CH713630B1 - Computer-implemented noise level simulation method and computer-readable medium and computer system therefor. - Google Patents

Abstract

The present invention relates to a method for calculating a noise level that is generated in an aircraft as a sound source (S), in particular an aircraft that moves along a defined trajectory (116) based on the at least one receiving point (R), for at least one Reception point (R), the method comprising the following steps: providing a sound propagation model for an air space which extends between the at least one reception point and at least one expected trajectory (115) along which the sound source is expected to move relative to the reception point; Calculating the noise level by entering the defined trajectory (116) into the sound propagation model and calculating a sound propagation between the sound source (S) and the at least one receiving point (R) for each of a number of different sound source locations (114) which are arranged along the defined trajectory , wherein the sound propagation is calculated based on at least one sound emission value of the sound source and the at least one sound emission value is calculated as a function of a sound radiation pattern which is assigned to a type of sound source for each of a number of different sound source locations which are arranged along the defined trajectory . The present invention also relates to a computer-readable medium containing computer-readable instructions to enable a computer to carry out a method according to the present invention. Finally, the present invention relates to a computer system, the means with instructions to carry out a method according to the present invention.

Description

The present invention relates to a computer-implemented method for calculating a noise level that is generated by a sound source, in particular an aircraft, which is moving with respect to the at least one receiving point along a defined trajectory, for at least one receiving point. The present invention also relates to a computer-readable medium that contains computer-readable instructions. Finally, the present invention relates to a computer system.

BACKGROUND OF THE INVENTION

In the current circumstances in densely populated areas such as Europe, traffic control and planning are subject to numerous factors influencing them. One of these influencing factors, which is of great importance and is being treated with increasing sensitivity, are the noise levels caused by traffic, as they are perceived by the public in residential and commercial areas. Particularly in the case of air traffic, noise levels generated by aircraft as a source of noise and perceived in areas near the airport determine the public acceptance of existing or planned take-off or landing routes, i. H. Aircraft trajectories during take-offs and landings.

[0003] The control and planning of such railways is moreover complicated, especially for traffic, in that the number of passengers is steadily increasing each year. In 2016, around 3.5 billion passengers worldwide used an aircraft. It is estimated that within 20 years, i.e. by 2036, the number of passengers in air traffic will increase to around 7 million passengers per year. This will most likely also increase the number of take-offs and landings, as the number of passengers increases more or less proportionally.

From a technical point of view, the trajectories of aircraft during takeoffs and landings are generally statically dependent on certain individual conditions of a particular airport and its surroundings, for example the orientation and length of runways, or geological conditions such as mountains, hills, etc. Furthermore, it results From a technical point of view, a dynamic dependence of the railways on certain temporal conditions such as weather conditions and traffic load. In addition to these technical conditions, the railways must comply with legal requirements that are specified by geographical parameters such as international borders and other restrictions, for example military areas, as well as time parameters such as night flight bans.

The latter in particular are mainly imposed by permissible noise levels in the vicinity of airports. The discussion about flight paths, especially take-off and landing routes, is already very emotional today between airport operators and airlines on the one hand and the public on the other, with politicians as a third party somewhere in between. Since the course of aircraft and thus their paths are more or less predetermined by predefined take-off and landing routes and routes cross borders between national and federal states, the respective discussions take place on an international level.

Because of the high level of emotionality, such discussions and thus also the planning and control of the railways are laborious and time-consuming. The parties involved in the discussion are often unable to find solutions or compromises because there is a lack of objective data on noise levels in affected residential and commercial areas. Instead, the discussions are based on mere assumptions and subjective interpretations of possible noise levels in relation to reception points.

An objectification of the discussions is desirable in order to streamline the decision-making process and make it more transparent, in order to be able to make fair decisions in which neither party feels ignored, and thus to increase planning security and to save costs and effort.

One way of providing a basis for objective decisions is the calculation of noise levels caused by aircraft at certain reception points. State-of-the-art aircraft noise calculations are already being used worldwide as essential tools for land use planning and control. They enable the assessment and optimization of noise abatement procedures for a quieter environment. Both topics represent important elements of the “Balanced Approach to Aircraft Noise” of the International Civil Aviation Organization ICAO. The first element is already well established through conventional immission models that are developed to calculate the noise level at the receiver . For this purpose, internationally harmonized procedures such as the publications ICAO Doc. 9911 or ECAC Doc. 29 is used, but national guidelines such as AZB in Germany and national programs such as FLULA2 in Switzerland also apply. With the latter element, more differentiated models, which describe the sound source and propagation separately, are necessary in order to precisely calculate individual flights. ANOPP, SIMUL and PANAM are currently available programs that meet this requirement. They all use similar semi-empirical emission models in their framework to describe the main sources of an aircraft.

[0009] Immission models for compliance with legal requirements and differentiated semi-empirical models for scientific purposes, however, have limitations in the evaluation of noise reduction methods. For example, the influence of the speed of the aircraft or its configuration on noise emissions is not taken into account in the noise forecast by means of emission models and is therefore not included in the calculation of the respective immissions, which is a considerable limitation. In addition, the acoustics description is simplified to generalized spectral classes (Doc. 9911 / Doc. 29) or attached to a standard atmosphere (FLULA2), which increases the model uncertainty in both cases. These inadequacies play no role in emission models that predict the spectra and directivity of each source in great detail. However, these require very detailed input of geometry data and physical flight parameters (e.g. primary jet speed or air flow mass) for accurate forecasts. Another disadvantage of noise calculation programs according to the prior art is the limited accessibility of these programs for other users and a very limited database.

SHORT REPRESENTATION

It is a concern of the present invention to remedy at least some of these disadvantages and shortcomings of the prior art noise calculation programs described above.

According to the present invention this object is achieved by providing a method according to main claim 1, a computer readable medium according to claim 6 and a computer system according to claim 7.

According to the present invention, the objectives listed above are achieved in that a method for calculating for at least one receiving point of a noise level that is generated by a sound source, in particular an aircraft, which is moving with respect to the at least one receiving point along a defined trajectory, consists of the following steps: Providing a sound propagation model for an air space which extends between the at least one receiving point and at least one expected trajectory along which the sound source is expected to move in relation to the receiving point; Calculating the noise level by entering the defined trajectory into the sound propagation model and calculating a sound propagation between the sound source and the at least one receiving point for each of a number of different sound source locations that are arranged along the defined trajectory, the sound propagation being calculated based on at least one sound emission value of the sound source and the at least one sound emission value is calculated as a function of a sound radiation pattern which is assigned to a type of sound source for each of a number of different sound source locations which are arranged along the defined trajectory.

In a computer-readable medium, the objects set forth above are achieved in that the computer-readable medium contains computer-readable instructions that enable a computer to carry out a method according to the present invention.

In a computer system, the objects set out above are achieved in that the computer system comprises means with instructions to carry out a method according to the present invention.

This solution offers the advantage over the prior art that in a first step general parameters of an immission model for the calculation of a noise level at the at least one reception point with the desired details of an emission model can be calculated for the expected trajectory and the spectra and directivity of the Sound source can be predicted at every point of the predefined trajectory with a high level of detail. Furthermore, a simulation of the sound propagation between the number of different sound source locations along the defined trajectory and the at least one receiving point can use a detailed acoustic description of the atmosphere and geological conditions with a level of detail that requires a lot of computing power that requires a spontaneous adjustment of model parameters based on relatively long calculation times of z. B. not possible for several hours or days.

If the sound propagation model is available for the air space extending between the defined trajectory and the at least one receiving point, an actual sound propagation between a defined trajectory and the at least one receiving point can be calculated based on said sound propagation model in a second step. The defined trajectory can deviate at least slightly from the expected trajectory. However, the sound propagation model for the airspace is applied to the defined trajectory, which can be easily changed and modified in order to simultaneously benefit from the desired depth of detail in the calculations of sound emissions and immissions. This enables a quick simulation of noise levels at at least one receiving point, which can be changed and recalculated almost immediately or at most within minutes and hours.

In other words, in a first step based on general parameters applicable to the expected trajectory, the framework for sound propagation is calculated in a sufficiently detailed manner. In a second step, additional model parameters such as fine adjustments to the trajectory of a vehicle and the vehicle type are defined, and the actual sound propagation between the vehicle running along the defined trajectory and the at least one receiving point is simulated within a very short time frame based on the defined framework. Various airspace scenarios can be calculated for expected orbits in range in preparation for simulating a noise level generated by different types of sound sources and expected orbits.

Thus, on the one hand, any planning and control devices can quickly simulate noise levels for certain reception points under different conditions, in particular using different defined paths, in order to map a route of a vehicle, e.g. B. an aircraft, to optimize in terms of noise pollution. On the other hand, respective simulations can be made available to the public quickly and changed according to public concerns about exposure to noise. In particular, this supports the objectification of discussions between operators, airlines, politicians and the public with regard to aircraft routes and thus facilitates the planning and control of air traffic.

The solution according to the invention can be combined and improved by the following further embodiments, each of which is advantageous in itself.

The sound propagation is calculated based on at least one sound emission value of the sound source. The sound emission values can include the total sound level, frequency spectrum and directivity characteristics of a sound source. Sound emission values of certain sound sources can be modeled based on backprobagized overflight measurements, which are separated into noise from the airframe and engines by means of a new separation method that is used during model development. Complex measurements with microphone arrangements are therefore not necessary, and the model can be applied to any type of aircraft powered by a turbofan.

The at least one sound emission value is calculated as a function of a sound radiation pattern which is assigned to a type of sound source for each of a number of different sound source locations which are arranged along the defined trajectory. If the sound source runs along the defined trajectory, its sound radiation pattern can change. Such changes can occur due to changes in operating modes and arrangements within the sound source as well as due to changes in an orientation and a location of the sound source within the airspace. With regard to aircraft, operating modes, parameters and arrangements can change along the defined trajectory, as well as the fact that throttle and flap settings as well as the status of the landing gear are changed, which changes the respective noise emissions. Furthermore, by different orientations and locations of the aircraft along the defined trajectory, the jet direction of the engines and the orientation of slats as sound sources are changed.

In some embodiments, the sound radiation pattern is based on a sound power level associated with the type of sound source. Different types of sound sources can be assigned to specific sound power levels. The sound power level can change along the defined trajectory.

In some embodiments, the sound power level is estimated based on course parameters of the sound source, which are derived from the defined trajectory. Such course parameters can be, for. B. climb angle, sink rate, top point of the landing approach, speeds and the like of an aircraft, but are not limited to these. These history parameters enable operating modes, parameters and arrangements such as weight, throttle, flap and landing gear settings to be estimated.

In some embodiments, the sound radiation pattern is based on a directional characteristic assigned to the type of sound source. Certain types of sound sources can have certain directional characteristics. For example, the nozzles or fans of an aircraft can direct the sound in a certain way. The different ways in which they direct sound and in which the direction of the sound has been changed according to operating and progression parameters can be taken into account when calculating the sound radiation pattern.

In preferred embodiments, the sound propagation model comprises a direct sound propagation scenario and a complex propagation scenario; wherein, in the direct sound propagation scenario, direct sound propagation between the sound source and the at least one receiving point is assumed; and wherein, in the complex propagation scenario, a complex sound propagation between the sound source and the at least one receiving point is assumed.

The direct sound propagation scenario can be used to calculate the direct sound propagation either as part of a complex sound propagation at a certain point in time and location of the sound source in relation to the receiving point or in addition to a complex sound propagation. With the complex sound propagation, all types of reflections and other effects on the sound passing through the air space can be taken into account. Thus, in terms of computing power, complex sound propagation is generally much more demanding than direct sound propagation. When deciding between scenarios, if complex sound propagation can be neglected due to the fact that direct sound propagation provides sufficient accuracy, omitting the calculation of complex sound propagation can assist in minimizing the computational effort and thus in simulating noise levels in accordance with a method according to the present invention Save time and energy.

In some embodiments, for each of the number of different sound source locations, a line-of-sight angle between the sound source and a horizon is determined; wherein the direct sound propagation model is applied to line-of-sight angles exceeding an insignificance threshold; and where it is assumed that at line-of-sight angles exceeding the insignificance threshold, complex sound propagation becomes negligible.

The use of the line of sight angle to differentiate the scenarios for the application of the Direktschallausbreltungsmodell and the complex sound propagation model represents a simple condition for switching between the models according to the different sound source locations along the defined trajectory.

In some embodiments, a homogeneous atmosphere within the airspace is assumed for the direct sound propagation scenario. Using a homogeneous atmosphere for direct sound propagation simulation helps minimize the complexity of the calculations involved. The complex sound propagation simulation is in turn preferably based on the assumption of an inhomogeneous, i.e. H. heterogeneous atmosphere within the airspace. Alternatively or additionally, if desired, the direct sound propagation scenario can also use calculations based on the assumption of a heterogeneous atmosphere within the airspace.

In some embodiments, the direct sound propagation scenario includes at least one of a geometric deviation model, a dissipation model, a barrier effect model, a fouling attenuation model, and a ground effect model for the airspace. Thus, the effects of geometric deviations and a dissipation of sound within the airspace can be observed. Barrier effects, d. H. Effects of any type of obstacle such as buildings, geological formations or the like that border on the airspace can be observed. Vegetation loss modeling is particularly helpful when providing calculations for various primary surfaces and / or seasons where it is assumed that the surrounding vegetation is or is not leafy. The modeling of ground effects supports the simulation of attenuation and reflection of sound according to individual characteristics of the ground adjacent to the airspace.

In some embodiments, the ground effect model includes at least one of spherical wave propagation, a rough terrain model, surface texture variation, and a loss of coherence model for modeling loss of coherence of various sound paths between the sound source and the receiving point. The determination of spherical wave propagation includes the assumption that sound is emitted by the sound source and, based on the sound source, passes through the air space as spherical wave h. Modeling an uneven terrain adjacent to the airspace enables the reflection and attenuation of sound on the ground to be simulated. This can include surface texture variations, as e.g. B. The roughness and texture of the ground can change along the airspace and over time. When running between the sound source and the at least one receiving point, the sound propagation can be divided into different paths by certain obstacles, which means that the sound waves running along these paths are separated so that they lose their cohesion,. Then the sound propagation must be calculated separately for each of the paths.

In some embodiments, the complex propagation scenario includes at least one of a weather correction model, an obstacle reflection model, and a forest diffusion model for the airspace. In the case of the correction model for the effects of weather, in particular changes over time in the atmosphere of the airspace are considered. In the obstacle reflection model, all reflections and obstacles between the sound source and the at least one receiving point are considered. In the forest diffusion model, diffusion effects are simulated in connection with sound waves hitting forest structures.

In some embodiments, the weather effects correction model adapts a calculation of sound dissipation within the air space to local temperature and humidity values of the air space. The propagation of sound within the air space depends on the temperature and humidity of the air. Thus, by looking at local temperatures and humidity values, the propagation of sound within the airspace can be precisely simulated for different weather conditions such as wind, precipitation, fog as well as different times of the day and seasons in which the propagation of sound is changed according to the respective weather conditions.

In some embodiments, the weather effects correction model adapts a calculation of barrier effects within the airspace to vertical gradients of at least one of wind speed values and temperature values of the airspace. Wind and temperature within the air space determine the reflection and diffraction of sound at obstacles adjacent to the air space. Taking temperature and / or wind speed values into account when calculating barrier effects supports the precise simulation of noise levels at the at least one receiving point under different weather conditions.

In some embodiments, the barrier effects include an acoustic shadow zone effect for at least one acoustic shadow area within the airspace in which the sound beams emitted directly and / or indirectly by the sound source reach the at least one receiving point based on updraft conditions that have been identified based on the vertical gradient of the at least one wind speed value , do not reach directly, a. In other words, under certain weather conditions, the at least one receiving point can be shielded from the sound waves by an obstacle in such a way that the sound waves pass above and / or next to the at least one receiving point located in the acoustic shadow area of the obstacle.

In some embodiments, a residual sound exposure of the at least one receiving point located within the at least one acoustic shadow area is based on at least one of a diffraction effect model and a scattering effect model applied to a sound beam passing along the at least one acoustic shadow area, calculated. Thus, even if the at least one reception point located in a sound shadow area is not directly influenced by a sound passing an obstacle, fractional and scattering effects from the obstacle on the sound can result in residual sound waves reaching the at least one reception point and at least partially closing them contribute to the noise level simulated for the at least one receiving point.

In a preferred embodiment, a method according to the present invention further comprises the following steps Dividing the airspace into adjacent subspaces; Calculating a subspace model for each of the subspaces to determine a finite sound propagation within each of the subspaces; Assembling the sound propagation model from the sub-room models, at least one boundary area between adjacent sub-rooms reproducing at least one of a virtual sound source and a virtual receiving point for a virtual sound transmission value that reproduces a virtual sound power level transmitted between at least two adjacent sub-rooms.

