EP3486897A1 - Work vehicle with noise reduction in a vehicle cabin - Google Patents

Abstract

The invention relates to a work vehicle having a driver's cab (10) which has a closed cabin interior which is enclosed by structurally stable cabin wall elements and at least one pane which closes a cabin wall opening released from the cabin wall elements, with an actuator device acting on the pane, a sensor device and a sensor connected to the sensor device and the actuator control unit, which is adapted to control in response to signals of the sensor device, the actuator device for noise reduction in the cabin interior, characterized in that the actuator is mounted on the inner surface of the disc inertial oscillator (4), the sensor device is a microphone (6) directed towards the cabin interior and the control unit is set up to control the inertial vibration exciter (4) with an adaptive and predictive control as a function of the M to control the microphone signals to perform an active noise reduction in the cabin interior.

Description

The present invention relates to a work vehicle with a driver's cab, which has a closed cabin interior, which is surrounded by structurally stable cabin wall elements and at least one disc which closes a cabin wall opening released from the cabin wall elements, with an actuator device acting on the pane, with a sensor device and with a the sensor device and the actuator device connected control unit, which is adapted to control in response to signals of the sensor device, the actuator device for noise reduction in the cabin interior.

The invention relates to work vehicles, including herein e.g. Forklift, excavator, combine harvester, bulldozer, rollers for road construction, tractors, etc. are understood. In a work vehicle, a closed driver's cab protects the driver from environmental influences such as noise, climate, rain, falling objects, etc .; In particular, because of the protection against noise in the working environment, the driver's cab is largely hermetically sealed.

With regard to noise in driver's cabs of work vehicles, attempts have also been made to take measures for active noise reduction by counter-sound, namely at low frequencies below 500 Hz (for higher frequencies sound-insulating materials on the interior walls of the cab are effective). Examples of active noise reduction are those in WO 94/29845 A and WO 94/29846 A described cabins in which a speaker is mounted on the ceiling of the cab. Furthermore, two microphones are mounted in the cab, detect the noise in the interior of the cab. In a control unit processes a filter algorithm the microphone signals in order to generate a control signal for the loudspeaker, so that the loudspeaker generates counter-sound which extinguishes or reduces the noise in the area of the microphones as far as possible. In this context, adaptive control algorithms are mentioned, which can adapt to a certain extent to changing noise situations. However, the method described is based on measuring the noise signal closer to the source before it arrives at the location where the noise is to be reduced, and in this way outputting the canceling or noise-reducing counter-sound from the loudspeaker, and then if it arrives in the desired noise reduction range, it causes a partially destructive interference or cancellation with the noise coming from the source and then arriving in the area of interest. This corresponds to a feedforward control, which is not stable and effective in all scenarios of time-varying noise situations. Moreover, the generation of counter-noise by a loudspeaker is not particularly effective for the described application in driver's cabs of work vehicles.

Out EP 2 101 316 A1 is known an agricultural work machine with noise-damped cabin. In this case, an external noise source outside the driver's cab is present, from which an oscillation is transmitted on a first path into the driver's cab. In order to reduce noise in the driver's cab, a compensator element is to be driven with a phase and amplitude adapted to the phase and amplitude of the transmitted vibration in order to emit a sound vibration which compensates for the vibration transmitted to the driver's cab. The noise of the external noise source is already measured at the source with a sensor, and it is assumed that the propagation of the sound from the external noise source on the first path is slower than the signal transmission from the sensor to the external noise source via the control device and to said compensator element, so that the desired extinction or reduction in amplitude and phase can be properly effected at the desired location within the driver's cab. This is also a feedforward control. Such open-type controllers do not work when there are jamming paths over which the noise propagates faster than the signal transmission on the second path to the compensator element.

The described system is particularly directed to the suppression of stationary tonal noise components that excite the cabin cabin resonances, such as the periodic noises generated by the rotating blades of a combine harvester. For noise components outside the resonance frequencies or in damped cabin modes, the described procedure does not result in the reduction of noise in the head area of the driver, since effective cancellation of the overlapping sound components is not possible due to the transit time difference from the loudspeaker to the head area and to the microphone. In principle, with the proposed sensor-actuator arrangement no noise can be suppressed in which the change duration of frequency and / or the amplitude of the noise is shorter than the transit time between the sensor and the loudspeaker.

