CN117574490A - BIM-based building light environment neural network control method - Google Patents
BIM-based building light environment neural network control method Download PDFInfo
- Publication number
- CN117574490A CN117574490A CN202311460071.5A CN202311460071A CN117574490A CN 117574490 A CN117574490 A CN 117574490A CN 202311460071 A CN202311460071 A CN 202311460071A CN 117574490 A CN117574490 A CN 117574490A
- Authority
- CN
- China
- Prior art keywords
- building
- light environment
- environment
- bim
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000005265 energy consumption Methods 0.000 claims abstract description 43
- 238000004088 simulation Methods 0.000 claims abstract description 30
- 230000008859 change Effects 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 10
- 238000005286 illumination Methods 0.000 claims description 38
- 230000006870 function Effects 0.000 claims description 18
- 238000012546 transfer Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 9
- 238000013461 design Methods 0.000 claims description 8
- 238000002834 transmittance Methods 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 239000013598 vector Substances 0.000 claims description 4
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 2
- 230000001960 triggered effect Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 18
- 239000000463 material Substances 0.000 description 10
- 238000012423 maintenance Methods 0.000 description 8
- 239000011521 glass Substances 0.000 description 6
- 230000035945 sensitivity Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000009413 insulation Methods 0.000 description 4
- 238000004321 preservation Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000010438 heat treatment Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 239000000178 monomer Substances 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 230000001105 regulatory effect Effects 0.000 description 3
- 239000002689 soil Substances 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000004907 flux Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000006798 recombination Effects 0.000 description 2
- 238000005215 recombination Methods 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 239000005413 snowmelt Substances 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000005034 decoration Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004313 glare Effects 0.000 description 1
- 238000005338 heat storage Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 230000000243 photosynthetic effect Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000005068 transpiration Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computational Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Structural Engineering (AREA)
- Civil Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Architecture (AREA)
- Pure & Applied Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Circuit Arrangement For Electric Light Sources In General (AREA)
Abstract
The invention discloses a building light environment neural network control method based on BIM, which comprises the following steps: step one, building light environment information acquisition; step two, building a BIM environment; thirdly, building a BIM model by using a revit, inputting collected surrounding environment information into software by using urban thermal environment system software Envi-met, and simulating the influence of the change of the surrounding environment data on a building; and step four, determining a light environment scheme mode according to light environment analysis, and carrying out refined control. The invention finds the optimal scheme with the lowest energy consumption under the condition of meeting the comfort level requirement of the light environment by adopting the BIM model for simulation. By adopting the building light environment neural network technology, a customizable inlet can be reserved for the building light environment neural network system, and different customizable options can be reserved, so that different requirements of different functional spaces on the light environment can be ensured.
Description
Technical Field
The invention relates to a building light environment control method, in particular to a building light environment neural network control method based on BIM, and belongs to the technical field of intelligent building.
Background
BIM (building information model) technology has the latest starting of operation and maintenance application in the application of design, construction and operation and maintenance stages, and the limitation of the current application is also the greatest. At present, application of BIM technology in operation and maintenance stage is mainly embodied in visual experience aspects of information storage, operation and maintenance. The vast amount of information in the BIM model is either filtered out during the lightweight process or is idle because it is not mined into the application scenario. The technology utilizes the BIM technology to utilize the light environment and the energy consumption in the operation and maintenance, and increases the application of the BIM technology in the operation and maintenance stage.
Meanwhile, the prior art has little research on BIM technology application and building light environment neural networks. The management of the existing intelligent building only makes intelligent judgment and execution aiming at inherent physical characteristics, technical characteristics and use characteristics inside the building, can be only applied to the application fields of extreme conditions such as fire alarm, security monitoring and air quality monitoring inside the building, lacks regulation and control on the use environment of daily people, and lacks real-time regulation and control on the change of the external light environment, the thermal environment, the wind environment, the sound environment and the air environment of the building. The invention combines and applies BIM technology, energy consumption calculation software, urban thermal environment system software and neural network technology to the building, so that the BIM technology can be used for adjusting and controlling the use environment of daily people besides the application field of extreme conditions.