Thus, arithmetic operations for simulating sound propagation within the airspace can be distributed to the subspaces in such a way that the total sound propagation is calculated iteratively. The border areas between the sub-spaces become virtual sound sources and virtual reception points via which sound waves are transmitted from one sub-space to the other. The border areas can, for. B. be corners of the subspaces that share the same or adjacent coordinates as neighboring subspaces, whereby the border areas of the subspaces form a three-dimensional grid. The sub-spaces can be viewed as cells, within which attenuation factors for sound propagation are calculated according to the properties of each cell. The airspace is thus represented as a cell model.

Attenuations can be calculated between each edge, vertex or corner point of the three-dimensional grid of the cell model and each reception point. Based on these iteratively calculated attenuations, an overall attenuation of the sound between the sound source and the receiving point can be calculated. The overall attenuation can then be used to calculate a noise level for the at least one receiving point that is generated by a sound source that moves along a defined trajectory relative to the at least one receiving point. This calculation based on the overall attenuation can be carried out immediately, i. H. Iteration calculations between corner points of the sub-areas are not necessary if the overall attenuation is present.

During the calculation of immissions, d. H. Noise levels at the at least one receiving point, which are caused by an aircraft as a function of the position of the aircraft at a certain point along a trajectory, can be attenuations related to the at least one receiving point or a plurality of receiving points from the attenuations of the eight surrounding corner points of the three-dimensional Grid can be interpolated. To calculate immissions, reception points can also be arranged along a grid of reception points arranged on the floor. A sound source grid and a grid of reception points can be independent of one another.

In some embodiments, a transmission of virtual sound power values is calculated for a number of possible combinations of virtual sound sources and virtual reception points. In particular after sound deflection, diffraction, scattering or the like, the sound waves can pass through the air space along different paths. These different paths lead to a number of possible combinations of sound propagation between different rooms. Taking into account all possible combinations, the propagation of sound within the airspace can be precisely simulated.

In some embodiments, a virtual attenuation is calculated for each of the virtual sound sources and / or virtual reception points, the virtual attenuation reproducing an attenuation of the virtual sound power value during transmission between the virtual sound sources and / or virtual reception points. In other words, within each sub-room and / or for each transmission of sound between sub-rooms, a virtual attenuation of the sound is calculated according to properties within the sub-room or between sub-rooms. Such properties can in turn include any type of weather or other values discussed herein for simulating the propagation of sound in air, taking into account boundaries and obstacles adjacent to the airspace.

In some embodiments, virtual attenuations are stored in a subspace database. By storing the virtual attenuations within a subspace database, the virtual attenuations can be called up immediately and quickly from the subspace database. This supports the rapid calculation of sound propagation within the airspace by using the virtual attenuations stored in the subspace database.

As already stated above, virtual attenuations can be used for calculating an overall attenuation. Reception points can be grouped into sub-territories into which a landscape bordering an air space around the railways can be divided. The sub-territories can form a tile pattern on the floor. For each sub-territory, total attenuations derived from the virtual attenuations of the subspaces and / or the overall attenuations can be stored in the subspace database and / or a supplementary database, which enables efficient parallelization of computation steps in a method according to the present invention. For reception points within a sub-territory, attenuations related to all corner points of the three-dimensional virtual sound source grid can be stored within the same database.

In some embodiments, the virtual attenuations are read from the subspace database while the noise level is being calculated. As a result, the noise level simulation can be carried out directly, i. H. more or less in real time based on the virtual attenuations stored in the subspace database. The subspace database can be updated in accordance with a change in the environmental conditions influencing the airspace.

In some embodiments, the propagation model includes an intermediate attenuation interpolated between the defined trajectory and the at least one boundary region. It can be assumed that the actually defined trajectory crosses the sub-spaces. The defined trajectory therefore does not coincide with the boundary areas for most of the locations of a vehicle. For the remaining sections of the sound paths between the actual location of the vehicle along a defined trajectory and the respective boundary areas, an intermediate attenuation is calculated in order to provide a precise simulation of the sound propagation.

In some embodiments, the air space is divided along a homogeneous horizontal grid with fixed spacings between the subspaces in a longitudinal direction and a transverse direction of the air space. The division of the air space along a homogeneous horizontal grid facilitates the definition of the subspaces, which are thus evenly distributed and can have the same sizes along the horizontal grid.

In some embodiments, the airspace is divided along a heterogeneous vertical grid with increasing heights of the subspaces along a height direction of the airspace. In other words, the height of the subspaces increases with an increasing distance of the subspaces from the ground. This helps minimize the total number of subspaces that must be created to represent the airspace. The computational effort is reduced by minimizing the total number of subspaces. Such a minimization of the number of sub-spaces takes into account that with increasing height above the ground, the air space and the atmosphere can be viewed as increasingly homogeneous. Furthermore, with increasing height above the ground, fewer obstacles have to be taken into account when modeling the airspace. Thus, with increasing distance from the ground, the sub-spaces can become larger and still provide sufficient increments for the simulation of sound propagation within the air space.

In some embodiments, the subspaces have a cuboid shape. The cuboid shape results from the fact that the horizontal grid and the vertical grid consist of straight lines that are arranged at right angles with respect to one another. In other words, the horizontal and vertical grids are arranged to form a Cartesian coordinate system, which makes calculations easier.

In some embodiments, the border areas are formed by eight corners of each of the cuboid-like subspaces. Thus, the boundary areas can be easily defined. A sound propagation within the airspace divided into partial spaces is simulated in that the corners of the partial spaces are viewed as virtual sound sources and virtual sound reception points. Virtual attenuations can be calculated between the virtual sound sources and the virtual sound receiving points. As already stated above, a grid of virtual sound sources can, however, differ from a grid of reception points. In addition, front points of surrounding buildings and / or other obstacles or any points located in a desired manner, e.g. B. measuring or recording points can be used.

In some embodiments, the sound propagation between the sound source and the at least one receiving point is calculated both in the time domain and in the frequency domain. Thus, time domain effects and frequency domain effects of the sound propagation, in particular a sound attenuation in the course to the sub-spaces, can be simulated.

In some embodiments, a frequency spectrum of the sound propagation in the frequency domain is divided into frequency bands, and the sound propagation is calculated for each of the frequency bands. By dividing the frequency spectrum to be considered for sound propagation in the frequency range into different frequency bands, different sound wave characteristics can be considered as a function of the wavelength. Correspondingly, certain parameters and / or constants of the propagation model can change as a function of the wavelength and thus the frequency band. Furthermore, the computation operations necessary for simulating the propagation of sound can be made easier and minimized to the effect that, for B. different algorithms are used for lower frequency bands than for medium and higher frequency bands. For example, the frequency bands can be divided into a spectrum of one-third intervals. This enables a third of an octave analysis and calculation in the frequency domain.

In some embodiments, the sound propagation is calculated with unweighted sound pressure levels and the noise level is displayed with A-weighted sound pressure levels. The calculation of sound propagation with unevaluated sound pressure levels enables a simulation of sound propagation on the basis of purely physical properties, whereas the display of sound pressure levels and / or noise levels with A-weighted sound pressure levels supports the visualization of the perception of the respective noise level by people. In this way, the actual perception of the noise level by people at a specific reception point can be precisely simulated.

In preferred embodiments, the noise level is weighted with a population value which represents an estimated population of the at least one reception point with a number of people at a predefined point in time at which the sound source moves along the defined trajectory. If the at least one reception point is located in a residential area, it can be assumed that the residential area has a certain population value corresponding to the number of people present at the at least one reception point at a certain time of the day and / or year. For example, residential areas are densely populated; H. the actual population figures correspond more or less to a theoretical number of people who are officially registered in the residential area, especially at night and / or on weekends, whereas residential areas tend to be less during working or business hours, for example from 9 a.m. to 5 p.m. are populated. In contrast to this, commercial areas are usually densely populated during working or business hours, so that it can be assumed that commercial areas are generally only sparsely populated at night and / or on weekends. Assessing the noise level with a population value that represents the estimated population of the at least one reception point at a predefined time thus helps to simulate the actual effect of the noise level on the population.

In preferred embodiments, a calculation of the sound propagation is distributed between a client device, which provides a number of client FLOPs per cycle, and a network of computers, which provides a number of cluster FLOPs per cycle, the number of clients -FLOPs per cycle is less than the number of cluster FLOPs per cycle. The client device can be a single-user computer in an office or at a service provider who, for example, provides engineering services including the simulation of noise levels discussed herein. The computer network can be any network of interconnected computers or data centers that provides computing power measured in FLOPS that exceeds the computing power of the client device.

It is particularly advantageous if the calculation of the sound propagation model for the air space extending between the at least one receiving point and the at least one estimated trajectory along which the sound source is likely to move based on the receiving point is carried out by the computer network is sufficient Provides computing power while being less accessible and flexible than the client device in terms of operation and intervention in the computations. The client device is then used to calculate the noise level by entering the defined trajectory and the sound propagation model provided by the computer network. The computing power of the client device should be used to calculate the cumulative propagation between the sound source and the at least one receiving point for each of a number of different sound source locations, which are changed along the defined trajectory, according to the respective requirements as desired for simulating different sound sources and defined trajectories can, are arranged, be sufficient.

In some embodiments, at least one total sound attenuation data set, which reproduces an attenuation of sound along at least one sound path that was created between the sound source and the receiver, is made up of a number of partial attenuation value data sets that were generated by the computer network and various Reproduce sound attenuation characteristics of the airspace for a respective scenario, calculated. The recipient can be the at least one receiving point. The partial attenuation value data sets can include the partial space data sets which provide virtual attenuations for the individual partial spaces of the airspace. In addition, the partial attenuation value data sets can reproduce different sound attenuation characteristics for respective scenarios and the fact that different weather and / or population scenarios are being considered. By preprocessing the scenarios in the computer network, they can be made easily accessible to the client device in order to quickly simulate the noise level at the at least one receiving point by using the at least one overall sound attenuation data set.

In some embodiments, the overall attenuation value data set is calculated by the computer network and transmitted to the client device or calculated by the client device. In other words, the total attenuation value data record can be calculated completely by the computer network, in order then to be transmitted to the client device, i. H. by downloading the total attenuation value data set from the computer network by the client device or uploading the total attenuation value data set to the client device by the computer network. Furthermore, each overall attenuation value record stored in a client device can be updated in a similar manner. On the other hand, at least some steps in calculating and updating the overall attenuation value data set can be taken on the client device. If, for example, a transmission band between the client device and the computer network has certain restrictions that do not allow the overall attenuation value data record to be regularly transmitted from the computer network to the client device, certain calculation and update processes can instead be carried out locally on the client device be e.g. By merely differential recalculation or updating of the total attenuation value data set.

In some embodiments, a sound profile is provided that includes at least one mean noise level at the at least one receiving point; and wherein the at least one mean noise level is associated with one of a bundle of sound source trajectories. The sound profile can be calculated with a plurality of estimated and / or defined paths. For each estimated and / or defined trajectory, an average noise level can be calculated for the at least one receiving point. Thus, when performing the actual simulation of the noise level, the sound profile can be used to quickly calculate the noise level based on the at least one mean noise level associated with at least one of the plurality of estimated and / or defined trajectories. This supports the minimization of computational effort and simplifies and thus accelerates simulation operations according to the present invention.

In other words, a calculation of sound propagation and a calculation of sound profiles, the average immissions generated by an aircraft of a certain type, which is traveling along a certain trajectory, ie. H. Route, running, caused, reproduced, based on attenuations calculated in advance, advantageously carried out on a computer network or central computer. Simulation of single flights and calculations of sound profile overlays based on traffic volume for the calculation of a traffic scenario can be carried out on a client device for easy accessibility and changeability.

In some embodiments, the sound profile is calculated by the computer network and transmitted to the client device. Thus, the computer network has two sound profiles available for quick changes, and a subsequent calculation of different scenarios of the sound propagation leads to respective noise levels at the at least one receiving point on the basis of the respective profile. For this purpose, the sound profile can be transmitted from the computer network to the client device as described above with reference to the overall attenuation value data record.

In some embodiments, an overlay of at least two sound profiles is computed by the client device. In other words, the client device can superimpose several sound profiles in order to interpolate simulations or quickly provide different but superimposed simulations of the sound propagation and thus the noise level at the at least one receiving point. This supports the rapid provision and rapid adaptation of noise level simulations according to the present invention.

In preferred embodiments, a method according to the present invention further comprises a data processing step in which a number of source points corresponding to the various sound source locations along the defined trajectory is defined. By defining the number of source points, sound propagation simulations can be calculated for each of the number of source points. In view of the high computational effort involved, such calculations can, for. B. be made by the computer network in order to be made easily accessible to the client device.

In some embodiments, the sound propagation model is calculated based on at least the number of source points, a number of reception points and a geodata set that represents a geological environment of the airspace. As a result, the sound propagation model initially considers the source points, the reception points and the geospatial data in a pre-calculated manner. Based on such calculations carried out in advance, all further data such as sound source parameters, paths and / or weather data can be entered into the sound propagation model in order to refine the model. Since the initial consideration of source points, reception points and geodata can require considerably more computing power than refining the model using the sound source parameters, paths and / or weather data, such a calculation carried out in advance helps to minimize computational effort and time when refining the model according to the desired source parameters, tracks and / or weather data.

The at least one sound source is an aircraft. Such an aircraft can be an aircraft, for example a jet aircraft for civil or military use, a helicopter or the like. Various different types of such aircraft can be modeled in accordance with the present invention. Acoustic overflight measurements can be performed for sound source modeling in order to measure sound emissions based on recordings of real air traffic in the vicinity of airports or the like.

BRIEF DESCRIPTION OF THE FIGURES

In order to describe the manner in which advantages and functions of the invention can be obtained, a more detailed description of the principles briefly described above is provided below with reference to the embodiments thereof which are illustrated in the accompanying drawings. These drawings merely depict exemplary embodiments of the present disclosure and are therefore not to be regarded as limiting the scope thereof.

The following applies to the figures: FIG. 1 shows a schematic representation of a beam tracking algorithm in order to derive the sound path from the source S to the receiver R in a method according to an embodiment of the present invention; 2 shows a schematic representation of a measuring arrangement in accordance with a method according to an embodiment of the present invention; 3 shows a schematic representation of a flight path axis system with polar angles θ and φ in the longitudinal and transverse directions; 4 shows a schematic representation of an angle coverage by any measurement setup for a short take-off (left) and two flights with different take-off points (right), for two microphones (| I: φ ≈ 40 °; II: φ ≈ 0 °); 5 shows schematic diagrams which represent the directional uncertainty in the case of radar data at a distance of 1 km (left) and 5 km (right) before touchdown; 6 shows a schematic representation of a setup for further distance measurements at an airport in accordance with a method according to an embodiment of the present invention; 7 is a flow diagram illustrating steps in data processing from measurements to an emission record at the source in accordance with an embodiment of the present invention; 8 is graphs showing the results of engine test runs on exemplary A330-300 aircraft; FIG. 9 shows two diagrams showing an exemplary influence of N1 on foreign language; FIG. 10 shows two diagrams which illustrate an exemplary influence of Ma on the sound emission level of the aircraft type A320; 11 shows two diagrams which represent an exemplary distribution of measured flap lever positions of the aircraft type A320 as a function of the Ma number; 12 shows two diagrams which illustrate an example of the influence of the landing gear on the noise emission of the aircraft type A320 during the approach for landing while idling; Fig. 13 is a diagram showing steps of a process for model development and data separation according to an embodiment of the present invention; 14 is three diagrams showing an example of data separation for the A320 at 100 Hz; 15 is a diagram illustrating an exemplary mean energy versus frequency correction factor for the A320 aircraft type; 16 shows two diagrams illustrating an exemplary determination coefficient versus frequency; 17 shows two diagrams which illustrate exemplary spectral directional characteristics of the A320 aircraft type during take-off with a high-performance setting; 18 shows two diagrams showing exemplary spectra for final approach and take-off with high-performance setting of the aircraft type A320; 19 shows two diagrams showing exemplary spectra for final approach and take-off with high-performance setting of the aircraft type E170; 20 shows two diagrams showing exemplary spectra of a landing approach of an aircraft of the type A320 with completely opened flaps and extended landing gear at θ = 130 °; 21 shows two diagrams illustrating exemplary spectral directivity characteristics for a departure with a low power setting; 22 shows two diagrams which illustrate an exemplary directivity in the longitudinal direction of the overall Lw for three different take-off power settings of the A320 aircraft type; 23 shows two diagrams illustrating an exemplary transverse directional effect of the overall aircraft for three different take-off power settings of the A320 aircraft type; 24 shows six diagrams which exemplify two landing approaches in the idle setting and with extended landing gear and high-lift elements on a receiver about 15 km before the landing threshold; 25 shows two diagrams which exemplify exemplary spectra of a landing approach of an aircraft of the type A320 idling about 15 km before the landing threshold; 26 shows two diagrams which show exemplary spectra of a landing approach of an aircraft of the type A330 idling about 15 km before the landing threshold; 27 shows two diagrams showing an exemplary radiation balance for an exemplary day in combination with the wind speed and weather categories representing different conditions during the day; 28 shows three diagrams showing exemplary temperature, wind and humidity profiles for an exemplary day at 10:00 a.m.; 29 shows three diagrams illustrating exemplary temperature, wind, and humidity profiles for an exemplary day at 9:00 a.m.; 30 shows two diagrams showing exemplary temperature, wind and humidity profiles for an exemplary day between 9:00 a.m. and 12:00 noon; 31 shows a schematic illustration of a calculation scenario based on a flight path; 32 is a diagram showing an exemplary variation in air absorption of a homogeneous atmosphere shifted from current conditions to a uniform atmosphere with averaged conditions; 33 is a diagram showing an exemplary variation in air absorption of a homogeneous atmosphere shifted from current conditions to a uniform atmosphere with averaged conditions; FIG. 34 shows two diagrams showing, by way of example, differences in air absorption between a homogeneous atmosphere under current conditions and COSMO-2 profiles; FIG. 35 shows two diagrams which illustrate, by way of example, differences in air absorption between idealized profiles and COSMO-2 profiles; 36 shows a diagram illustrating an exemplary influence of the dissipation variance on the A-weighted spectrum at a receiver; 37 shows a schematic perspective view of a partial space used in a model for calculating attenuations in accordance with a method according to the present invention; Figure 38 is a schematic perspective view of an air space divided into subspaces in accordance with a method of the present invention; and FIG. 39 is a schematic diagram illustrating a system for performing a method in accordance with an embodiment of the present invention.