In the cited document EP 2 101 316 A1 It is also mentioned that a stimulable resonator, which is attenuated more strongly than the window pane, can be coupled to a window pane of the driver's cab. In this way, the energy of the vibration of the window pane should be distributed to the window pane and the resonator. The addressed excitable resonator is therefore only a passive vibration absorber.

Out US 5,812,684 B A device for reducing noise in a driver's cab of a motor vehicle is also known, which does not expressly refer to the driver's cab of a work vehicle, but otherwise has the features of the preamble of claim 1. At a lower edge region of a side window planar piezoceramic actuators are attached. By deliberately applying tension to the piezoceramic actuators, these can be purposefully expanded and contracted, whereby the disk on which the planar piezoceramic actuators is attached can be deformed. On the disc, a piezoceramic sensor is further attached, detected with the vibrations of the disc and sent to a control unit. The control unit is configured to generate control signals for the piezoceramic actuators on the basis of the vibrations of the window pane detected by the sensor in such a way that the vibrations of the window pane detected by the sensor are minimized. In this way, the sound level emitted by the oscillating window pane into the interior of the driver's cab should be reduced by suppressing the vibrations of the windowpane as much as possible. Although generation of counter sound for active noise reduction is also mentioned as basically known in this document, it should be replaced by a simpler procedure (vibration suppression of the window panes).

As mentioned above, the cabs of work vehicles are largely hermetically sealed because of their various protective functions. In work vehicles, the driver must be able to observe his surroundings very well in order to be able to detect accident risks. Therefore, such cabs have next to structurally fixed cabin wall panels one or more larger, released by the cabin wall elements openings which are closed by slices. The discs are made in the Usually made of transparent glass or plastic. Most are therefore provided on all sides of the cab wall openings with windows as windows, so that the driver can perceive its environment forward, backward and on both sides. In terms of vibration, the closed air volume in the cabin interior together with the panes in the cabin wall openings forms a vibro-acoustic system (VAS), which is weakly damped at low frequencies and therefore susceptible to vibration. Suggestions of the VAS take place at low frequencies, in particular by an internal combustion engine of the work vehicle, by hydraulic units and by tire-roadway interactions, such as the impact of the distributed around the circumference of the tire lugs of the profile on the running surface. Therefore, high sound levels of more than 110 dB can occur in resonance ranges of the VAS at approx. 40 Hz, which are perceived by the driver as unpleasant hum or noise.

Another problem is that the vibrations transmitted to the driver's cabin can change rapidly according to frequency and amplitude composition as a function of time. For example, with an accelerating forklift, the engine firing frequency and engine speed increase. Significantly faster over these increases the tire-road contact frequency; if e.g. the tires distributed over their circumference each have 28 wheel profile cleats, the wheel contact frequency increases by a factor of 28 faster than the wheel rotation frequency. The resulting complex evolution of the frequency and amplitude composition of the driver cabin vibration components can not be counteracted with measures operative for stationary external sound and vibration sources, such as the feedforward control described above.

It is an object of the present invention to provide a work vehicle with a driver's cab with an actuator device and a to design this operating control unit for active noise reduction in the cab so that the above-described significant amplitudes of the vibrations of the vibratory acoustic system from the cabin air volume of the cab and the windows of the cab in the low-frequency range and the associated noise are effectively suppressed and that the control unit can adjust itself to the time-varying conditions of the frequency and amplitude composition of the vibration acting on the vibrating acoustic system during operation of the working vehicle.

To solve this problem, the working vehicle with the features of claim 1. Advantageous embodiments of the invention are set forth in the dependent claims.

According to the invention, it is initially provided that an inertial vibration exciter is attached to the disc as the actuator device, which vibrator can set the disc in vibration. Inertial-Schwingerreger (which are also referred to as shakers) are based in their operation on a movable in the housing of the vibrator along an axis inertial mass (actuator), which is driven to oscillate electromagnetically, wherein the oscillation frequency is adjustable by adjusting the electromagnetic drive frequency. The inertial vibration exciter is mounted on the disc in such a way that the axis of the oscillation movement is perpendicular to the disc plane in order to be able to couple mechanical vibrations into the disc effectively. In this way, the disc itself is the sound source, which generates in a manner to be described counter-noise for noise reduction in the interior of the cab.