Disclosure of Invention
The prior art has little research on BIM technology application and building light environment neural networks. The invention provides a building light environment neural network control method based on BIM, which is used for combining and applying BIM technology, energy consumption calculation software, urban heat environment system software and neural network technology to a building, so that the BIM technology can regulate and control daily use environment except the application field of extreme conditions, and can detect and cope with changes of external light environment, heat environment, wind environment, sound environment and air environment of the building in real time.
The invention is realized in the following way:
a building light environment neural network control method based on BIM comprises the following steps:
step one, building light environment information acquisition;
step two, building a BIM environment;
thirdly, building a BIM model by using a revit, inputting collected surrounding environment information into software by using urban thermal environment system software Envi-met, and simulating the influence of the change of the surrounding environment data on a building;
and step four, determining a light environment scheme mode according to light environment analysis, and carrying out refined control.
According to the invention, the BIM technology is adopted to perform simulation calculation and comparison of environment-related information and building-related information of different buildings around the built building address, and the light environment measure of the building with the lowest energy consumption is obtained, so that the building is intelligently operated, the use efficiency of the light environment of the actual building is improved, and the overall energy consumption of the building is reduced.
The further scheme is as follows:
the first step specifically comprises:
and determining a light environment acquisition range. The method comprises the steps of dividing a light environment acquisition area by adopting a 3 m-3 m square grid within 1 km with a building as a center, respectively acquiring lighting coefficients and illumination in each acquisition area, and analyzing light comfortableness.
The complexity of building light environment information acquisition comes from the complex and diverse requirements of indoor environment comfort level, building heating and air conditioner energy consumption control: the light environment information acquisition requirements become regional, complex and diversified due to the factors such as the interaction among different building groups, the interaction among different building monomers in the building groups, the different directions of different rooms in the monomer building, and the different peripheral protection forms and materials of the rooms in the same direction. And, as the refinement of the region division is enhanced, the complexity and diversity thereof are increased. The acquisition of the light environment coefficients therefore employs a relatively accurate 3m x 3m square grid division.
The building external light environment information acquisition device is called a light environment information acquisition device (hereinafter referred to as acquisition device), and the acquisition device forms a building external light environment acquisition system: the collector is composed of a sensor and a protection shielding structure, wherein the sensor mainly senses the intensity of light, and is used for assisting in monitoring the sensor by radiant energy in a certain spectral range.
The collector structure is provided with a reasonable light receiving angle, a reasonable installation environment is selected, only direct sunlight and energy radiation are collected, and the influence of local environment light and reflected light is avoided or greatly reduced.
The accuracy of the collection is influenced by external factors such as the shielding of direct sunlight and the influence of ambient light, and the sensors in the collector take various measures to exclude the accuracy and the persistence of the collection of any external factors. The specific measures are as follows:
in order to remove sky-falling foreign matters and flying birds from falling and shielding, two collectors with a certain distance (3 m) are adopted at one collecting point, the two collectors can be compared with each other, once the data are obviously different for more than 3 times, the 3 rd collecting point is required to be started, the data are measured again, the difference of the collected data is judged, and if the difference occurs again, manual intervention is required, and the collectors are warned and indicated to be maintained.
In real life, the collector arrangement is also of significance in practical use when surrounding buildings and ambient light are unavoidable to be shielded (for example, no shielding during daytime is unavoidable in a cold day).
In the region that has snow, customization collector setting is carried out, sets up snow melt device and adopts snow melt structure can avoid snow on the collector.
Because the number of the sensors is limited and the sensors are located at the highest point, the light environment information of each part of the body building can not be collected, and therefore the sensors are only limited to the collection of direct sunlight, and the surrounding building curtain wall collects parameters in a BIM simulation mode for the influence of the surrounding building curtain wall on the light environment. Aiming at the light pollution of the peripheral building curtain wall, as the incidence direction of the curtain wall primary reflection light pollution is opposite to the direct sunlight direction, and the influence time and the pollution intensity have fixed characteristics, the collector is specially designed, one collector adopts multiple sensors to combine and shelter and separate, each sensor is responsible for a certain light receiving angle, and the multiple sensors are combined into a complete light receiving angle. See fig. 5 and 6.