DETAILED DESCRIPTION OF THE FIGURES

One embodiment of the present invention is a method and a corresponding computer-implemented time step program for calculating aircraft noise, in which the sound source and the propagation calculation are strictly separated from one another. In contrast to the simulation model FLULA2, which was developed at the Eidgenössische Materialprüfungs- und Forschungsanstalt (EMPA) for the acoustic investigation of complex scenarios such as annual air traffic at an earlier point in time, the program, according to one embodiment of the present invention, focuses on individual flight events in order to use from either generic data, e.g. B. from a flight simulator (FFS), or cockpit data of real flights to carry out investigations and to optimize noise reduction procedures.

The aircraft as a sound source is described by physical laws and scale empirical data relating to flight parameters such as power setting or speed and aircraft configuration (slats, flaps, landing gear). In order to obtain a sufficient data basis for various types of aircraft and engines, extensive measurements of real air traffic were carried out at Zurich Airport.

A highly developed sound propagation model is adapted to the special conditions of the aircraft noise calculation. The sound emission and propagation models are combined in a geographic information system. This interface enables you to prepare projects, perform calculation tasks and obtain useful tools for the analysis and presentation of results. In addition to individual flight events, the algorithms and the program structure also enable the calculation of complex scenarios such as annual air traffic. Although the emissions model causes more effort in the preparation of the input data, this does not have any relevant effects on the calculation time. In contrast, the efficiency of the dispersion model plays a crucial role.

Acoustic emission model

A semi-empirical acoustic emission model according to an embodiment of the present invention is based on a combination of data measured in real air traffic with generalized physical laws that establish the relationship between flight configuration and acoustic emission, including information on frequency spectrum and directivity. Thus results and experiences from the research program MODAL of the German Aerospace Center are considered. At greater distances from the airport, it is not possible to cover a wide range of polar angles. Sound directional effects may not be able to be created. However, measurements are still of interest in order to determine the sound emission for different flight conditions. For this purpose, mobile measuring stations are used at numerous different locations at distances of up to 25 km from the airport. If the directivity cannot be determined at distant locations, spectral sound level differences are used as a substitute solution in order to take changes in the flight configuration into account.

Cockpit data is used to determine the flight configuration of the measured aircraft events. Such cockpit data are available for example for the Swiss air fleet, namely the aircraft types of the Airbus A320 series, the A330-300 type, the A340-300 type and the Avro RJ100 type. The cockpit data cover all the necessary information in high time resolution, for example flight path, spatial orientation, real air speed, rotational speeds of the engines and the position of slats, flaps and landing gear. In the case of aircraft from other airlines for which no cockpit data is available, the ground speed and the inclination can be derived from the flight path. In addition, the speed of the low-pressure turbine (N1) is estimated by spectral analysis of the acoustic data as an indicator of the power setting. This procedure works exactly in the immediate vicinity of the airport. At a greater distance with strongly weakened signals, the evaluation of N1 presents a greater challenge.

Sound propagation model

The calculation of the sound propagation in a method and computer-implemented program according to an embodiment of the present invention is based on a model called sonX, which has a propagation core that is optimized from an acoustic point of view and with regard to performance.

The sonX propagation model relates to point sources. Direct sound is calculated on the basis of vertical sections of the terrain from the source to the receiver including buildings and other barriers. The calculation is carried out in two steps. In a first step, a calculation is made assuming a homogeneous atmosphere. Geometric deviation, dissipation according to ISO 9613-1, barrier and soil effects as well as vegetation damping according to ISO 9613-2 are taken into account. For example, barrier effects can be calculated as implemented in ISO 9613-2. For the calculation of ground effects, an analytical solution for spherical waves is used, which has been extended for uneven terrain and varying surface properties. In a second step, the weather influences on sound propagation, in particular the influence of local temperature and humidity on dissipation and the consequences of vertical gradients of wind and temperature on the effects of the screen, are determined. The latter is done using a ray tracking algorithm that derives the sound path from the source to the receiver along possible barrier edges from the effective sound velocity for any profile.

Figure 1 shows a ray tracking algorithm to derive the sound path from source S to receiver R. The emergence of acoustic shadow zones can be derived as an additional effect. Reflections on buildings and walls are taken into account. The model differentiates between coherent and mirrored reflections and scattering. Diffuse reflections from forest edges and cliffs are represented by two separate models. The sound propagation modeling is done in 1/3 octave bands. A frequency range of 20 Hz to 5 kHz is used for aircraft noise.

Simulation tool

According to the time step concept, a single flight is reproduced by source positions which follow the flight path in predetermined time steps, usually one second. The angle-dependent sound emission is calculated at each position with the current power setting, aircraft configuration and orientation. At any recipient location, the amounts from each source position are summarized in chronological order. Acoustic values such as equivalent continuous sound pressure level or maximum sound pressure level can be derived from the resulting level-time protocols.

The resulting simulation tool is used for the detailed analysis of individual flights and the calculation of complex scenarios with processing of several tens of thousands of flights in order to create noise maps of large areas with several tens of square meters. In the latter case, the computation time is a critical point that can be reduced by two measures.

First, a differentiation of the propagation situation is initiated. If sound propagates in the vicinity of the ground, a direct model as described above with reference to FIG. 1 is used. At higher altitudes, on the other hand, the situation is much simpler, since there are no shield effects and the time-consuming processing of terrain profiles can be omitted. Furthermore, the influence of temperature and humidity on the dissipation is the only relevant weather influence. In addition, direct sound dominates and reflections can be neglected. A simplified approach is used for this, which only takes into account geometric deviation, dissipation and a standardized ground effect. The line of sight angle with respect to the horizon line is used as a criterion for distinguishing complex and simplified situations.

Second, the detailed or complex sound exhaust calculations are made prior to the calculations of the actual aircraft noise, and the resulting attenuations are stored in a demand database. For this purpose, an air space is divided into basic rectangular shapes, each of which forms a cell-like sub-space. A sound propagation calculation is carried out for each of the eight corners of those cells through which the aircraft actually flies. During the simulation of individual flights, the relevant attenuations are derived by interpolating at the actual source position from the values of the cell corners according to the database.

In order to avoid inconsistencies in the transition from the simple to the differentiated or complex model, dissipations per meter with sonX are calculated in advance for given weather conditions and stored in a look-up table for different layers of the atmosphere. During the single flight simulation, attenuations can either be accessed from the database or, in the simple case, they can be calculated directly using the latter dissipation values.

Measurement concept

Measurements are carried out at a public airport, for example Zurich Airport, with real air traffic, which enables the most important commercial aircraft types that operate at comparable airports to be recorded. Due to different lengths of runways, operating concepts and destinations at the airport, a wide range of flight parameters can be collected for the development of the semi-empirical sound source model. Independent measurements are taken both at short and long distances from the airport.

FIG. 2 shows a measuring arrangement on the runways 16, 28 and 34 of the airport. In the immediate vicinity of the airport, data for 3D directivity patterns are collected by placing six to eight microphones next to the runways. The angles covered depend on the microphone position and the take-off point, which can differ by up to 500 m for the same aircraft type and by up to 1500 m for different aircraft types, mainly due to different take-off weights and thrust performance. The microphones are not only located at the end of the runway, but also along the runway to detect premature take-off. During the day, most of the data can be recorded on runway 28 (departure to the west) and an additional runway (approach from the north, not shown).

Due to German air traffic restrictions in the early hours of the morning (6:00 a.m. to 7:00 a.m.), some types of aircraft only arrive from the south (runway 34). Heavy aircraft types such as the A330 or A380 also normally take off in a southbound direction from runway 16. Since these types are acoustically relevant, measurements are also carried out on runways 16/34. The locations of the microphones can be optimized to cover a wide range of polar angles and thus to accumulate sufficient data for the 3D directional characteristics.

For greater distances, approximately twelve microphones are installed at locations up to 25 km from the airport, which cover various flight configurations. Departures are measured along the regular flight route west of runway 28 and east of runway 16. Landing approaches are only measured on runway 34, since sufficient data is available there for all types of aircraft.

Flight paths are determined in the immediate vicinity of the airport (time and position) by an optical tracking system. At greater distances from the airport, the requirements for the accuracy of flight path location are lower and radar data should provide sufficient accuracy. These systems are particularly necessary for the aircraft of airlines where cockpit data is not available. A mobile multilateration system is used to validate tracking data derived from different sources.

Prediction of the variety of angles for 3D directional effects

In order to be able to create reliable 3D directional characteristics, a wide range of polar angles must be covered. The microphone positions were optimized accordingly using a Matlab program that predicts the coverage of the polar angles. Since the departure positions can differ greatly, flight paths which are derived from radar data from various aircraft types and are provided with an early and a late departure point for each type are used as input data for the forecasting tool. For each discrete source position (s) of the selected flight path, a relative vector rg, mic for a microphone was determined, as described in airborne normal world coordinates (index g) as defined in DIN9300-1 according to equation (1) below. Since all sample data are given in Swiss national coordinates CH1903, the xg-axis is aligned in the direction of true north.

The above equation (2) is used to transform the vector rg, mic into the Cartesian flight path axis system (index k, see FIG. 3) in order to determine the polar angle from each source position to the microphone. Since information on the angle of inclination is usually not available, the influence on the aircraft orientation is neglected, but is the subject of future studies. The transformation matrix is converted into the flight path.

3 shows a schematic representation of a flight path axis system with polar angles θ and φ in the longitudinal and transverse directions. The polar angles θ and φ for each discrete point of the flight path are shown by Lambert's cosine law with the vector rk, mic (Eq. 2) or vector rk, mic, yz and the unit vectors x and zk as shown in Fig. 3 and according to the equations below ( 3) and (4).

The calculations consider the portion of the overflight from shortly after take-off to an altitude of 1,500 feet above the runway. This flight segment reproduces an almost stationary flight with constant speed and constant power settings. The aerodynamic noise of the landing gear, which is retracted in this phase, can be neglected due to the dominance of the engines. A height of 30 feet or a slope angle of 10 ° are suggested as criteria for starting the evaluation, while the throttling at 1,500 feet is the criterion for ending the measurement.

4 shows, by way of example, results of the forecasting tool, in particular the resulting polar angles for any flight paths, which are graphically represented as flight path axes with normalized vectors on one side of the hemisphere that represents the underside of the aircraft. (Only one side is graphed as the sound directivity is modeled symmetrically along the longitudinal axis, xk). In the case of a flight path with a short take-off, almost all microphones show good coverage of θ due to their location in front of the take-off point. The data is limited to θ angles between 15 ° and 170 °. However, these angles are negligible in relation to their sound contribution. In addition, even if the angular coverage of a single overflight is insufficient, the variation in flight paths of a statistically sufficient number of events will fill this gap, as shown in FIG. 4 (right). Good data coverage can be predicted within the marked surfaces of two exemplary microphone positions.

For the optimization of the microphone positions shown in FIG. 2, all eight microphones of any arrangement are calculated. The positions are optimized for different flight paths of different types of aircraft on each of the runways covered by the measurements. In addition, an example of a final arrangement as shown in FIG. 2 is also dependent on various other conditions such as security requirements at the airport, inaccessibility of some areas and acoustically disadvantageous locations with reflections or strong background noise.

Directional uncertainty

Reliable 3D directional characteristics require precise determination of the aircraft position, which affects the accuracy of the polar angles. The standard inaccuracies uθ and uφ (confidence interval 68%) can be estimated by the error propagation law based on the uncertainties of the horizontal and vertical aircraft position by applying equations (5) and (6) below to equations (3) and (4) above. where xk is the longitudinal axis, yk is the transverse axis and zk is the vertical axis of the flight path axis system (Figure 3). The quantities of uθ and uφ are quantified for radar data with a lateral tolerance of 230 m and a vertical tolerance of 46 m. This estimate gives an upper limit for uθ and uφ since other positioning systems such as cockpit data are more accurate. Assuming a rectangular distribution within the tolerance limits, the conversion of tolerances t into standard uncertainties can be calculated according to equation (7) below as follows

5 shows schematic diagrams which represent the directional uncertainty for radar data at a distance of 1 km (left) and 5 km (right) before touchdown. The microphone position is shifted sideways from 0 to 100, 200 and 300 m. Above: glide path; Middle: uncertainty uθ; Below: Uncertainty uφ. (Note: different scales of the standard uncertainties for 1 km.)

Interim conclusion on data collection

The above-described model for a method and a computer-implemented program according to an embodiment of the present invention takes into account flight configuration parameters and thus meets the requirements for acoustic optimization of flight methods. The sound emission model as a function of the flight configuration and the underlying sound source database are decisive for the development of such a model. Comprehensive measurements have to be carried out in the immediate vicinity as well as at a great distance from the respective airport, e.g. B. Zurich, can be made. A special focus is on the development of reliable 3D sound directional characteristics. The latter require optimized microphone positioning for broad coverage of polar angles. The prognosis of the polar angle coverage of a certain measurement arrangement is of great help in optimizing the measurement setup. Furthermore, the determination of the polar angle itself must be reliable. Their uncertainties are therefore estimated as an example for a typical landing approach. The results show that radar data for measurement positions at great distances from the airport (uncertainties of 0.6 °) are sufficiently accurate, while other systems are required in the immediate vicinity of the airport.

Using a hybrid case-dependent propagation model in combination with an attenuation database, it is possible to provide a simulation tool with high flexibility and accuracy without increasing the computational effort. A significant advantage of the attenuation database is that it can be reused for numerous calculations once it has been created for a particular airport. Further steps are the evaluation of the sound source data in combination with the tracking data, the development of the acoustic emission model and the integration into the simulation tool in accordance with a method and a computer-implemented program according to an embodiment of the present invention.

Sound source model for aircraft as a function of flight conditions

Based on the measurement setup and data sources described above, a sound source model for vehicles, in particular aircraft or aircraft, in accordance with a method and a computer-implemented program according to one embodiment of the present invention takes into account the respective flight conditions of the aircraft.

Input data - measurements and data sources

As already stated above, acoustic overflight measurements of real air traffic around a public airport, e.g. B. Zurich, made. In order to cover a wide range of typical flight conditions, microphones are placed both in the immediate vicinity of the airport, but also at distances of up to 20 km.

6 shows the measurement setup at a further distance from the airport including landing approaches from the south and two take-off routes that cover large and standard fuselage aircraft. Microphone locations 1-10 (dots) and typical flight routes for departures from runway 16 (continuous lines) and runway 28 (dashed line) as well as approaches to runway 34 (dash-dotted lines) are shown in FIG. 6. The ten independent locations are equipped with omnidirectional free-field microphones, which are arranged at a height of 10 m above the ground.

The microphone locations for departures are chosen such that they cover flight conditions after throttling, during acceleration and during continuous climb in cruise configuration. During the landing approach, the microphone locations were distributed along the glide slope while the aircraft was slowing down for the final approach.

Measurements in the immediate vicinity of the airport, within a distance of 2.5 km from take-off or touchdown, provide data for both the final approach and the initial climb in a wide range of radiation angles. Eight omnidirectional free-field microphones were installed at a height of 4 m above the ground for each of the three measured runway directions in the vicinity.