As inertial mass of the inertial vibration exciter can be used for oscillation drivable actuator with an electromagnet, which is suspended on springs and on the two axially spaced ferromagnetic attraction elements act. An example of an inertial vibration exciter is in EP 3 217 053 A1 described.

Furthermore, according to the invention, the control unit is adapted to process the signals of the microphone directed to the interior of the driver's cab with an adaptive and predictive control algorithm, thereby generating a control signal for the inertial oscillator, around the disc to which the inertial vibrator is attached is to vibrate so that the disc generates counter-noise in the interior of the cab, which minimizes the error signal picked up by the microphone. The use of an adaptive and predictive control algorithm makes it possible to counteract the frequency and amplitude composition temporally variable noise in the interior of the cab during operation of the work vehicle so that the counter sound already arrives at the location of the microphone when the then prevailing noise situation by this counter sound is destructive to compensate.

It has been found that the coupling of the inertial vibration exciter to a window of the driver's cab effects a particularly effective and direct mechanical coupling to the vibroacoustic system comprising interior air volumes and windows of the driver's cab which provide particularly effective noise suppression in the interior the driver's cab allows. As a result, not only the resonance vibrations of the disc and the trapped air volume of the driver's cabin counteracted, but operated the disc itself as a counter-noise source.

The inertial oscillator should be able to be operated below 100 Hz, in particular in the low-frequency range, since in this Frequency range critical noise situations occur in cabs of work vehicles.

The attachment of the electrodynamic inertial vibration exciter on the disc realized with a compact actuator (Inertial-Schwingerreger) a structure that can be generated by the deliberately excited to vibrate disc sound with low frequencies and high sound levels in the interior of the cab, which is to extinguish of the noise in the area of the microphone.

Adaptive and predictive control or filtering algorithms are known in the art. Control units that are set up to implement an adaptive and predictive control algorithm are also known by the term adaptive internal model controller (IMC).

In a preferred embodiment, the inertial vibrator is mounted horizontally and vertically outside the center of the disk to the disk. Here, a horizontal direction is a direction parallel to the bottom surface on which the work vehicle stands, and a vertical direction is a direction perpendicular to this bottom surface. The advantage of placing the inertial vibrator eccentrically on the disk is that the vibration mode whose wavelength is equal to the length of the disk in the horizontal direction and the vibration mode whose wavelength is equal to the height of the disk in the vertical direction, exactly the center of the disc have nodes of vibration. An inertial oscillator mounted centrally on the disk could not effectively couple to these low-frequency vibration modes. In addition, an off-center inertial vibrator located off center of the disk interferes less with transmission through the disk.

According to a further preferred embodiment it is provided that the microphone is placed on or in the housing of the inertial vibration exciter. If the microphone is located inside the housing, the microphone should be in the area of a housing opening, which allows the microphone to detect the sound in the interior of the driver's cab. In this embodiment, it is further advantageous for the microphone to be aligned in such a way that a central surface normal on the membrane of the microphone is perpendicular to the oscillation axis of the inertial vibration exciter. This avoids that the vibrations of the inertial vibration exciter additionally vibrate the membrane of the microphone, which could falsify the microphone signal.

The arrangement of the microphone in the immediate vicinity of the inertial vibrator has the further advantage that thereby the acoustic transit time between microphone and inertial vibrator is very small, which makes the scheme more stable and robust against movements of the driver's head, because movements of the The driver's head would change the control path more when the microphone is placed closer to the driver's head.

Usually, it would make sense to first subject the microphone signal to low pass filtering because of the low frequency range of frequencies less than 100 Hz, which is of interest here alone, since only these frequencies must be included in the control. In a preferred embodiment it is provided that instead of a low-pass filtering of the analog microphone signal, the control unit is set up to digitize the microphone signal at a sampling frequency of at least 20 kHz for digitization. Due to the oversampling, the low-pass filtering is required only from 10 kHz. However, this is already done physically by the inertia of the microphone membrane.