And the computer is used for identifying, filtering and algorithmic combination of the data collected by each sensor so as to eliminate and weaken the influence of the environmental light pollution source.
The layout of the collector corresponds to the service scope and service object of the neural network. The accuracy of the acquired information is ensured, and the accuracy of the sunlight environment information is ensured; and the global property is ensured.
The service range can be formed by combining and integrating a plurality of acquisition areas, and is formed by a city patch area, a neighborhood, a district and a building monomer from large to small. One service area may be shared by multiple building neural networks. The collector is placed at a high point within the service area.
The further scheme is as follows:
the second step specifically comprises:
the building light environment acquisition system can be unified and global based on the assistance of BIM technology. BIM environment includes a series of objects with fixed characteristics such as building, structure, ground, road, greening and landscape. The BIM environment is the result of the building neural network service object body and the environment virtualized by adopting the BIM technology, and consists of a building and various BIM models influencing the external light environment of the building. And building a BIM model by using a revit, inputting the collected surrounding environment information into the software by using urban thermal environment system software Envi-met, simulating the influence of the change of the surrounding environment data on the building, and realizing digital twin in a real sense.
The further scheme is as follows:
the third step specifically comprises:
and building a BIM model by using the revit, inputting the collected surrounding environment information into the software by using urban thermal environment system software Envi-met, and simulating the influence of the change of the surrounding environment data on the building. After the building external light environment acquisition system and the BIM environment are built, the building of the building BIM model light environment data neural network is deepened and supplemented according to building light environment simulation requirements. And after being combined with various executing mechanisms, the intelligent adjusting function can be realized.
Mainly uses buildings, structures, ground, roads, landscapes and greening near the service body and objects which can generate light pollution to the service body in a long distance.
When the service environment range is larger, the service environment model can acquire basic information based on the existing regional GIS system, select potential influencing objects, such as a high-rise glass curtain wall building which is far beyond the service ontology by a few kilometers, finely adjust light environment data of the high-rise glass curtain wall building, and bring the light environment data into the service environment BIM model creation range. And adopting a function gradient grid in software outside a range of 1 km, connecting the Bezier curve grid with a gradient formula, and increasing the grid spacing.
The precision of the service body is far greater than the precision of the service environment, the service body is determined by integrating all building operation and maintenance requirements, the precision of a general model is not less than LOD500, the LOD500 corresponds to a completion stage, and the model can be used for completion settlement and is integrated into a building operation and maintenance system as a central database. The creation of building body models (including decoration) is the most important content in the step, and the key point is that sunshade and shading components such as doors and windows are required to be built; besides the material information, the material information also comprises physical information such as a thermal resistance value, a heat storage coefficient and the like of the enclosure structure. It should be noted that the service ontology may also be one of the influencing factors of the service environment.
And (3) linking the obtained data into a BIM model in real time, and performing indoor heat energy simulation calculation on the light environment structure with the surrounding landscape and the building by using Design Builder software to obtain light environment energy consumption simulation data. And analyzing the light environment simulation data by using SPSS data analysis software to obtain a light environment comparison result.
In nature, LSG takes the maximum value of 2.13 and the minimum value of 0.77
And comparing the heat conduction through the light-transmitting enclosure structure, the solar heat obtaining and the heat obtaining, and evaluating the heat preservation and insulation performance of the window in winter and the daylighting and sunshade performance in summer through two indexes of a heat transfer coefficient K and a light-heat ratio LSG. And meanwhile, comparing the data simulated by the specific model in the BIM to obtain an energy consumption comparison result, analyzing the data, and comparing the optimal energy consumption simulation data of 3 groups of light environment parties by using the BIM energy consumption to simulate the window under the condition that the energy consumption comparison and the light shielding requirements are violated, for example, the energy consumption is low, the light shielding is large, the evacuation is small, and the coupling relation exists between the two.