An optical tracking system and multilateration (MLAT) provide position data with great accuracy in the vicinity of the airport. In the further distance where the accuracy of radar data is sufficient, the latter were used. Flight recorder data with routes based on GPS are provided by international airlines (e.g. SWISS). The flight recorder data also provide air speed, engine parameters, aircraft orientation, configuration of the airframe, environmental conditions, etc. For aircraft types without available flight recorder data, the rotational speed of the engine N1 is extracted from short-term spectra of the acoustic measurements. In the present example measurements, flight recorder data are available from a total of six aircraft types with 161 to 673 flights and thirteen combinations of aircraft and engine types, so-called reference types, which are based on the N1 determination with 27 to 334 flights each (see Table 1 below).

Input data data processing

FIG. 7 shows the data processing that was used for the back-propagation of the measurements and the creation of the input data set for the acoustic emission model. The sound wave files are analyzed with a fixed time interval of 50 ms and filtered to third octave bands with 24 center frequencies from 25 Hz to 5 kHz in order to obtain the sound pressure level Lp (f). This frequency range is chosen to cover the characteristics of aircraft noise, which includes high acoustic energy in the low frequency range from the nozzle. Frequencies above 5 kHz are quickly attenuated due to the usually relatively large distances between the aircraft and the receiver and can therefore only be measured very close to the source.

In order to prevent backpropagation of background noise, only that part of the level-time protocol (of each third of an octave band) which is 6 dB above the minimum level before and after the event is selected for the analysis. In addition, events with undesirable noises are discarded. The sound propagation model sonX is modified and used to calculate the corresponding attenuations and the speed of the sound for each source-receiver combination. Source positions and flight parameters are synchronized to the acoustic data and corrected for the delay in the sound as it travels through the atmosphere. Geometric deviation, atmospheric absorption, ground effect, damping of vegetation and the influence of vertical gradients of wind, temperature and relative humidity are taken into account. For each flight, individual weather profiles from the well-known numerical weather forecast model COSMO-2 are used in order to reproduce the actual atmospheric conditions as precisely as possible. The coefficient of atmospheric absorption is calculated for certain plane surfaces with a maximum step size of 100 m.

Effects due to the movement of the source are corrected by applying a frequency shift and a level gain. In equation (8) below, the Doppler factor DF is defined as a function of the relative Mach number Ma from the source in the direction of the receiver, where θ is the radiation angle between the flight path and the vector from source to receiver. DF = 1 - Ma · cos (θ) (8)

Flight effects (FE), which can be categorized for sound sources that move with the aircraft, consist of the kinematic effect, which corresponds to the movement of the source in relation to the receiver (Doppler), and the dynamic effect, which corresponds to the Corresponds to movement of the source in relation to the medium of propagation. The combination of both effects leads to the level gain ΔLFE, which is defined in equation (9) below. ΔLFE = 40 log10DF <-1> (9)

Equation (10) below summarizes the backpropagation process for obtaining the sound emission levels Lem for each third octave band. According to the present invention, the sound emission level Lemals is to be regarded as equivalent to a sound power level Lw, the directivity D already being included. This explanation is necessary because only the lower hemisphere of the aircraft can be measured from the ground.

The backpropagation from the receiver to the source is carried out for all receivers (microphone locations) based on short-term sound pressure levels Lp which are corrected for the propagation attenuation ΣA and the flight effect ΔLFE. In accordance with the definition of the sound power level, the geometric deviation in Σ A includes the transformation constant log10 (4π). The frequency shift is defined by the frequency ratio of the source and receiver, which corresponds to the inverse Doppler factor. For third-octave bands, the frequency shift was implemented on the assumption that the energy was evenly distributed over each band. The frequency shift is applied after the backpropagation to the source position according to equation (10) above.

Input data - resulting data set

As a basis for the model development, a data record is prepared for all events of each individual combination of aircraft type and engine type. The level of detail required for classification is limited to general aircraft types, so that no distinction is made between optional equipment such as winglets or double ring combustion chambers.

For example, the A320 family from the aircraft manufacturer Airbus is divided into the types A319, A320 and A321, which differ primarily in terms of length and maximum take-off weight. Different engine options, either CFM56 or V2500, are available for each type within the group, implying six possible data sets when all combinations are measured.

Each data set consists of 24 subsets for the evaluated third-octave bands with center frequencies of 25 Hz to 5 kHz. A subset includes the corresponding emission level Lemin dB from the above equation (10), the calculated radiation angles θ, φ in degrees, flight parameters such as the rotational speed of the engines N1 in%, the Mach number Ma and the atmospheric parameters pressure p in Pa, temperature T in ° C , Density p in kg / m 3 and speed of sound c in m / s. If flight recorder data is available, the angles of attack and sideslip as well as the configuration settings are also available. In flight recorder data, the Mach number Ma could be related to the actual air speed (related to the moving air) of the aircraft, but for reasons of consistency it is related to the flight path speed (related to the ground) for all aircraft.

In addition, the identification number for the event and microphone are appended to each data point (line) for traceability during model development. An event with a flight segment of 60 s adds 1,200 data points per microphone to the data set. For all flights and measurement locations, this can amount to one to two million data points per subset.

For a comparability of the model prognoses with the measured data in section V, the data set is filtered according to the corresponding flight conditions and directional angle in each frequency band. For example, flight parameters for a typical flight phase and a relevant directional angle are selected in order to predict the sound emission level. Then the same parameters are used with a certain interval around each parameter in order to generate a subset from the complete data set (φ = 60 ° ± 5 °, N1 = 93% ± 2% etc.). Then the arithmetic mean Lembe is calculated and compared with the arithmetic mean of the forecast values (see below). The comparisons with measured data are not independent as the model is tuned to the same data set, but they allow an assessment to be made as to whether the model approach is suitable.

Model development

The model is created using multiple linear regression. This procedure enables the identification of the effects of various influencing parameters and their interactions in great detail. It can be applied to the sound emission level because the logarithmic levels are normally distributed. In a first step, outliers are removed from the data set (see section below). Then the parameters for the models are selected (see below). A novel process is then carried out to separate the data set into airframe and engine noise (see below). This separation of the two main source mechanisms is advantageous as it allows a more precise description of each mechanism. For example, both models can include different parameters or take into account different relationships of the same parameter.

Data processing, data analysis and adaptation of the source models can e.g. B. by using the standard software Matlab 2014b for mathematical calculations. The models are z. B. using the "Statistical Toolbox" from Matlab via the "fitlm" command, which uses a smallest square adjustment (OLS) and enables an individual evaluation of the data points, balanced.

Runaway

Before the model coefficients are estimated, outliers are removed by an adaptive outlier detection which uses the robust Mahalanobis distance (RD) for the automatic detection of outliers. One advantage of this method is the adapted threshold, which is adapted to the sample size.

If the data set originates from a multivariate normal distribution, no outliers would be recognized in relation to a fixed threshold.

Data exploration

The available parameters of the presented data set are analyzed by means of exploratory analysis and knowledge from the literature. The findings are used to select the most important parameters for the statistical model presented below. In addition, parameters are discarded if they correlate with other parameters in order to ensure reliable model coefficients. Finally, the relationship of the parameters to Lem is disclosed in order to satisfy the linearity of the statistical model.

In many engines, N1 is the power setting control parameter. In contrast to the thrust or jet speed of the engine, this is a directly measurable parameter. It correlates with the jet speed, which can be regarded as the main physical cause of nozzle noise. At frequencies below 1 kHz, where mixed jet noise dominates, N1 can be used as a substitute for jet velocity. In addition, the leaf passing frequency (BPF) and the fan broadband noise with a center frequency of 2.5 times the BPF [20] are directly related to N1. For most engine types, the BPF and thus the broadband noise are above 1 kHz, so that N1 represents an appropriate prediction value for the entire engine spectrum.

Figure 8 is graphs depicting the results of engine test runs performed on the exemplary A330-300 (TRENT772B) aircraft for two exemplary third-octave bands with Lp measured over a 170 m radius in four different directions. The test runs were carried out by SWISS, measurements were taken by EMPA and Zurich Airport. Permission to use the data was kindly given. An exemplary engine test run of an A330-300 with TRENT772B is evaluated to establish the functional relationship between sound pressure level and N1 for each third of an octave band, since no such relationship is known. The stationary aircraft excludes all airframe noise sources and flight effects from the measurements. Short-term linear sound pressure levels (Lp) are measured at four microphone locations within a radius of 170 m around the aircraft. The engine test is performed twice up and down by ramping up six different engine loads from idle to the highest possible engine pressure ratio on the ground. Thus, for each time interval with constant N1, two to four mean Lp are available (Fig. 3). A regression model was tuned for each direction, 0 ° corresponds to the nose of the aircraft, N1 is tuned as a second-degree polynomial.

According to diagrams a) and b) in FIG. 8, which show two exemplary frequency bands, the functional relationship changes with the frequency and the direction. At 31.5 Hz, the fit curve is slightly parabolic at the front and almost linear towards the rear of the aircraft. In addition, the gradient increases towards the rear. At some frequencies, for example 2 kHz, the parabola actually opens downwards for about 15 °. For frequencies in between, which are not shown here, the relationships are very similar with predominantly linear or slightly quadratic behavior (open upwards). Thus, a flexible second-degree polynomial enables the relationship between Lpbzw. Lemund to represent N1.

9 shows two diagrams which illustrate an exemplary influence of N1 on Lemdes exemplary aircraft E170, equipped with the turbo engine CF34-8E, for 2 kHz. The data points are filtered to 0.2 <Ma <0.24, but airframe noise can affect levels at low N1. In general, the turbofan engines of today's civil aircraft are very similar, and the mechanisms of noise generation are the same. It can therefore be assumed that the quadratic approach is also valid for other turbojet engines. This assumption can be tested exploratively using the data set backprobagged in FIG. 9 for the Embraer E170 (CF34-8E) engine at 2 kHz. Although airframe noise is included in Lem for a low N1, the same trends can be seen. In diagram a) in FIG. 9, the sound emission towards vom is a downwardly open parabola, while the opposite applies to the radiation angle in diagram b) of FIG. 9. This fits well with the results of the test run.

The Mach number Ma = U / c was selected in order to take into account the speed-dependent sound sources. It shows the mean flow velocity U at the source and the local velocity of the sound c in a single, dimensionless quantity. Furthermore, the Mach number is an aerodynamic figure that is interpreted as a compressible flow condition and thus ensures comparable flow phenomena.

The dependence of the sound emission on Ma can be provided by an aeroacoustic analogy. The generation of sound by the fluctuating fluid is described by the classical wave equation, which is expanded to include three basic source terms: monopole, dipole and quadrupole. The theoretical free-field solutions, which are obtained, for example, using equation (11) below, show that the sound power W is proportional to the air density p, a characteristic variable of the source D, the mean flow velocity U and the Mach number Ma. The exponent x depends on the source mechanism (monopole x = 1, dipole x = 3, quadrupole x = 5). W α ρD <2> U <3> Mα <x> (11)

In order to derive the relationship of the sound power level LW, the logarithm for base 10 is used in accordance with equation (12) below. The units of the parameters are normalized with ρ0, D0, U0.

The dependencies of equation (12) are carried over to Lem, which as defined in equation (10) is proportional to LW. In a statistical model, the implementation of U and Ma is problematic because these parameters are highly correlated (multi-collinearity). Multicollinearity can have a profound effect on the estimates of the regression coefficients. Even if the model equation is still useful in its known intervals, the individual effects of the parameters can only be estimated poorly and would lead to incorrect extrapolations. In order to prevent multicollinearity, only log10 (Ma) is taken into account instead of log10 (U). Thus, the regression coefficient reflects the power x of Ma at which the total airframe noise scales. In contrast, engine noise is primarily influenced by the effects of the reduced relative jet speed. To enable an extrapolation of the model for the take-off run, a simplified correlation of L ∝ U by [22] can be used. Thus Lem∝ Ma is assumed for engine noise.

10 shows two diagrams which illustrate an exemplary influence of Ma on the sound emission level of the aircraft type A320 during a) take-off and b) approach for landing at 250 Hz. The dashed line is a linear regression in a) and a logarithmic regression in b). The exemplary dependence on Lemvon Ma at 250 Hz for typical flight conditions at take-off with high power is shown in diagram a) and the landing approach with idling is shown in diagram b) in FIG. 10. It is to be expected that engine noise dominates in diagram a) , and a linear regression seems adequate to extrapolate the decrease in Ma to zero. In contrast to this, it is to be expected that the airframe noise dominates during the landing approach in diagram b). However, a regression with the logarithm to the base 10 of Ma is appropriate as a linear regression, but a linear approach would not reflect the physical properties of the airframe noise and thus overestimate the levels for lower and higher Ma.

An air density effect is included in the airframe model to accommodate the generalized aeroacoustic theory in equation (12) above. The air density correlates strongly with air pressure p and temperature T, which are excluded from the regression model in order to prevent multicollinearity of the forecast values. The multicollinearity can be checked by means of the variance inflation (VIF), which can be determined for each possible parameter. After discarding p and T, there is no strong multicollinearity left. Nevertheless, it can be decided to reject the speed of the sound c, since it is already included above Ma. Indeed, after discarding c, the VIFs of Ma and p continue to decrease. It is thus possible to select only p as a variable for the statistical model. From the above equation (10), the logarithmic conversion is applied to p. The transformation may be necessary to ensure linear behavior with respect to the airframe model coefficients. The Mach number and the air density can be converted logarithmically, i. H. the linear coefficients represent the exponents of the variables within the logarithm. The air density can be normalized by the density at mean sea level, as defined by the International Standard Atmosphere ISA (ρ0 = 1.225 kg / m <3>), in order to obtain a dimensionless size of the variable. At zero air velocity or density, the deformations tend towards minus infinity, which is physically appropriate. In practice Ma can be set to 10 <-3> to get a real value, and p should never be extrapolated that much. lMa = log10 (Ma) (13)

The sound emission of an aircraft has a directional effect that can best be described by means of spherical coordinates. In particular, the longitudinal radiation, which is represented by the polar angle θ, changes greatly with the type of aircraft, the frequency band and the power setting, as shown in FIG. 10. The transverse radiation, represented by the azimuth angle φ, is also taken into account as a function of the aircraft type, frequency band and power setting. The directional effect in the transverse direction can lead to level differences of up to 3 dB above φ. In addition, the directional effect in the transverse direction can have considerable discrepancies with the generalized corrections, which only differentiate between engines mounted on the wings and on the fuselage.

A second order Fourier series can be selected to represent the directivity in the longitudinal direction. A higher order was also tested during model development, but resulted in problematic slopes at the boundaries where less data is available. The directivity at high frequencies was particularly critical for conditions at a great distance from the receivers. The directivity in the transverse direction is modeled with a half Fourier series (second order) in order to simplify the number of terms and also to prevent problematic slopes in areas with poor data coverage.

11 shows two diagrams which represent an exemplary distribution of measured flap lever positions of the aircraft type A320 as a function of the Ma number. In contrast to the landing approach, the flap lever position 1 relates to a different deflection angle of the flaps during take-off. The configuration of the aircraft is modeled by three categorical variables: position of the landing gear (retracted: 0, extended: 1), position of the flap lever (0 to 4, fixed combinations of slat and flap deflection) and use of the airbrakes (deactivated: 0, activated: 1). This information is taken from the flight recorder data. Of course, due to the measurement of actual air traffic, the data is not balanced and it may not be possible to obtain data for all configuration combinations. Furthermore, the flap settings correlate strongly with different intervals of the Mach number due to processes and structural constraints, as shown in FIG. 11. In addition, the flap lever position 1 for take-off is specified as 1 + F, which corresponds to a different deflection angle of the flaps than for the landing approach (10 ° instead of 0 °). Despite all the difficulties, the influence of the aircraft configuration is of interest and is therefore taken into account with the help of a specific model structure, which is described in more detail below.

12 shows two diagrams which illustrate an example of the influence of the landing gear on the noise emission of the A320 during the approach to landing in idle for 250 Hz in (a) and 2 kHz in (b). Measured data and the regression lines show a clear effect on the noise emission when the landing gear is extended. The data points shown in FIG. 12 for representing the effect of the landing gear come from measurements for approach situations with the landing gear retracted and extended. Each data set is matched with a simple logarithmic regression to show the influence of the chassis. In diagram a) of FIG. 12, a slightly larger gradient of the regression with the landing gear extended can be seen. At low Ma the emission levels are similar, but at high Ma of 0.3 the level difference is 2.6 dB. At 2 kHz in diagram b) of FIG. 12, the effect of the chassis on the Ma-dependency of the sound generation is significantly higher. At Ma = 0.3 the difference is already 6 dB.