Low-pass filtering is also not provided on the output side of the control unit because the inertia of the oscillating body in the inertial vibrator effectively acts as low-pass filtering.

In a preferred embodiment, the control unit is configured to adaptively perform the adaptive and predictive control algorithm only up to a predetermined cutoff frequency. This cut-off frequency can be adjusted depending on the distance between the microphone and the place where suppression by counter sound is supposed to be most effective (ears of the driver) and the cabin shape. By determining the cut-off frequency, it is possible to avoid producing counter-noise components which would lead to a reduction of the noise at the location of the microphone, but not at the actual operating location (ears of the driver).

So that after the sampling of the microphone signal at 20 kHz, the computational effort by the high sampling frequency does not lead to a significant increase in the computational time in the rest of the control algorithm, the control unit is provided after sampling at 20 kHz with a down sampler, which is adapted to the digitized Signal down-sampling to a maximum of 2 kHz. This down-sampling does not have the negative time-delay effect of low-pass filtering an analog signal, since this down-sampling is performed on the digitized signal, resulting in no significant delay.

In order to accelerate the adaptive and predictive control algorithm, it is further provided that the control unit is provided with an equalization filter in the adaptive and predictive control algorithm that is inverse to the minimum-phase component of the frequency response of the controlled system by the convergence time of the adaptive and predictive control algorithm frequency independent. In this way, the convergence speed of the adaptive and predictive control algorithm can be increased, since there are no frequency ranges with very high convergence times on the one hand and other frequency ranges with very low convergence time, which would mean that the longest convergence time would be considered decisive, but only one smaller mean convergence time, which then applies to all frequencies.

In a preferred embodiment, not only is the microphone attached to or integrated with the housing of the inertial vibrator, but also the control unit, with its electrical circuits and digital processor units as a control module, is mounted on or integrated into the housing of the inertial vibrator. In this embodiment, a working vehicle with active noise reduction in the cab can be realized by attaching the inertial vibrator with the associated microphone and the associated control unit to the window of the cab and then connected to the DC electrical power supply in the cab. This makes it possible to retrofit a work vehicle without active noise reduction by installing and connecting a single component in a very simple manner.

As far as in the present application of a "disc" is mentioned, this will usually be a transparent glass or plastic in practice. In principle, however, the invention can also be implemented if, in addition to openings with windows, an opening of the driver's cab is closed with a non-transparent pane on which the inertial vibration exciter could then be fastened in order to realize the active noise reduction in the driver's cab of the working vehicle.

• Fig. 1 zeigt die Motorzündfrequenz eines mit einem Verbrennungsmotor angetriebenen Gabelstaplers bei der Beschleunigung als Funktion der Zeit,
• Fig. 2 zeigt die Radstollenfrequenz des beschleunigenden Gabelstaplers als Funktion der Zeit,
• Fig. 3 zeigt eine schematische perspektivische Ansicht und eine Querschnittsansicht einer Fahrerkabine eines Arbeitsfahrzeuges,
• Fig. 4 zeigt ein schematisches Blockschaltbild einer Fahrerkabine mit Komponenten zur aktiven Lärmreduzierung,
• Fig. 5 zeigt ein Blockschaltbild des Regelkreises mit aktiver Lärmreduzierung und
• Fig. 6 zeigt ein Blockdiagramm des Regelkreises als digitales System.

The invention is explained below on the basis of an exemplary embodiment in conjunction with the drawings, in which:

• Fig. 1 shows the engine firing frequency of a forklift powered by an internal combustion engine as it accelerates as a function of time;
• Fig. 2 shows the wheel lug frequency of the accelerating forklift as a function of time,
• Fig. 3 shows a schematic perspective view and a cross-sectional view of a driver's cab of a work vehicle,
• Fig. 4 shows a schematic block diagram of a driver's cab with components for active noise reduction,
• Fig. 5 shows a block diagram of the control loop with active noise reduction and
• Fig. 6 shows a block diagram of the control loop as a digital system.