Further parameters can be considered or discarded during model development due to insignificance or for practical reasons. In particular, the angle of attack and the slip angle, which are available from the flight recorder data, can be excluded, since no correlations to the emission level can be found. In addition, discarding both angles is also appropriate as they are usually not available for forecasting.

Data separation process

Fig. 13 is a diagram showing steps of a process for model development and data separation according to an embodiment of the present invention. The process is repeated for each third octave band. After the exemplary process shown in Fig. 13 to show the model development and the separation of the total emission levels from a frequency band in the data set in airframe and engine noise, in a step 1 the data set is separated into two parts: Part one contains all Data when the engines were idle, i.e. H. only landing approach situations, and part two contains all other data from landing approach and take-off with engines under load. A reasonable threshold for the division is determined by data plots of Lem over N1, as in Figure 14, where no correlation below 40% can be seen for N1. Thus, it can be assumed that the airframe noise dominates the overall Lem for this subset.

In step 2, an airframe base model and an engine base model are matched to their respective data sets to reveal the main effects of each source. The starting models comprise only minor changes compared to the models described in more detail below. With the aid of the predicted initial sound emission levels for airframe noise and engine noise, a source ratio can be calculated for each data point in the original data set (step 3). The ratio q '(see equation (15) below) is defined as the predicted noise emission of the engine divided by the sum of the predicted engine and airframe emissions. Thus, a ratio of 0 means that only airframe noise contributes to the total emission, and 1 corresponds to an exclusive engine noise emission. It should be noted that all forecasts from these models are marked with a caret to distinguish them from input data; the superscript i indicates that the original models are used.

However, this method implies that the output models must be extrapolated. For example, the initial airframe model is based on approaches with a Mach number less than 0.35. In contrast, starts are measured up to a Mach number of 0.45. As a result, the model must be extrapolated to forecast for each data point in the full data set. However, the extrapolation of the Mach number is based on physical knowledge (see above), so that a plausible first estimate is possible.

Based on the ratio q <l>, two separate data sets, each including all measurements for landing approach and take-off, are generated for each third octave band (step 4). One gives the noise emission from the engines (see equation (16) below), and the other gives the noise emission level from the airframe (see equation (17) below).

Figure 14 shows three diagrams illustrating a data separation example for the A320 at 100 Hz. The backproagged data in the upper area is split to estimate the noise emissions associated with airframe noise (bottom left) and engine noise (bottom right). In Fig. 14, both data sets of the original learning are compared with one another. The airframe levels (bottom left) dominate for N1 <40%, which corresponds to the implication of the assumption in step 1. In the example they are about 20 dB below the total levels for departures. In contrast to this, the engine levels dominate the take-offs (bottom right) and lose their influence as the N1 falls. Overall, the original data set can be reconstructed by adding the individual levels of both data sets energetically.

In step 5, the final airframe and engine noise models as defined herein are matched to the separate data sets presented in FIG. Steps 3 to 5 are repeated once in order to improve the estimate of the ratio between airframe and engine noise, since the regular models, in contrast to the original models, have been matched to the entire data set. Finally, the energy sums of the airframe model and the engine model add up to the forecasted Lem. The complete process of data separation and model adaptation is done 24 times for each third octave band.

Source models

Source models according to an exemplary embodiment of the present invention based on findings as described above are presented below. The models are based on knowledge of the main parameters and their interactions, e.g. B. the Mach dependency of the chassis built. From a statistical point of view, this approach is similar to the forward selection with a scope that only includes the relevant parameters for both sources. Each model with a new parameter or a new interaction is compared with the coefficient of determination R <2>, the mean square deviation σe and the Akaike information criterion (AIC). All criteria should be compared across all frequencies in order to find a global optimum. The agreement with the model assumptions was confirmed visually by means of residual plots.

By way of example, the source models are built on the A320 data set, which provides flight recorder data and a large number of flights. They can then be tested and further improved using the five other exemplary aircraft types with flight recorder data in order to confirm that the models can be applied to different aircraft and engine types. The resulting models are presented as advanced models (see below). If no flight recorder data are available, the configuration is unknown to the aircraft. b. thirteen reduced models are created (see below).

Advanced models

The sound emission level of the airframe can be modeled by the sum of the source terms and the radiation angle terms as summarized in equation (18) below. The dependence on the frequency f indicates that all coefficients of the source and radiation angle terms are matched to all third-octave bands, even if this is not expressly noted for reasons of readability.

Equation (19) below gives the source terms of the airframe model. La0 is the captured intercept, and aa1 to ea3 are the frequency dependent coefficients of all model parameters. The main parameters are the logarithmic transformations IMa and Ip, which reflect the behavior of the aeroacoustic sound generation in accordance with known semi-empirical models. In addition, every change in the configuration of the landing gear, flaps or airbrakes (SB) is modeled in discrete steps, as these parameters are categorical. In the case of flaps and airbrakes, interactions with the landing gear are also considered in order to take into account changes in the absolute effect if the noise emission level is increased by the extended landing gear. In addition, the landing gear and airbrakes interact with IMa in order to take into account the speed-dependent generation of noise. In the case of the flaps, the interaction with IMa is neglected, since each flap position is only used in a certain small range of Mach numbers (see FIG. 11), so a Ma interaction can only be determined with great uncertainties. Nevertheless, the very different combinations of flap positions and Mach number range for landing approach and take-off in FIG. 11 must be taken into account in one way or another in the model. For this purpose, another categorical parameter, the flight procedure Proc (departure: 1, approach: 0) can be introduced, which also takes into account the various deflection angles. The parameter also takes into account the observation on the A320 family that frequency bands with cavity tones can show a sharp rise in level during landing approaches with increasing Mach numbers, while this is not the case during takeoffs. It can be assumed that the local flow field differs from the take-off due to different angles of attack and flap positions during the landing approach.

The directivity of the airframe model (see equation (20) below) is expressed as axially symmetrical radiation along the longitudinal axis of the aircraft, since the directivity in the transverse direction is mainly associated with reflections of the engine noise at the airframe (see below). The polar angle θ is taken into account with a second order Fourier series in order to model the directivity in the longitudinal direction. The airframe directivity coefficients are kabis na. Interactions are not included; H. the shape of the emission directivity is the same for all flight conditions. This simplification is justified because the data set has already been corrected with regard to the flight effect (see equation 10 above).

The sound emission level of the engine noise Lbem, eng (f) is modeled by the sum of the source terms and a more detailed approach for the radiation angle terms as summarized in equation (21) below.

Source terms for engine noise (see equation (22) below) include the intercept Le0 and three parameters with their coefficients ae1, be2. The main source term for engine noise is N1. The square approach for N1 reflects the nozzle and fan noise as detailed below. In addition, the parameter Ma takes into account the change in the source strength of the jet mixing due to the aircraft speed.

The exemplary engine test run shows that the relationship of Lem to N1 is heavily dependent on the polar angle θ. Thus the Fourier terms of the directivity interact with N1 as well as with N1 <2> (see equation (23) below). The corresponding model coefficients are kej to ne, J with the index j for each interaction. The directivity in the transverse direction (see equation (23) below), which reflects the installation effect, is a half-Fourier series of the second order, i.e. H. included exclusively with sine mini of φ. Analogous to the directivity in the longitudinal direction, each term has an interaction with N1 with the coefficients oe, J, peJ. Including the interaction with N1 <2> can also be tested, but no significant improvement is to be expected from the present example.

With the approach of equations (23) and (24) listed above, the shape of the 3D directivity is enabled to change with the engine setting N1. Since each third octave band is tuned separately, the spectral content of the overall directivity also varies with the power of the engines.

The airframe base model used for the first calculation of the ratio q 'for the data separation as set out above is identical to equation (19) above, but the parameter Proc is omitted because all data points of the base model are related to landing approaches belong. In addition, the directivity in equation (19) above is not changed. The engine output model as in equation (22) above is reduced by the parameter N1 <2>. This prevents an increase in the sound emission level for the extrapolation for N1 below 40%. For this purpose, the directional effect in the longitudinal direction in the above equation (23) is also reduced by N1 <2>.

Reduced model

If no flight recorder data are available, all models must be reduced by the unknown parameters. If, for example, the configuration parameters for airframe noise are missing, they must be removed from the model approach (see equations (25) and 26) below). The effects of the configuration are still implicitly present in the data record. In contrast to the further developed model, the Mach dependency on Lemnun takes the averaged configuration of all measured flights into account. Thus, the parameter Proc is retained in the model for the reasons outlined above. The radiation angle terms remain unchanged (see equation (20) above).

In the case of engine noise, the parameter originating from the flight recorder data is N1. This parameter is crucial because it has the greatest impact on Lem. For this purpose, in the case of missing flight recorder data, N1 can be determined based on spectral analyzes. The engine model of the reduced model is thus the same as that of the further developed model (see equations (21) to (24) above).

weighting

During measurements, the polar angle θ changes only slowly in the case of a distant aircraft, whereas it changes quickly when the aircraft flies over the microphones. As a result, only a few data points of the evenly spaced acoustic samples in the most relevant range of θ are available and vice versa. For this, the data must be weighted in order to reduce an influence of the inhomogeneous distribution of data points over θ, which is inversely proportional to the time derivative θ '= dθ / dt. To do this, the model is provided with a weighted least squares (WLS) algorithm analogous to the linear sinus regression of the usual least squares method.

Since each flight-receiver combination has a different geometry, θ 'is standardized by the maximum value per event and receiver, notated as w0 in equation (27) below. The standardization prevents a higher weighting of measured levels when an aircraft is closer to a receiver than when it is further away and θ 'is generally lower. The weights are then normalized by their mean w0 to ensure that the sum of all weights w1 used for the WLS algorithm corresponds to the number n of observations included in the analysis (see equation (28)). w0, i = θ '/ max (θ') event, receiver (27)

Energy correction

As a result of the least squares estimation, the model predicts the arithmetic mean of the sound emission level Lemin dB. A correction is necessary in order to forecast the averaged energy Le, which corresponds to the arithmetic mean of the noise emissions in watts. Since Lemnormal is distributed, which is a prerequisite for a linear regression, the energy correction can analytically be 0.115. σ <2> can be determined. The variance σ <2> can be represented by the error variance estimate, applied to each model according to equations (29) and (30) below.

FIG. 15 shows a diagram illustrating an exemplary correction factor for a mean energy versus frequency for the aircraft type A320. It can be seen from FIG. 15 that the error variance in microphones at a greater distance is significantly higher than in microphones in the close range, which is primarily caused by a higher variance due to turbulence and uncertainties in the back propagation. In general, microphones for departures are further away from the source, which supports the interpretation that the variance is distance-dependent. The energy correction should only include the variance of the sound emission level and the uncertainty of the measurement. For this purpose, the correction is applied to the error variance of the data in the close range (total error of landing approach and take-off).

The energy sum (⊕) of the agreed end models for airframe (see equation (29) above) and engine noise (see equation (30) above) gives the predicted total emission spectra of the modeled aircraft according to equation (31) below.

Reference types

In the present example, a total of 19 acoustic reference types are created on the basis of the input data presented above. Table 1 below gives an overview of the reference types with details on the corresponding aircraft and engine types, the origin of the data and the number of flights on which the models are based. For example, the entry A320_CFM56-5B, the data of which is often used in the description below for illustrative purposes, is based on flight recorder data from the CFM56-5B-equipped A320-200 with a total of 673 flights. The entry E170 without flight recorder data is based on a total of 89 flights and is used to prove the feasibility of the reduced model.

Table 1: Reference types with grouped aircraft and engine types and their database (number of flights). If the input data are based on the flight recorder, the further developed model (Adv.) Is created, for other types with N1 determination the reduced model (Red.) Is created.

A319_CFM56-5B Airbus A310-100, CFM56-5B FDR Adv. 120 41 161 A320_CFM56-5B Airbus A320-200, CFM56-5B FDR Adv. 424 249 673 A321_CFM56-5B Airbus A321-200, CFM66-5B FDR Adv . 300 126 426 A32X_CFM56-5A Airbus A320-Family, CFM56-5A N1 Red. 57 16 72 A32X_V2500 Airbus A320-Family, V2500 N1 Red. 198 33 231 A333_TRENT7 Airbus A330-300, TRENT7 FDR Adv. 249 136 385 A343_CFM56-5C Airbus A340-300, CFM50-5C FDR Adv. 166 120 286 A388_GP7270 Airbus A380-800, GP7270 N1 Red. 26 2 28 A388_TRENT9 Airbus A880-800, TRENT9 N1 Red. 38 20 58 B737_CFM56-3 Boeing B737 Classic (-300 to -500), CFM50-3 N1 Red. 84 39 123 B737_CFM56-7B Boeing B737 NG (-600 to -900), CFM56-7B N1 Red. 297 37 334 B763_PW4060 Boeing 787-300, PW4060 N1 Red. 9 34 43 B76X_CF6 -80C2 Boeing 767-Family (-200 to -400), CF-80C2 N1 Red. 19 57 76 CRJ9_CF34-8C5 Bombardier Regional Jet CRJ-900, CF34-80C2 N1 Red. 71 22 93 E170_CF34-8E Embraer ERJ 170, CF34 -8E N1 Red. 62 27 89 E190_CF34-10E Embraer ERJ 190, CF34-10E N1 Red. 243 49 292 F100_TAY650- 15 Fokker 100, TAY050-15 N1 Red. 234 61 295 FA7X_PW307 Dassault Falcon 7X, PW307 N1 Red. 17 10 27 RJ1H_LF507 BAE SYSTEMS AVRO RJ-100, LF507 FDR Adv. 324 202 526

If the number of flights measured is small, some types of aircraft from the same aircraft family can be grouped. In general, only types with an identical engine are grouped, since the engines represent the main source of sound, which can lead to considerable deviations in noise emissions. For example, all sub-types of the B737 are grouped with the classic CFM56-3 engine option, while all types of the new generation equipped with the modern CFM56-7B are grouped separately. Grouping a family of aircraft is appropriate and improves the model, as a larger range of parameters is covered due to different take-off weights and processes. In particular, different types of aircraft use different N1s for take-off, since different take-off weights require a different thrust.

Performance of the models

16 shows two diagrams which illustrate an exemplary determination coefficient over frequency. Airframe and engine models are compared to their separate datasets, while the overall model is compared to the original dataset. FIG. 11 shows the coefficient of determination R 2 versus frequency for all third-octave bands. R <2> total represents the performance of the model from equation (24) in order to reproduce the original data set of the backprojected data Lem. Overall, the goodness of fit of the overall models shows good informational value of the regression adjustment with values between 0.7 and 0.8 for the A320 (a), but also for the E170 (b). R <2> af and R <2> closely describe the quality of fit of the source models to the separate data sets. For most of the frequency bands, the engine model shows R <2> eng values that are slightly above 0.8. In contrast, the airframe model of the A320 R <2> afM values between 0.2 and 0.6 with a much higher fluctuation between the different frequency bands. In the case of the E170, R 2 varies mainly between 0.4 and 0.7.

A specific aspect of R 2 is related to the frequency range in which the sound sources radiate. For example, R 2 narrow of the A320 in diagram a) of FIG. 16 is between 50 and 400 Hz high when the nozzle noise is dominant. Similarly, R <2> is close to 2 to 3 kHz, the bands that contain the A320's Blade Passing Frequency (BPF) at departure. Airframe sound sources can be identified in the same way. In accordance with measurements on a full-size A320 wing in the low-speed DNW-LLF wind tunnel, the slats (including the Flaps parameter) emit considerable sound between 100 and 300 Hz. Furthermore, a distinctive cavity tone can be seen in the wing at 500 and 630 Hz. And finally, there is additional noise on the side edge of the slat between 1 and 1.6 kHz.

It is precisely in these frequency ranges that R 2 has local maximum values. In contrast, there are no explicit sound sources for 50 to 100 Hz and above 1.6 kHz, and consequently R 2 is afm low. No corresponding measurements in the wind tunnel as for the A320 are known for E170, but the measurements indicate a tone at 100 Hz, which also increases R <2>. In general, the E170's airframe sound sources contribute across all frequencies because R <2> is afm high.

Model comparison

Exemplary results of spectra and directional characteristics are shown for different flight conditions and compared with measurements.

17 shows two diagrams which illustrate exemplary spectral directional characteristics of the aircraft type A320 during departure with high-performance setting. In FIG. 17, spectral directional characteristics are predicted and compared with averaged measurement data with flight parameter settings similar to those set out above. A departure with a high performance setting of N1 = 93% is shown. The directivity in the longitudinal direction in diagram a) of FIG. 17 is presented to a viewer on the side at φ = 60 °. All frequency bands show good agreement between predicted and measured data. Frequencies in the low range such as 125 and 250 Hz show a typical nozzle jet characteristic that is clearly pronounced towards the rear. At high frequencies such as 2 kHz, the characteristic shows local maximum values in the front and rear areas due to the fan. Thus, the Fourier series and the interactions with N1 and N1 <2> enable the model to accurately reproduce the directivity in the longitudinal direction.

In the same way, the directivity in the transverse direction, which is shown in diagram b) in FIG. 17 from the rear at θ = 130 °, agrees well with the measurements. Overall, the radiation in the transverse direction is less important and less variable across the frequency bands. Thus, the half-Fourier series with interaction with N1 represents a sound approach. Directional effects in the longitudinal and transverse directions also agree well with the overall learning, which supports the chosen model approach for the directional effect.

18 shows two diagrams showing exemplary spectra for final approach (N1 = 55%, Ma = 0.23) and take-off with high-performance setting (N1 = 93%, Ma = 0.24) for the A320 aircraft. FIG. 17 shows spectra for typical flight conditions at take-off and final approach for θ = 90 ° in diagram a) of FIG. 18 and θ = 130 ° in diagram b) of FIG. 18 pictured. As expected, the predicted dominant sound source at start-up is the engine. The airframe model is not shown in this case in order to avoid an overlay with the spectra of the final approach. In addition to the strong, low-frequency broadband noise of the nozzle in the rear area, the leaf passing frequency at 2.5 kHz is also well available. In contrast, the airframe model is relevant for the final approach. At 90 ° the middle frequencies are still dominated by engine noise, presumably turbo engine noise. At 130 ° (rear), airframe noise dominates the medium frequencies. Interestingly, while the nozzle noise is very low for N1 = 55%, the fan broadband noise is in the range of about 2.5 times the leaf passing frequency (2.5 × 1.65 kHz = 4.1 kHz) by about 4 kHz in the engine spectra in diagrams a) and b) of FIG. 18 can still be seen.

19 shows two diagrams which represent exemplary spectra for final approach (N1 = 50%, Ma = 0.2) and take-off with high-performance setting (N1 = 86%, Ma = 0.24) for the aircraft E170. The corresponding spectra of the aircraft E170 in FIG. 19 depict results that are very similar to those shown for the A320 in FIG. 18. The overall spectra agree well with the measurements, and take-offs are dominated by engine noise. With typical final approach settings, in contrast to the source-dependent contributions as in the A320, engine and airframe noise contribute equally across all frequencies. The leaf passing frequency for takeoffs at 3 kHz is not observed in both diagrams a) and b) of FIG. 19, which is in agreement with the measurements. For the final approach, the blade passing frequency in diagram a) of FIG. 19 can be found at 1.7 kHz both in the prognosis and in the measurements.

The fact that each frequency band is tuned in turn enables the model to form different spectral shapes. For example, the quadratic dependence on N1 enables the engine spectra to change from high nozzle noise to low power settings, as is shown in FIGS. 18 and 19 in particular in the rear area of the aircraft. In addition, bands with tones that occur from a uniform broadband spectrum are also taken into account.

As a result, according to the present invention, a new acoustic emission model is presented in order to overcome the limitations in the prognosis of aircraft noise models according to the prior art with a moderate number of necessary input parameters. As a significant advantage of the present invention in comparison with known models such as Doc. 9911 or FLULA2 Airframe noise, represented by the influence of the aircraft's Mach number and configuration, can be modeled separately from the engine noise. At the same time, only a minimum of process parameters are required for engine noise, which do not require detailed knowledge of the engine power such as flow rate or jet speed, as is the case in semi-empirical models according to the state of the art such as ANOPP or PANAM.

According to the present invention, jet and fan noise of the engines with N1 are considered as the main parameter that can be determined acoustically in order to develop the respective models. A novelty according to the present invention enables the overall acoustic emission levels in airframe and engine noise to be separated. Thus, no complex microphone arrangement measurements are required to separate the two sound sources. The separation enables airframe and engine noise to be forecast independently of one another, which is an advantage when assessing noise abatement procedures. On this basis, new combinations of modeled airframes and engines for which input data are missing or which have not yet been flying in such a combination can be created. A further refinement towards the detailed modeling of individual sound sources still represents a challenge, especially in the scope of the coverage of a wide range of aircraft types with different dimensions.

The parameters chosen for the exemplary embodiments described herein make it possible to adequately reproduce the directivity and spectra for typical flight conditions. In particular the aircraft type A320 with flight recorder data and a high number of flights and the aircraft type E170 without flight recorder data and with only 89 flights show similarly good results. This is the case for all types listed in Table 1 above. Thus the further developed models as well as the reduced models are comparable and enable the creation of models for different input data and aircraft types.

Prerequisites for creating the model according to the present invention for other types of aircraft are (i) measurements at various locations close to the airport and at greater distances, (ii) backpropagation to the source and (iii) spectral analyzes to determine N1 or processing of black box data. One limitation of the separation of airframe and engine noise is the assumption that airframe noise dominates when the engine is idling. In addition, the validity of the separation cannot be proven because no data are available.

Since the source model is based on the sound emission level, it can be combined with any sound propagation model in order to calculate the sound immission at the receiver. However, the dispersion model should take into account all effects that were used in backpropagation (e.g. ground effect). Using a simulation program, new procedures such as the CDA landing approach procedure can be optimized acoustically, which could expand current studies on path optimization with noise as a cost function. The sound source model thus makes it possible to calculate various noise levels such as the actually perceived noise level (EPNL), the event level and the maximum level with a free choice of spectral weighting.

A sound source model according to the present invention for turbofan-powered aircraft closes the gap between conventional models and top-of-the-range models according to the prior art. It provides two separate models for airframe and engine noise emissions. Both were created based on knowledge of explanatory data analysis, physical knowledge and statistical models. Source models for a wide range of relevant aircraft and engine types have been created.

The flight parameters that link the sound emission level with the current flight conditions are, in the best case, flight recorder data or, alternatively, radar data with an additional analysis to determine N1 via the fan speed. If available, it is recommended to use the advanced model for investigations of single flights. Nevertheless, the reduced model can also be used in a similar way, although the effect of the flight configuration cannot be represented exactly. The reduced model would improve the accuracy of the current noise maps at least for airport scenarios and annual calculations.

The presented exemplary embodiments demonstrate the ability of a model according to the present invention to conduct studies on noise abatement methods. It is possible to provide data from optimizations or flight simulators in order to calculate and compare the noise mass and the affected population. The presented methodology can also be used without separating the sources to develop acoustic emission models for helicopters, propeller-driven aircraft or military jets.

Three-dimensional spectral directivity depending on flight conditions

As described above, a semi-empirical sound source model according to the present invention is constructed using multiple linear regression. This requires an adequate empirical data set to match the model coefficients to a typical range of flight conditions. In particular, measurements of different flight conditions and a wide range of angle covers are necessary to create a model according to the present invention.

Data processing

The sound pressure levels (SPL) at the measurement sites can be backpropagated to the source to obtain source power distribution levels SPL using the sophisticated sonX propagation model and taking into account the actual atmospheric conditions. Then the Doppler effect due to the movement of the source must be corrected by applying a frequency shift and an intensity gain. The resulting acoustic data, directional angle and Lw, must be synchronized and supplemented with flight recorder data, which provide a variety of parameters such as the course of the path, the rotational speed N1 of the turbine and changes to the configuration of the landing gear and flaps.

As already stated above, a complete model can be formulated for each third octave band from 25 Hz to 5 kHz, noted as parameter f in the description below. The directivity is modeled three-dimensionally using spherical coordinates consisting of the polar angle θ and the azimuth angle φ. Both angles are defined in relation to the flight path axis system as shown in FIG. 1 in order to simplify the requirements for input data for a prognosis, since the actual orientation of the aircraft is generally not known. The azimuth φ is also corrected by the bank angle of the aircraft, which can be calculated from the trajectory.

The sound source model can be divided into two sub-models, one for engine noise and another one for airframe noise, which, as explained above, each take into account different source mechanisms. An engine model according to the present invention, as described below in equation (32), includes both angles of directivity in order to take into account the directivity in the longitudinal direction and the lateral installation effect of the engines. The most important parameter is N1, which reflects the power of the engine and is therefore directly related to the jet speed and the rotation of the fan and turbine. With the help of engine test runs, as described in detail above, the conclusion can be drawn that Lwauch is dependent on N1 <2>. Using interactions between N1 and θ as well as φ, the directivity can change with the power setting. In addition, the Mach number Ma of the aircraft, which influences the flow and in particular also the source strength of the nozzle due to the surrounding flow, is included in the model.

Airframe noise can primarily be modeled with a two-dimensional directional effect, since the measurement setup for landings in which all aircraft follow approximately the same trajectory does not allow a reliable determination of lateral effects. The most important constant explanatory variables are the Mach number Ma and the density p of the surrounding medium, both of which are transformed by the logarithm to the base 10. The transformations are required to linearize the variables and correspond to semi-empirical models for airframe noise. In addition, the model includes various categorical variables to allow for changes to the configuration, i. H. Undercarriage position, flap position and airbrakes SB to reproduce. The complete model also includes interactions, e.g. B. of the landing gear and the Mach number, in order to take into account the difference in the dependence of the air speed with and without the landing gear extended. Finally, the “Procedure” (Proc) factor is included because it has been shown that some effects such as the change in level of the flaps are different in take-off and landing situations. Another angle of attack and thus the lift coefficient could be the cause, which has a direct impact on the noise generation.

Both models for airframe and engine noise are finally combined energetically in order to obtain the total source power distribution level as a function of the parameters considered. Since the model parameters can be obtained from an adaptation by a mean-least-squares algorithm and thus reproduce the arithmetic mean of Lw, an energy correction is introduced in order to correct the mean source power as described in detail above. The Doppler frequency shift and intensity amplification are then applied to take the flight effect into account.

The complete model (see equations (32) to (34) above) is created for six exemplary aircraft and engine type combinations from Swiss International Air Lines for which flight recorder data were available (see Table 2 below). A large number of flights are measured in order to create the present models on a statistically relevant basis, but a smaller number of flights would also be acceptable.

Table 2: Overview of acoustic reference types with flight data recorder and number of flights that were integrated into a model according to the present invention.

A319_CFM56-5B Airbus A319-100 with CFM56-5B 120 41 161 A320_CFM-5B Airbus A320-200 with CFM56-5B 424 249 673 A321_CFM56-5B Airbus A321-200 with CFM56-5B 300 126 426 A333_TRENT7 Airbus A330-300 with TRENT 700 249 136 385 Λ343_CFM56-5C Airbus A340-300 with CFM56-5C 166 120 286 RJlH_LF507 BAE SYSTEMS AVRO RJ-100 with LF507 324 202 526

As already stated above, model approaches according to the present invention make it possible to reduce the depth of detail if the input data lack parameters, which is the case when no flight recorder data are available. As a result, the number of parameters in the airframe model decreases.

The engine model remains in shape as N1 is the most important parameter that should not be removed. If no flight recorder data are available, it can be determined by recognizing the leaf passing frequency with the help of extended signal analysis. In the event of insufficient coverage of the transverse directional straightening angle, the model can simply be reduced to two-dimensionality by removing the azimuth angle φ from the engine model.

Comparisons

Three exemplary cases are presented in order to demonstrate the capabilities of the source model for three-dimensional directional characteristics according to the present invention and to compare it with exemplary measurements: a) spectra of airframe and engine noise on final approach, b) directivity at take-off, and c) Simulation of landing approaches with the landing gear extended.

The comparisons (a) and (b) can be made for the reference type A320_CFM56-5B (A320) on the basis of unevaluated Lwan of the source in order to show the influence of flight conditions on spectra and directivity. In (c) both the simulated sound pressure level and LAS of A320 and A333_TRENT7 (A330) are compared to show the strong effects of the landing gear.

Spectral comparison

A decisive property of a model for three-dimensional directional characteristics according to the present invention is the spectral resolution in third-octave bands. Furthermore, the source model and thus the spectra are separated into airframe and engine noise.Due to the dominance of engine noise during departures, spectra for the final approach of the A320 can be predicted in order to show the interaction of engine and airframe noise with the help of two different flight conditions: Idle power (N1 = 30%) at Ma = 0.26 compared to N1 = 50% at a final approach speed of Ma = 0.21. All flight parameters used for the forecast with the complete model are listed in Table 3 below. At the same time, the empirical data set, which is also used to adapt the model, can be filtered for the same flight conditions with a certain tolerance of the parameters. For each class of data, the arithmetic mean Lw can be determined to test how accurately the regression model can reproduce the input data.

Table 3: Parameter settings for the forecast of final approaches and tolerances of the measurement data.

[0189] θ [°] 130 ± 5 φ [°] 0 ± 90 N1 [%] 30.50 ± 5 Ma [-] 0.26,0.21 ± 0.02 p [kg / m <3>] 1.12, 1.15 - / - Flaps / Gears / SB 4/1/0 - / -

Directivity comparison

To compare the directivity with measured data, typical parameter settings can be selected for a take-off, since the directivity is most pronounced at high engine power. Analogous to the description in the section below, the regression data set can be filtered within tolerance ranges of the predicted mean values (see Table 4 below) in order to check the reproducibility of a model according to the present invention.

First, the directivity in the longitudinal direction of the complete model (3D, see equation (32) above) is shown for four exemplary frequencies. Second, the integrated Lw is shown over all frequencies for three power settings: a high power setting at 93%, a medium power setting at 90% and a low power setting close to power throttling (87%). In addition, the influence of the reduced model (2D, see equation (36) above) is shown.

Table 4: Parameter settings for the forecast of departures and tolerances of the measurement data.

Θ [°] 90, 130 ± 5 φ [°] 0.50 ± 5 N1 [%] 87.90.93 ± 1 Ma [-] 0.24 ± 0.02 p [kg / m 3] 1.15 ± 0.02 Flaps / Gears / SB 2/0/0 - / -

Simulation of actual approach situations

Finally, an exemplary landing approach situation for a receiver below the glideslope at a distance of 15 km from the landing threshold 34 of the exemplary airport, i. H. Zurich Airport, shown. In this situation the azimuth angle φ is below 5 ° and the aircraft flies at an altitude of about 600 m above the receiver. Similar situations can be selected for the aircraft type A320 as well as the aircraft type A330 in order to show the influence of the extended landing gear during the overflight.

The level-time protocol (LAS) of the full 3D model according to the present invention (aww via equation (33)) and the reduced 3D model (without configuration, see equation (35) above) are used with the measurement compared. In addition, the simulated flights are displayed as unweighted spectrograms with the change in sound pressure level in third-octave bands over time, just as a sophisticated level meter would measure them.

Results-spectral comparison

20 shows two diagrams which represent exemplary spectra of a landing approach of an aircraft of the type A320 with completely opened flaps and extended landing gear at θ = 130 °. Phase 1 (left) with idle power and Ma = 0.26 compared to phase 2 (right) with N1 = 55% and Ma = 0.21. Each total spectrum is divided into forecast contributions from the airframe (dash-dotted, blue) and engines (dashed, magenta). Thus, in Figure 20, two spectra of the aircraft type A320 on the final approach with the landing gear extended with the total level and their respective contributions from noise from the airframe (dash-dotted, blue) and engine (dashed, magenta) are shown. In the first phase (left) the engines are idling and the Mach number is 0.26. In this case, the model suggests that the full spectrum of airframe noise is dominated. Indeed, due to the development of the model based on measurements of the entire aircraft, it is inevitable that the airframe model still includes the idling of the engines. In the second phase (right) the aircraft has reached landing speed and the engines are running at 55% to maintain the glide slope. The engine model Ma = 0.21 thus predicts a spectrum that dominates slightly below 160 Hz due to the increased nozzle broadband noise and above 2 kHz due to the increased fan broadband noise. The airframe spectrum still dominates the medium frequencies, with the level being reduced by 4 dB compared to phase 1 due to the lower Mach number.

The empirical mean values show good agreement with both phases. Deviations at frequencies below 100 Hz and above 3 kHz can be explained by a higher uncertainty of the measurements due to the back propagation with high air absorption. In particular for frequencies between 125 Hz and 3 kHz, the interval that is decisive for A-weighted levels at the receiver, the deviations are, however, below 1.1 dB.

Results-directivity comparison

21 shows two diagrams which illustrate exemplary spectral directional characteristics for a departure with a low power setting (N1 = 87%). The directivity in the longitudinal direction against the polar angle θ (left) and for φ = 130 ° and the corresponding directivity in the transverse direction against the azimuth angle φ (right) are shown. During a typical take-off, the total air weight is primarily a result of the engine noise. As expected, the 125 Hz band is clearly pronounced in the rear area and has a peak around 140 °, which corresponds to the typical behavior of nozzle noise as shown in FIG. At higher frequency bands, the maximum level shifts to 90 ° and changes its shape. In particular for 2 kHz the shape changes completely and reaches the peak at 40 °. It is believed that it is fan noise that dominates the front area. The complete model reproduces the measured data exactly with deviations of mostly below 2 dB. These deviations are within the standard deviation of the measurement, which varies between 1.6 dB and 3 dB depending on the frequency and θ.