In the driver's cab of a work vehicle, the primary signal (background noise) is measured with a microphone, also referred to herein as an error microphone. The background noise is a multi-tone signal to which various frequency components contribute, which in turn are functions of time such as when the work vehicle is accelerating. An example of a contributing signal component is in Fig. 1 the engine ignition frequency of an accelerating forklift shown as a function of time. Another example is in Fig. 2 the wheel lug frequency of the accelerating forklift as a function the time shown, the forklift tires with 28 Radstollen, which are distributed over the circumference and come in contact with the driving surface successively in contact. The Radstollenfrequenz is then 28 times the wheel speed. Other signal components are the engine speed, which is 2/3 of the engine ignition frequency in a 3-cylinder, 4-stroke engine and harmonics of the components mentioned. The amplitudes of the individual components in the primary signal are dependent on the structural and cavity resonances of the driver's cab and also change depending on the driving behavior. Thus, the primary signal is a multitonal time-varying signal with frequency and amplitude modulation.

An acoustic problem consists in the excitation of the cavity resonances of the driver's cab in the frequency range of 30 Hz to 50 Hz by the resulting from the engine rotational frequency and the wheel rotation frequency sinusoids of the primary signal. In order to sufficiently excite these relatively low frequencies by means of an actuator, a disk with an inertial oscillator mounted on its surface is used as the actuator device.

Fig. 3 schematically shows a driver's cab with a seat 10. By a dashed line a rear window is indicated, which is closed by a disc. Other windows on the sides and in the front of the cab are not shown for ease of illustration. Attached to the rear disc off-center, close to a corner, is an inertial vibrator 4 whose axis of oscillation is perpendicular to the surface of the disc and which typically has a resonant frequency in the range of 20 to 90 Hz. On the housing of the inertial vibration exciter 4, an error microphone 6 is mounted, that receives the error signal in close proximity to the oscillator and directs to a control unit described below. The control unit then controls the Operation of the inertial vibration exciter 4 in such a manner that the vibrations of the rear disc caused by the inertial vibrator 4 generate counter-noise such that the error signal received by the error microphone 6 is minimal.

In Fig. 3 In addition to the error microphone 6 is still a monitor microphone 8 indicated. However, the signal of this monitor microphone 8 is not included in the control, but is analyzed in this embodiment only for control to check how the active noise reduction near the head of the driver, at the top of the backrest of the driver's seat.

In the following, the regulation of the inertial vibration exciter as a function of the error signal will be described. The control signal is generated by the inertial vibration exciter on the inside of the rear window. The error microphone 4 is placed in close proximity to the inertial vibrator 4 thereto. The controller is currently implemented with Matlab Simulink on RCP platform (DS1202, dSpace); in the final execution, the controller is implemented on a serial DSP. The block diagram of the structure with primary signal generation, control loop and evaluation of the sound reduction by the monitor microphone is in Fig. 4 shown.

In the block diagram in Fig. 4 the cabin is shown schematically as a dashed block. Disturbances such as engine and tire noise arrive via an interference transmission to and into the driver's cab and are detected there by the error microphone. The signal from the error microphone is amplified and then fed to an analog-to-digital converter. The digitized signal is passed to a controller in which an adaptive and predictive control algorithm is performed. The output of the regulator is then returned to a digital-to-analog converter transferred an analog signal, which is supplied to the actuator after amplification. The actuator is formed in the present case by the inertial vibrator on the disc. The controller controls the control signal for the inertial oscillator so that the generated by the inertial oscillator and the vibrating disc caused by vibration produced counter-structure after the structure-acoustic transmission 1 is superimposed with the external interference and extinguished at the location of the error microphone. Structure-acoustic transmission means that the mechanical speed of the vibration exciter is not transferred directly to the speed of the adjacent air layer via a rigid piston surface. The Inertial-Schwingerreger brings simplified a point force on the disc. The disk responds via the location and frequency-dependent transmission mobility with the surface speed, which in turn causes a sound pressure at different locations in the cabin via the location and frequency-dependent transmission impedance. Accordingly, since the monitor and the error microphone are located at different locations in the cabin, there are two different transmission mobility transmission impedance paths from the actuator (vibrator) to the sensors located in Fig. 4 are referred to as structure-acoustic transmission 1 and 2. Analogously, the interference transmission to two different locations takes place via two different paths, so that it must be distinguished between interference transmission to the error sensor and the monitor sensor, although the source of interference is identical.