In FIG. 21, the directivity in the transverse direction for θ = 130 ° is shown on the right-hand side. The directional effect in the transverse direction is more pronounced at frequencies above 125 Hz, since only frequencies with short wavelengths are reflected on the wing surfaces. Maximum levels for higher frequency bands that exceed the level at 0 ° by about 2 dB occur at azimuth angles of about 30 ° to 60 °. The model predicts the measured values precisely with deviations for φ ≤ 60 ° below 2 dB, while Lw for φ> 60 ° is overestimated. However, these angles are less important for noise calculations, as they only contribute in distant lateral areas.

22 shows two diagrams which illustrate an exemplary directivity in the longitudinal direction of the overall Lw for three different take-off power settings of the aircraft type A320: below the aircraft at φ = 0 ° (left) and to the side at φ = 50 ° (right ). In Fig. 22, integrated, unweighted Lw for φ = 0 ° (left) and φ = 50 ° (right) are shown. For relevant polar angles from 60 ° to 120 °, a 3D model according to the present invention predicts the measured data well with deviations of less than 1 dB. For polar angles outside this range, the deviations are for both models, especially for φ = 0 °. but bigger. The 2D model shows slightly larger deviations on the left side, which become considerably larger for φ = 50 ° on the right side.

23 shows two diagrams which illustrate an exemplary directivity in the transverse direction of the overall Lw for three different take-off power settings of the A320 aircraft type: perpendicular at θ = 90 ° (left) and rearward θ = 130 ° (right). Results similar to those described above with reference to FIG. 22 can be seen for the transverse directional effect shown in FIG. 23 for θ = 90 ° (left). It turns out that the levels predicted with a 2D model would systematically overestimate the levels directly below the trajectory and underestimate the levels for lateral positions with φ between 20 ° and 60 °. In contrast to this, for θ = 130 ° in FIG. 23 (right) the deviations of the 2D and 3D models can be neglected at least for φ <60 °. Every 3% step of N1 leads to an increase in the total Lwum of approximately 1 dB for θ = 90 ° and approximately 2 dB for 0 = 130 °, which is well reproduced by both models.

Simulation of actual approach situations

24 shows six diagrams which exemplify two landing approaches in the idle setting and with the landing gear extended (dash-dotted vertical lines) and high-lift elements on a receiver about 15 km before the landing threshold. Flight parameters of the flight recorder data and the forecast and measured LAS are shown for the aircraft types A320 (left) and A330 (right). Thus, two landing approaches and their flight parameters are shown in FIG. The A320 on the left approaches with idle power, flap position 2 and extends the landing gear with a brief application of the airbrakes at the marked point. The same procedure applies to the A330 (right), but without using the airbrakes. The level-time logs of the measurement and the 3D model essentially agree very well. However, this does not apply to the reduced model (3Dred), in which the configuration of the aircraft is not modeled. After the landing gear is extended, the level of the A320 is only slightly lower, but the level of the A330 is considerably lower from 30 s to 50 s and does not agree well with the measurement.

FIG. 25 shows two diagrams which show exemplary spectra of a landing approach of an aircraft of the type A320 idling about 15 km before the landing threshold. Predicted (left) versus measured (right) spectrogram in third-octave bands from 25 Hz to 5 kHz on a receiver. The model presented provides many details for a single overflight, comparable to a real measurement. FIG. 7 shows the simulated (complete model, left) and measured (right) spectrograms of the same event as in FIG. 24 left. The sound pressure level over time shows very good agreement, with low frequencies in the rear area (A) being overestimated. The air absorption of high frequency bands (B) and the ground effect (C) are clearly visible and correlate with the measurements. The extension of the landing gear at (D) shows a slight increase in the level for frequencies below 100 Hz in the simulation as well as the measurement. A specific cavity tone of the A320, which occurs at 800 Hz and falls to 400 Hz due to the Doppler shift, is reproduced well, although it blurs over two frequency bands during the first 15 s (E).

26 shows two diagrams which exemplarily show exemplary spectra of a landing approach of an aircraft of the type A330 idling about 15 km before the landing threshold. Predicted (left) versus measured (right) spectrogram in third-octave bands from 25 Hz to 5 kHz on a receiver. In the same way as in FIG. 26, FIG. 26 shows the landing approach of the A330 as shown on the right in FIG. Again, there is good agreement between simulation (left) and measurement (right) with a slight overestimation at (A). The propagation effects (B, C) correlate with the measurement. The effect of landing gear extended at (D) is striking across the frequency range and shows the importance of including configuration parameters in the airframe model.

As a result, all comparisons of the flight conditions, namely take-off, landing approach and final approach, agree very well with the measured data. In addition, it was found that reducing the model to a 2D directivity or neglecting the configuration, although adversely affecting the accuracy, nevertheless represent feasible options if this data is not available in order to create a model according to the present invention.

Influence of the atmospheric stratification on the sound propagation of single flights

Sound propagation through the atmosphere is influenced by local conditions of temperature, humidity and wind speed. In the case of aircraft noise, the prevailing weather effect on sound propagation is dissipation. Second-order effects include level fluctuations caused by turbulence and, in rare cases, in sound paths close to the ground, the creation of acoustic shadow zones and influences on barrier effects as a result of temperature and wind gradients with height gradients.

One element of a method according to the present invention is the separation between the source model and the propagation model. The source model is based on a semi-empirical approach and takes into account airframe and engine noise of different flight conditions. The sonX sound propagation model is adapted to the specific problems of aircraft noise calculation. In addition to the detailed propagation, calculations are also used to forecast individual flights in order to convert data measured at the receiver back to the source. The exact reproduction of the atmosphere is particularly important here. A differentiated dispersion model also requires a greater depth of detail and higher quality of the input data. Appropriate vertical atmospheric profile data can be obtained from various sources: From weather balloons or from measurements made by the aircraft itself. From simulation results of numerical weather forecast models such as COSMO for many European countries. Based on measurements from ground stations in combination with idealized profiles.

According to the present invention, weather data from the three different sources are presented and compared in order to measure their effect on the resulting sound attenuation with different source-receiver geometries. A method according to the present invention is described for deriving idealized profiles from the weather data from ground stations. Based on these results, the benefit of a detailed modeling of the atmosphere compared to the assumption of a homogeneous atmosphere is discussed.

Dispersion model

All the propagation calculations according to the present invention can be carried out by the sonX model. The calculation from a point source to a receiver is carried out in two steps. First, the direct sound propagation for a uniform atmosphere with averaged conditions is calculated according to equation (37) below. This takes into account the geometric deviation (Adiv) and atmospheric dissipation (Aatm, f) depending on the frequency f according to ISO 9613-1. The model also takes into account barrier and soil effects (Agr / bar, f) as well as damping of vegetation (Afol, f), which are negligible when considered according to the present invention. ABasic, f = Adiv + Aatm, f + Agr / bar, f + Afol, f (37)

Then additional weather effects such as atmospheric dissipation (Aatm, Meteo) due to local temperature and humidity conditions are calculated by equation (38) below. The attenuations are reported as level differences compared to the basic attenuations of equation (37). In addition, changes in screen effects and the formation of acoustic shadow zones due to temperature and wind gradients (Dmet) are calculated using a ray tracking algorithm. These effects are not relevant for the current analysis. AMeteo = (Aatm, Meteo- Aalm, Basic) + (Afol, Meteo-Afol, Basic) - Dmet (38)

Information about the vertical profile of the atmosphere can be provided as individual profiles (i.e. from forecast models or flight record data) or as idealized profiles (see Section 2.3). A classification scheme has been introduced for the latter (see Table 5 below). The classification is reduced to the three main classes unstable (U), neutral (N) and stable (S). Depending on the wind speed at a height of 10 m and the current radiation balance, a corresponding class can be determined for each specific atmospheric condition.

Table 5: Classification scheme for different weather conditions.

0-1 0.0 U0 N0 S0 1-2 1.5 U1 N1 S1 2-3 2.5 U2 N2 S2 3-5 4.0 U3 N3 S3> 5 6.0 U4 N4 S4

Radiation balance and development of idealized profiles

The measurement of the radiation balance is possible, but not available in every case, since several sensors are necessary. In Switzerland, for example, only the short-wave incoming radiation that is measured with a pyranometer is usually available. The determination of the radiation balance follows VDI Standard 3789 Part 2. With regard to a horizontal area, the radiation balance is the sum of the solar short-wave radiation and terrestrial long-wave radiation, see equation (39). The solar shortwave radiation is the difference between the global radiation (G) and its reflection (R), which depends on the shortwave albedo of the earth's surface. The emitted heat radiation (E) of the earth can be simplified for natural soil surfaces as a black body that radiates energy with the fourth power of the surface temperature. Atmospheric gases and clouds reflect back to the earth (A). Q = (G - R) + (A - E) (39)

27 shows two diagrams which represent an exemplary radiation balance for an exemplary day in September 2013 at Zurich Airport (above). In combination with the wind speed, the weather categories reflect the different conditions during the time of day (below). The radiation budget can be calculated for each day of the acoustic measurements in order to collect source data (Fig. 27, top). Weather data at station height (KLO station at Zurich Airport, 426 m above sea level) were used in ten-minute resolution as input for the radiation balance. The cloudiness N in eighths is a visually recorded parameter with a resolution of one hour, therefore it was interpolated with a piecewise cubic Hermitian interpolation polynomial in order to ensure a smooth curve for the counter radiation. Using the radiation balance and the wind speed, the classification scheme from Table 5 above was implemented as in FIG. 27 (bottom).

For each class in Table 5, predefined Lin-Log profiles for temperature, wind and humidity were calculated, which reflect idealized profiles for the weather category and different soil types. For use under specific conditions, the profiles have been shifted in the direction of the current temperature and humidity on the ground. The weather data used herein had reference heights of 2 meters for temperature and humidity and 6 meters for wind speed. The profiles show considerable fluctuations near the ground. At higher altitudes (more than 100 m), however, they show a uniform behavior with constant wind speed and direction and an adiabatic temperature gradient of 9.8 K / km for unsaturated air and 6.5 K / km for saturated air. The absolute humidity is assumed to be unchanged, and thus the relative humidity increases with altitude up to a maximum value of 100%.

Profile data

28 shows two diagrams showing exemplary temperature, wind and humidity profiles for an exemplary day in September 2013 at 9:00 a.m. at Zurich Airport. The example shows idealized profiles (blue solid line) in high divergence to profiles from the numerical model COSMO-2 (magenta-colored dashed line) and flight recorder data (red dash-dotted line). The dashed black line stands for a homogeneous atmosphere with values measured at station height. Exemplary studies on which the present invention is based include flight data recordings (flight data recorders) of 223 departures at Zurich Airport, which were provided by Swiss International Airlines. From this data, air temperature and wind speed and direction were processed in order to generate vertical profiles as in FIG. 28. In a method according to the present invention, as stated above, the flight data recorder is further processed in combination with the acoustic measurements in order to develop the acoustic emission model as a function of the flight configuration.

All the data considered in this paper come from flights measured within four weeks between August 21 and September 13, 2013. Most A320 Family departures took place between 9:00 a.m. and 4:00 p.m. Although variations in atmospheric conditions are thus limited, the temperature still varied between 12 ° C and 28 ° C and the humidity varied between 20% and 80%. In addition, ten flights by A340-300 aircraft and three other types were measured at a night time around 11 p.m. when the temperature was still between about 17 ° C and 20 ° C and the humidity was about 65%.

As an alternative, profile data from the numerical weather forecast model COSMO-2 were used as input data for the dispersion calculation. COSMO is the “Consortium for Small-Scale Modeling” of the national weather services in Germany, Greece, Italy, Poland, Romania, Russia and Switzerland. MeteoSwiss, which provided the data, uses the local scaling model COSMO-2 with a grid spacing of 2.2 km, which also includes the Alpine arc (17). Indeed, the numerical model integrates atmospheric observation data from radiosondes, aircraft, wind profiler and surface level data. The hourly profiles provided include temperature, humidity, and wind speed and direction for surface levels 24 to 60, which correspond to heights above ground of approximately 10 m to 4,900 m.

In FIG. 28, the four profile types are compared with one another by way of example. For a comparison with a homogeneous atmosphere on the basis of current conditions, weather data at Zurich Airport were used at a reference height of 2 m above the height. Figure 2 shows an unstable situation at 10:00 a.m. for which all data sources show a similar pattern over the entire altitude range up to 800 m.

29 shows three diagrams showing exemplary temperature, wind and humidity profiles for an exemplary day in September 2013 at 10:00 a.m. at Zurich Airport. The example shows idealized profiles (blue solid line) in good agreement with profiles from the numerical model COSMO-2 (magenta-colored dashed line) and flight recorder data (red dash-dotted line). The dashed black line stands for a homogeneous atmosphere with values measured at station height. In contrast to FIG. 28, FIG. 29 depicts a situation in which the various profiles differ considerably from one another. In the case of the COSMO-2 and flight recorder profiles, the temperature decreases with increasing altitude, which indicates an unstable stratification. In contrast, the idealized profile already assumes a neutral state. In the same way, the moisture profiles also differ greatly.

30 shows two diagrams showing exemplary selected temperature and humidity profiles for an exemplary day in September 2013 between 9:00 a.m. and 12:00 noon at Zurich Airport. The numerical model COSMO-2 (magenta dashed line) suggests that the stratification from a stable boundary layer at night to a typically unstable layer on a sunny day continues until midday. The classification scheme already assumes an unstable stratification (blue solid line). The idealized profile could not always correctly reproduce the temporal transition from a stable boundary layer at night to a typically unstable layer on a sunny day. In FIG. 4, the vertical profiles of the idealized profiles and COSMO-2 data are compared for three different times of the day between 9.30 a.m. and 12 noon. On this sunny day (see Fig. 27) the sun rises at 7:30 a.m. and the radiation balance leads to an unstable stratification two hours later if the classification scheme is applied. In contrast, the data from COSMO-2 suggest that the change in stratification continues up to 500 m until noon and that the temperature and humidity profiles only slowly converge.

In summary, the wind and temperature profiles of the flight data recorder and COSMO-2 are basically very consistent. The extrapolation of temperature from soil conditions to higher altitudes, as done by the idealized profiles, appears to be valid in most cases. However, wind and moisture profiles differ considerably compared to the differentiated profiles from the COSMO-2 model.

Propagation calculation

31 shows a schematic illustration of a calculation scenario based on a flight path. Although the flight path of the 223 used departures is available, the same generic source points of a virtual flight path can be used for all flights in order to avoid differences in propagation due to different flight path geometries. The scenario depicted in FIG. 31 shows three different source positions (S), one at 500 m above receiver R1 and the others at 45 ° and 30 ° angles in relation to the direction of flight. A second receiver R2 was set up 500 m to the side. The receivers are placed 4 m above a level grassland area.

The results are presented below as attenuations AMeteo (see equation (38) above) of the dispersion calculation in order to show the differences between a homogeneous atmosphere based on current local conditions (LC) and averaged conditions for Switzerland. Furthermore, the results of the COSMO-2 data are compared with those of the homogeneous atmosphere based on current conditions (see equation (40) below). In addition, the results of the idealized standard profiles (IP) are compared with those of the COSMO-2 profiles (see equation (41) below). The differences in attenuation in each of the third octave bands up to 5 kHz are detailed below. In particular, the influence on the A-weighted sound exposure level LAE is discussed.

Since the flight recorder data do not provide any information on moisture data, this information was taken from COSMO-2 for the sound propagation calculation. However, it turned out that the results are very similar to the calculations of the COSMO-2 profiles and are therefore not shown and discussed in more detail. ΔALC-COSMO = AMeteo, LC- AMeteo, COSMO (40) ΔAIP-COSMO = AMeteo, IP- AMeteo, COSMO (41)

Results

Fig. 32 is a diagram showing an exemplary fluctuation of air absorption of a homogeneous atmosphere with a shift from current conditions at 2 m to a uniform atmosphere with averaged conditions.

Results of the propagation from S1 to R1 at a distance of 500 m

Distributions of the calculated differences in air absorption are presented for the 223 exemplary departures or their associated weather situations. For each third octave band, the differences are shown as box whisker plots (see legend in FIG. 32). Because it provides the most reliable data, COSMO-2 is used as a reference for comparing results.

Differences between homogeneous atmospheres

In a first step, the weakening of a homogeneous atmosphere based on the current conditions on the ground is compared with the weakening of a homogeneous atmosphere with averaged values of 8 ° C. and 76% for Switzerland. The results for the point of closest approximation (CPA) in FIG. 32 show only slight fluctuations of less than 0.6 dB below 500 Hz. Between 500 Hz and 1.6 kHz there were mean values of approximately +0.5 dB with maximum values of +1.9 dB and positive minimum values (except at 1.6 kHz). Thus, the current conditions always led to higher attenuations than the mean atmosphere. At high frequencies the trend changed towards negative differences, but showed significantly higher fluctuations in both directions. The mean value of the 5 kHz band is -4.3 dB with a minimum value of -10.3 dB and a maximum value of +15.0 dB.

33 is a diagram showing an exemplary fluctuation in air absorption of a homogeneous atmosphere with a shift from current conditions at 2 m to a uniform atmosphere with averaged conditions. Results of the propagation from S3 to R2 at a distance of 1,118 m. With this, the attenuation spectra for the greatest propagation distance of 1,118 m can be derived from Fig. 33. The same trend as for the CPA can be observed with a turn to negative values at 2 kHz. The mean values for 500 Hz to 1.6 kHz ranged from +0.5 dB to +1.0 dB with maximum values of +2.0 dB to +4.3 dB and also positive minimum values. The high frequencies showed a higher negative mean dissipation for an atmosphere based on current conditions, again with strong deviations between -23 dB and +34 dB.

Differences in profile data sources

34 shows two diagrams which illustrate, by way of example, differences in air absorption between a homogeneous atmosphere under current conditions and COSMO-2 profiles. Results of the propagation from S1 to R1 (500 m) and S3 to R2 (1,118 m). In accordance with the present exemplary embodiments of the present invention, results are only presented for 250 Hz to 5 kHz, since the fluctuations from low frequencies are negligible. 34 (left) shows the fluctuations in air absorption between the homogeneous atmosphere at current conditions and the COSMO-2 profiles for S1 to R1. Below 2 kHz the mean values were slightly positive. In contrast, above 2 kHz the averaged differences increased up to -3.6 dB for 5 kHz with strong fluctuations from -17.4 dB to +5.7 dB. Fig. 34 (right) compares the propagation from S3 to R2. The trend is similar, but shows greater fluctuations at medium and high frequencies as a result of the greater distance from which it is propagated.

35 shows two diagrams which illustrate, by way of example, differences in air absorption between idealized profiles and COSMO-2 profiles. Results of the propagation from S1 to R1 (500 m) and S3 to R2 (1,118 m). The variance of the idealized profiles compared to the COSMO-2 profiles is presented in FIG. For both spectra there were no fluctuations of more than 0.5 dB below 1 kHz. At higher frequencies, the mean values for Figure 35 (left) decreased to 4.2 dB at 5 kHz with a fluctuation between -17.3 dB and 2.3 dB. In Figure 35 (right) the mean values also decreased to 3.4 dB, but fluctuated from -15.1 dB to 8.1 dB.

Influence on the resulting sound exposure level

The results discussed above clearly show considerable fluctuations in air absorption for frequencies above 250 Hz. The question thus arises of how relevant these fluctuations are for the resulting LAE. Sound emission directivities from 35 Airbus A320s can therefore be averaged, and the resulting A-weighted spectra and the total LAE at receiver R1 were calculated (FIG. 10). Then the differences from LAE for the mean difference and the minimum and maximum fluctuations can be determined as described above.

Fluctuations in the frequencies above 2.5 kHz have almost no influence on LAE (<0.1 dB (A)), in particular due to the high absolute atmospheric absorption. The same applies to frequencies below 125 Hz, which are greatly attenuated by the A-weighting. The influence between a homogeneous atmosphere under local and averaged conditions on LAE for 500 m (S1R1) is -0.2 dB (A) as an average, but fluctuates between -1.4 dB (A) and 0.3 dB (A) for single flights. At 1,000 m (S3R1) the mean fluctuation of the LAE is -0.6 dB (A) with a range of -2.6 dB (A) to 0.2 dB (A).

In the same way, the influence of the differences in air absorption between a homogeneous atmosphere at current conditions and COSMO-2 profiles is -0.6 dB (A) to 1.3 dB (A) at 500 m and -0.9 dB (A) to 0.4 dB (A) at 1,000 m. The differences in the calculated air absorption of idealized and COSMO-2 profiles change the LAE from -0.2 dB (A) to 1.5 dB (A) 500 m and from -0.5 dB (A) to 0.7 dB (A) at 1,000 m.

As FIG. 35 shows, the differences between idealized and COSMO-2 profiles at frequencies above 1 kHz are in the same size range as the variance between the homogeneous atmosphere under current conditions and COSMO-2. However, at medium frequencies below 1 kHz, which have a strong influence on the LAE, the fluctuations are small.

36 shows a diagram which shows an example of an influence on the variance of the dissipation in the A-weighted spectrum at the receiver R1 of the A320 for the occurrence of S1 (90 °) and from the front at S3 (30 °). The indicators represent the minimum and maximum fluctuations between the homogeneous atmospheres under current or averaged conditions.

As a result, a considerable dependence of the sound level at higher frequencies on varying conditions of temperature and humidity can be seen. For the frequency bands that significantly influence the resulting A-weighted level, the differences are smaller, but still have a range of several decibels. When looking at the mean values, it becomes apparent that the deviations between the various data sets show a clear trend and thus also have a systematic influence on the long-term averages.

For the comparison of different sources of weather data, forecast models such as COSMO-2 appear to be the most reliable. While black box data shows good agreement with COSMO-2 for wind and temperature, it does not provide humidity data. In addition to the fact that flight data recorders cannot be used as the only data source, the availability of flight recorder data is usually limited and the accuracy of the input data also depends on the type of aircraft. Thus, black box data alone is not an adequate source of weather data.

The use of standardized profiles of the atmosphere normalized to conditions on the ground, as shown above, has a beneficial effect on the accuracy of the sound propagation calculation compared to the assumptions of a homogeneous atmosphere. The advantage of this approach is that the profiles can be generated with very little, easily accessible input data.

A variance in air absorption can be expected to be of the same order of magnitude as the results presented here. In contrast, the fluctuations between the homogeneous atmosphere at current conditions and idealized profiles compared to the COSMO-2 profiles are not expected to change significantly, and the following conclusions can also be considered generally applicable.

The use of current ground conditions of temperature and humidity is an important improvement over a uniform atmosphere with averaged conditions for the calculation of individual flights. In addition, a correct reproduction of the stratification in the dispersion calculation also has a beneficial effect on the accuracy of the resulting LAE on. in particular, the use of data from weather forecast models such as COSMO-2 is believed to be the most accurate solution. Particularly for the development of an emissions model according to the present invention, in which individual flights are reversely converted to the source, it is recommended to use such detailed input data. Although high frequencies above 2.5 kHz are negligible for the LAE above 500 m, the strong fluctuation in air absorption could lead to large errors in this band.

It can be assumed that idealized profiles cannot replace the more detailed profiles from COSMO-2. The assumptions, especially for moisture, but also for the temperature above 100 m, often lead to an extrapolation of the soil conditions, which is not valid at higher altitudes. Thus, the differences in air absorption compared to the COSMO-2 profiles, which are significant at frequencies above 1 kHz, change the LAE and lead to a less accurate forecast for single flights.

Models for reflections on forests, cliffs, buildings, walls and other rigid surfaces

A method according to the present invention can also include models for reflections on buildings, walls, and other rigid surfaces, as well as diffuse reflections on forest edges and cliffs. For a detailed description and references to such extensions of a propagation model on which a method according to the present invention is based, reference is made to the “Documentation of the sonX model” of September 12, 2016 and to scientific publications published in ACTA ACUSTICA UNITED WITH ACUSTICA, especially “An Extended Model to Predict Reflections from Forests” (Jean Marc Wunderli; Vol. 98 (2012) 263-278; DO) 10.3813 / AAA.918510); “A Model to Predict Sound Reflections from Cliffs” (Reto Pieren, Jean Marc Wunderli; Vol. 97 (2011) 243-253; DOI 10.3813 / AAA.918404); “Calculation of Reflections in an Urban Environment” (Kurt Heutschi; Vol. 95 (2009) 644 - 652; DOI 10.3813 / AAA.918193); and "An Engineering Model for Sound Pressure in Shadow Zones Based on Numerical Simulations" (Jan Hofmann, Kurt Heutschi; Vol. 91 (2005) 661 - 670), which are included herein by reference , referenced.

Subspace concept

37 shows a schematic perspective view of a vehicle 111, for example an aircraft, in particular an aircraft, for which noise emissions and immissions are to be calculated in accordance with a method according to the present invention. An air space through which the vehicle 111 passes is divided into subspaces 112 which form cells. Each of these sub-spaces 112 can have a cubic shape so that it provides eight corners 113.

Each of these subspaces 113 represents a potential source location. From each corner 113 of one of the subspaces 113 through which the vehicle 111 passes to each receiver point, an attenuation is stored in a database. During individual flights, attenuations from corners 113 are looked up in the database. The resulting attenuation is expressed as the linear interpolation of the attenuations of the corners 113 versus the effective source position, i.e. H. the position of the vehicle 111.

37 shows a schematic perspective view of an air space which is divided into subspaces 112 in accordance with a method according to the present invention. Source locations 114 of a sound source such as a vehicle 111 form an expected trajectory 115 or a defined trajectory 116. The air space and thus the subspaces 112 extend along a longitudinal direction X, a transverse direction Y and a height direction Z, which together form a Cartesian coordinate system. In a projection along the height direction Z, each of the subspaces 112 preferably has a square shape. In particular, vertical edges of the subspaces that extend parallel to the height direction Z can, however, have different and / or changing lengths and coordinates along the height direction Z.

By dividing the air space into the subspaces 112, considerably fewer attenuations and thus corresponding calculations have to be made than in this simulation in full format. This enables an efficient calculation of different traffic scenarios in a method according to the present invention. Maximum level calculations and simulations of individual flights as well as real-time simulations of noise pollution are possible. This enables precise calculations with a method according to the present invention, which integrates spectral noise propagation models, considerations of weather influences, coherent and incoherent reflections on buildings, shielding effects through buildings and / or considerations of ground reflection / noise propagation influences depending on earth types as described above. Furthermore, a precise emission model is made possible, which can be integrated into a method according to the present invention for the simulation of low-noise landing approach and departure procedures of vehicles to and from the airport or the like.

Computing facilities, systems and processes

According to at least one aspect of a method and system according to the present invention, interfaces for noise calculations on computing devices that work in parallel with one another can be provided. Central data storage for easier administration can be implemented. Flexibly configurable interfaces for data import can be made available. Methods and systems according to the present invention can provide functionalities for data preparation and homogenization. Furthermore, modules for evaluation and reporting for generating reports and standard file formats with regard to the noise exposure of people, buildings, apartments, workplaces, areas, etc. can be provided.

Figure 39 is a schematic diagram illustrating a system 117 for performing a method in accordance with the present invention. System 117 comprises an external module 118 and an internal module 119. External data such as cockpit data, radar data, transponder data, emission data and evaluation data as well as statistics can be provided and processed in the external module 118. The statistics include, in particular, certain numbers of flight movements for certain types of aircraft and routes. On the basis of such statistics, paths are superimposed in order to generate the sound profiles and to derive and / or calculate total sound immission values for the at least one reception point R or generally a number of reception points arranged along the paths.

In the internal module 119, calculations are carried out according to models integrated in a method according to the present invention, which generate flight events and geometries, source locations, receiving locations and sound propagation based thereon, as well as attenuation data derived therefrom, and a simulation of individual flights and the generation of Sound profiles that include a bundle of estimated and / or defined paths and finally noise exposure maps. The internal module 119 comprises a central computer and / or computer network 120 and at least one client device 121. The most demanding arithmetic operations for simulating models according to the present invention are carried out in the computer network 120 as described above. For this purpose, the computer network 120 preferably comprises a plurality of digital processors for performing the calculations, which central computers, servers or the like can be implemented. The client device 121 can be a stand-alone computer or the like on which less demanding calculations and operations are carried out. The system 117 according to the present invention is adapted to carry out a method according to the present invention, which comprises the following steps: On the side of the external module 118, for example by external data providers and providers of social services, in a first step S1 location data such as Cockpit data, radar data, transponder data and other related data are provided. In a second step S2, the location data are processed in order to be input into the internal module 119.

In the internal module 119, in particular in the computer network 120, flight events including geometries are processed in a third step S3. Based on the flight events, sound source positions are calculated in a fourth step S4. In parallel with or following the third step S3, reception points are calculated in a fourth step S4. The source positions and reception points then leave the computer network 120 in order to be displayed on the client device 121. In the client device 121, initial calculations of the actual sound propagation can be carried out in a sixth step S6 in such a way that they can be adapted and alternated by a user of the client device 121. The propagation calculations are then fed back from the client device 121 to the computer network 120.

In the computer network 120, the attenuation is calculated in a seventh step S7. The attenuation data are returned to the client device 121, where single flight simulations are carried out in an eighth step S8. Results of at least one single flight simulation, preferably a plurality of single flight simulations, are fed back to the computer network in order to generate sound profiles with clean simulation data that relate to several orbits, taking into account emission data that are provided by the external module 118 in a ninth step S9, to generate.

Based on the generation of sound profiles in a tenth step S10, which can be carried out by the computer network 120 and / or the client device 121, in combination with movement statistics of a population that is obtained from the external module 118 in an eleventh step S11 noise exposure maps are created in a twelfth step S12. Based on the noise pollution maps, a discussion of railways, in particular flight routes, can be objectified with the help of a method according to the present invention, so that all parties involved, such as operators, airlines, the public and politicians, can implement feasible and acceptable solutions for the planning and control of traffic, in particular Air transport, can find.

REFERENCE MARK

1 Location microphone one 2 Location microphone two 3 Location microphone three 4 Location microphone four 5 Location microphone five 6 Location microphone six 7 Location microphone seven 8 Location microphone eight 9 Location microphone nine 10 Location microphone ten 16 First runway (departures) 28 First runway (landings) 34 Second runway 111 Vehicle 112 Subspace 113 Subspace corner 114 Location of the source 115 Expected trajectory 116 Defined trajectory 117 System 118 External module 119 Internal module 120 Computer network / central computer 121 Client device S Source R Receiver X Longitudinal direction J Cross direction Z Elevation direction θ polar angle φ azimuth angle ρ medium density

Claims (7)

1. Computer-implemented method for calculating a noise level that is generated by an aircraft as a sound source (S, 111), in particular an aircraft that moves along a defined trajectory (116) based on the at least one receiving point (R), for at least a receiving point (R), the method comprising the following steps: Providing a sound propagation model for an air space which extends between the at least one receiving point and at least one expected trajectory (115) along which the sound source is expected to move relative to the receiving point; Calculate the noise level through Inputting the defined trajectory (116) into the sound propagation model; and Calculating a sound propagation between the sound source (S, 111) and the at least one receiving point (R) for each of a number of different sound source locations (114) which are arranged along the defined trajectory, wherein the sound propagation is calculated based on at least one sound emission value of the sound source (S, 111) and the at least one sound emission value depending on a sound radiation pattern that corresponds to a type of sound source (S, 111) for each of a number of different sound source locations (114), the are arranged along the defined trajectory (116), is assigned, is calculated. 2. The computer-implemented method of claim 1, wherein the sound radiation pattern is based on a sound power level associated with the type of sound source (S, 111). 3. The computer-implemented method according to claim 1, wherein the sound propagation model comprises a direct sound propagation scenario and a complex propagation scenario; wherein in the direct sound propagation scenario a direct sound propagation between the sound source (S, 111) and the at least one receiving point (R) is assumed; and wherein in the complex propagation scenario a complex sound propagation between the sound source (S, 111) and the at least one receiving point (R) is assumed. 4. The computer-implemented method of claim 3, wherein for each of the number of different sound source locations (114), a line-of-sight angle between the sound source (S, 111) and a horizon is determined; wherein the direct sound propagation model is applied to line-of-sight angles exceeding an insignificance threshold; and where it is assumed that at line-of-sight angles exceeding the insignificance threshold, complex sound propagation becomes negligible. 5. The computer-implemented method of any one of claims 1 to 4, further comprising the following steps Dividing the air space into adjacent sub-spaces (112); Calculating a partial space model for each of the partial spaces (112) for determining a finite sound propagation within each of the partial spaces (112); Composition of the sound propagation model from the sub-room models, at least one boundary area between adjacent sub-rooms (112) reproducing at least one of a virtual sound source and a virtual reception point for a virtual sound transmission value which reproduces a virtual sound power level transmitted between at least two adjacent sub-rooms (112) . 6. Computer readable medium comprising computer readable instructions that appear at the execution by a computer cause the latter to carry out a method according to any one of claims 1 to 5. 7. Computer system (117) comprising means with instructions for carrying out the method according to one of claims 1 to 6.