In the block diagram in Fig. 4 The two amplifiers, the analog-to-digital converter, the controller, and the digital-to-analog converter together form the unit otherwise referred to in this application as the controller.

In Fig. 5 is shown a block diagram of the control loop, wherein compared to Fig. 4 the controller off Fig. 4 in several Subcomponents is subdivided. The signal of the error microphone is subjected to an analog-to-digital conversion after passing through an amplifier. An antialiasing filter then acts on the digitized signal. It should be noted here that the anti-aliasing filter is not realized here as a low-pass filter for the analog signal, but causes a sampling with a frequency of at least 20 kHz of the digitized signal. This has the advantage that the delay caused by the electrodynamic inertia of an analog low-pass filter can be avoided because the high-frequency sampling of the digitized signal is not associated with such a delay. In order to limit the computational effort in the actual control algorithm, a down-sampler is applied, which effects a down-sampling to 2 kHz. The actual control algorithm then runs in the block designated IMC (Internal Model Control). The output of the control algorithm then goes through an up sampler and then an interpolation filter.

Finally, the output of the interpolation filter is subjected to a digital-to-analog conversion, the analog signal amplified and fed as a control signal to the inertial oscillator. Then, the structure-acoustic transmission of the vibrated disc over the volume of air in the interior of the cab takes place to the error microphone, where the counter sound generated in this way is to cancel out the external interference sound by superposition.

It is a multirate processing (20 kHz A / D - D / A - conversion, 2 kHz signal processing) according to the block diagram in Fig. 5 realized.

The controller H ( z ) is implemented as an adaptive (normalized FxLMS) IMC with pre-equalization of the secondary path. Under assumption error-free A / D and D / A conversion is the control loop as discrete-time system in Fig. 6 shown.

The adaptive unit update specification (nFxLMS) c = [ c ₀ , c ₂ , ..., c _{M -1} ] ^T $$\mathit{c}\left(n+1\right)=\mathit{c}\left(n\right)-\mathit{\mu }\left(n\right)\hat{\mathit{d}}\text{'}\left(n\right)e\left(n\right)\mathrm{,}$$

gebildet, wobei * der Faltungsoperator ist. Hierbei ist $${\hat{s}}_{\mathrm{ges}}\left(n\right)=\hat{s}\left(n\right)\ast {\hat{s}}_{\mathrm{inv}}\left(n\right)$$

die Impulsantwort der Gesamtsekundärstrecke mit ŝ(n) als Schätzung der Impulsantwort der Sekundärstrecke s(n) und ŝ _inv(n) als Impulsantwort eines Entzerrungsfilters für die Sekundärstrecke. Die Referenzsignalsynthese erfolgt mit dem Ausgangssignal der adaptiven Einheit und dem Fehlersignal $$\hat{d}\left(n\right)=e\left(n\right)-\tilde{y}\left(n\right)\ast {\hat{s}}_{\mathrm{ges}}\left(n\right)\mathrm{.}$$

The vector of the filtered reference signal d ' ( n ) = [ d ' ( n ), d '( n -1), ..., d' ( n - M + 1)] ^T is obtained from the reference signal $$\hat{d}\text{'}\left(n\right)=\hat{d}\left(n\right)*{\hat{s}}_{\mathrm{ges}}\left(n\right)$$

where * is the convolution operator. Here is $${\hat{s}}_{\mathrm{ges}}\left(n\right)=\hat{s}\left(n\right)*{\hat{s}}_{\mathrm{inv}}\left(n\right)$$

the impulse response of the total secondary distance with ŝ ( n ) as an estimate of the impulse response of the secondary path s ( n ) and ŝ _inv ( n ) as an impulse response of an equalization filter for the secondary path. The reference signal synthesis is done with the output of the adaptive unit and the error signal $$\hat{d}\left(n\right)=e\left(n\right)-\tilde{y}\left(n\right)*{\hat{s}}_{\mathrm{ges}}\left(n\right)\mathrm{,}$$

$$P_{\hat{d}ʹ}\left(n\right)=\left(1-\beta \right)P_{\hat{d}ʹ}\left(n-1\right)+\beta {\left\{\hat{d}ʹ\right\}}^2\left(n\right),$$

mit dem Glättungsfaktor β und der additiven Konstante P _min für die Begrenzung des Nenners bei zu niedrigen Signalleistung. Entzerrungsfilter Ŝ _inv(z) wird eingesetzt, um die Nachteile der starken Frequenzselektivität der Sekundärstrecke S(z) für die Adaption zu verringern. Die Berechnung des Entzerrungsfilters (FIR-Filter) $${\hat{S}}_{\mathrm{inv}}\left(z\right)={\displaystyle \sum _{k=0}^K}{\hat{s}}_{\mathrm{inv},k}\left(z^{-k}\right),$$

als lineares Prädiktionsfilter erfolgt aus dem Modell der Sekundärstrecke $$\hat{S}\left(z\right)={\displaystyle \sum _{l=0}^L}{\hat{s}}_l\left(z^{-l}\right)$$

durch Lösen der Gleichung (lineare Prädiktion mit LMS) $$\left[\begin{array}{cccc}r\left(0\right)& r\left(1\right)*& \cdots & r\left(K-1\right)\\ r\left(1\right)& r\left(0\right)& \ddots & ⋮\\ ⋮& \ddots & \ddots & r\left(1\right)*\\ r\left(K-1\right)& \cdots & r\left(1\right)& r\left(0\right)\end{array}\right]\left[\begin{array}{c}{\hat{s}}_{\mathrm{inv},1}\\ {\hat{s}}_{\mathrm{inv},2}\\ ⋮\\ {\hat{s}}_{\mathrm{inv},K}\end{array}\right]=\left[\begin{array}{c}-r\left(2\right)\\ -r\left(3\right)\\ ⋮\\ -r\left(K\right)\end{array}\right],$$

mit dem Autokorrelationsvektor r = [r(0) r(1) ... r(K)]^T des Koeffizientenvektors ŝ = [ŝ ₀ ŝ ₁ ... ŝ _L]^T des Sekundärstreckenmodells. Hierbei steht (.)* für konjugiert komplex. Es gilt immer _{ŝ inv,0} = 1 und K ≤ L. Um unerwünschte Dämpfung der Sekundärstrecke durch den Entzerrer (Prädiktionsfilter) zu vermeiden, werden die Koeffizienten ŝ _inv = [1 ŝ _inv,1 ... ŝ _inv,K ]^T auf die minimale Prädiktionsfehlerleistung $$J=\sqrt{{\displaystyle \sum _{i=0}^I}{\hat{s}}_{\mathrm{ges},i}^2}$$

normiert. Hierbei sind ŝ _ges,i die Abtastwerte der endlichen Impulsantwort ŝ _ges(n) = ŝ(n) * ŝ _inv(n).Normalization of the convergence factor is $$\mathit{\mu }\left(n\right)=\frac{{\mathit{\mu }}_0}{P_{\hat{d}\text{'}}\left(n\right)+P_{\min }}.$$

$$P_{\hat{d}\text{'}}\left(n\right)=\left(1-\beta \right)P_{\hat{d}\text{'}}\left(n-1\right)+\beta {\left\{\hat{d}\text{'}\right\}}^2\left(n\right).$$

with the smoothing factor β and the additive constant P _min for limiting the denominator when the signal power is too low. Equalization filter Ŝ _inv ( z ) is used to reduce the disadvantages of the strong frequency selectivity of the secondary path S ( z ) for the adaptation. The calculation of the equalization filter (FIR filter) $${\hat{S}}_{\mathrm{inv}}\left(z\right)={\displaystyle \underset{k=0}{\overset{K}{\Sigma }}}{\hat{s}}_{\mathrm{inv}.k}\left(z^{-k}\right).$$

as a linear prediction filter takes place from the model of the secondary route $$\hat{S}\left(z\right)={\displaystyle \underset{l=0}{\overset{L}{\Sigma }}}{\hat{s}}_l\left(z^{-l}\right)$$

by solving the equation (linear prediction with LMS) $$\left[\begin{array}{cccc}r\left(0\right)& r\left(1\right)*& \cdots & r\left(K-1\right)\\ r\left(1\right)& \mathrm{<figure-callout\; id="0"\; label="r"\; filenames="imgf0001.png"\; state="\{\{state\}\}">r</figure-callout>}\left(0\right)& \ddots & ⋮\\ ⋮& \ddots & \ddots & r\left(1\right)*\\ r\left(K-1\right)& \cdots & r\left(1\right)& r\left(0\right)\end{array}\right]\left[\begin{array}{c}{\hat{s}}_{\mathrm{inv}.1}\\ {\hat{s}}_{\mathrm{inv}.2}\\ ⋮\\ {\hat{s}}_{\mathrm{inv}.K}\end{array}\right]=\left[\begin{array}{c}-r\left(2\right)\\ -r\left(3\right)\\ ⋮\\ -r\left(K\right)\end{array}\right].$$

with the autocorrelation vector r = [ r (0) r (1)... r ( K )] ^{T of} the coefficient vector ŝ = [ ŝ ₀ ŝ ₁ ... ŝ _L ] ^{T of} the secondary line model. Where (.) * Stands for conjugate complex. It is always _{ŝ inv, 0} = 1 and K ≤ L. To avoid unwanted attenuation of the secondary path through the equalizer (prediction), the coefficients ŝ _inv = [1 ŝ _{inv, 1} ... ŝ _{inv, K} ] ^T on the minimum prediction error power $$J=\sqrt{{\displaystyle \underset{i=0}{\overset{I}{\Sigma }}}{\hat{s}}_{\mathrm{ges}.i}^2}$$

normalized. Here ŝ _{ges, i are} the samples of the finite impulse response ŝ _ges ( n ) = ŝ ( n ) * ŝ _inv ( n ).

Claims (8)

A work vehicle having a driver's cab (10) having a closed cabin interior enclosed by structurally stable cabin wall elements and at least one pane closing a cabin wall opening left free by the cabin wall elements, actuator means acting on the pane, sensor means and sensor means and The control unit connected to the actuator device, which is set up to control the actuator device for noise reduction in the cabin interior in dependence on signals of the sensor device, characterized in that the actuator device is an inertial vibration exciter (4) fastened on the inner surface of the disc, the sensor device on the cabin interior directed microphone (6) and the control unit is adapted to drive the inertial vibrator (4) with an adaptive and predictive control in response to the microphone signals n to perform an active noise reduction in the cabin interior. Work vehicle according to claim 1, characterized in that the inertial vibration exciter (4) horizontally and vertically outside the center of the disc is fixed to the disc. Work vehicle according to one of the preceding claims, characterized in that the microphone (6) is placed on or in the housing of the inertial vibration exciter (4) and is aligned so aligned that a central surface normal on the membrane of the microphone perpendicular to the axis of oscillation of the Inertial vibration exciter (4) stands. Work vehicle according to one of the preceding claims, characterized in that instead of a low-pass filtering of the microphone signal, the control unit is adapted to sample the microphone signal at a sampling frequency of at least 20 kHz. Work vehicle according to claim 4, characterized in that the control unit is adapted to perform after sampling the microphone signal down-sampling to at most 2 kHz, thereby reducing the computational effort in the remaining control algorithm. Work vehicle according to one of the preceding claims, characterized in that the control unit is provided with an equalization filter in the adaptive and predictive control algorithm, which is inverse to the minimum phase portion of the frequency response of the controlled system, to make the convergence time of the adaptive and predictive control algorithm frequency independent and shorten , Work vehicle according to one of the preceding claims, characterized in that the control unit is adapted to adaptively perform the adaptive and predictive control algorithm only up to a predetermined cutoff frequency. Work vehicle according to one of the preceding claims, when dependent on claim 3, characterized in that the control unit is mounted together with their electrical circuits and processor units as a control module to the housing of the inertial vibration exciter or integrated therein.

Images