The building envelope performance can be comprehensively evaluated by using two parameters of heat preservation and insulation performance K in winter and daylighting and sun shading performance LSG in summer, whether a window is opened and the opening angle is regulated by the size of the K value, whether a window curtain adopts a semitransparent form or a full transparent form, whether an air conditioner is required to be opened indoors, intelligent household setting is carried out, indoor light is ensured to be sufficient, if insufficient, a local light source is automatically opened for intelligent control, the internal daylighting requirement is ensured, and the energy consumption requirement is reduced.
According to different climatic regions, different requirements are adopted, and taking a winter-cooling-summer-heating region as an example, a scheme with an adjustable heat transfer coefficient and a photo-thermal ratio index in a simulation model is the optimal choice. According to the building sensitivity, a light environment strategy and a light environment combination scheme, such as a sensitivity high-area superposition recombination comprehensive light environment scheme, for example, artificial light environment and natural light environment unified consideration, and spatial light environment unified consideration are established. The corresponding low sensitivity zone adopts a basic light environment scheme.
(4) And according to the light environment prediction and deduction analysis, determining the continuous feedback and continuous adjustment of the light environment scheme mode.
Constructing a fuzzy neural network control system, wherein the fuzzy neural network control system comprises a neural network prediction model and a fuzzy controller; the neural network prediction model takes the illumination intensity of an actual collector, the illumination intensity simulated by BIM, the visible light transmittance and the heat transfer coefficient as input vectors, takes the actual illumination intensity after corresponding sun-shading measures as output vectors, and predicts the effectiveness of the sun-shading measures; the fuzzy controller uses the deviation Lux and the deviation change rate Lux of the actual illumination intensity and the expected illumination intensity (taking expert experience and national illumination standard as references) after taking corresponding sun-shading measures c And taking the sunshade area difference delta s and the indoor temperature difference delta t as output quantities as input quantities, thereby obtaining a sunshade area deviation delta s, adding the sunshade area deviation delta s and the sunshade area s to obtain a final sunshade area s+delta s, and taking the final sunshade area s+delta s as a sunshade area value input in the next cycle of the neural network prediction model. The fuzzy controller adopts a comprehensive generation method of a language type fuzzy system reasoning method to obtain a fuzzy result.
The neural network prediction model adopts a BP algorithm of a learning process of an error counter-transmission error counter-propagation algorithm, and is of a three-layer structure, wherein the first layer is an input layer, the number of nodes is 4, and the nodes are respectively the illumination intensity of an actual collector, the illumination intensity of BIM simulation, the visible light transmittance and the heat transfer coefficient; the second layer is an hidden layer, and the number of nodes is 4; the third layer is an output layer, the number of nodes is 1, namely the actual illumination intensity after the corresponding sun-shading measures are taken.
The hidden layer transfer function of the neural network prediction model adopts an S-shaped transfer function logsig; the momentum batch gradient descent method (TRAINGDM) trains function triggers; the training algorithm adopts a quasi-Newton method; the maximum training frequency is set to 10 3 Secondary times; training object of 10 -2 The method comprises the steps of carrying out a first treatment on the surface of the By first calculating the direction optimization parameters of quasi-Newton training and thenFinding the appropriate learning rate.
The BP algorithm uses the square of the network error as an objective function, and adopts a gradient descent method to calculate the minimum value of the objective function.
A database is established, sensing data acquired by a sensor arranged on a building roof is stored in the database, and a neural network prediction model selects representative data from the sensing data and continuously carries out autonomous learning.
And (3) performing real-time control according to the optimal scheme analyzed and evaluated in the step (3), and taking corresponding light environment measures. And feeding back the light environment effect to the building BIM model in real time through the sensor. And determining a plurality of light environment combination schemes according to the light environment comfort level requirement. And (3) linking the obtained data into a BIM model in real time, using the BIM to simulate the influence of the sunlight intensity change of the micro environment on the energy consumption of the building, comparing the energy consumption simulation data of different light environment combination schemes, simultaneously comparing the data to obtain an energy consumption comparison result, determining a final light environment scheme under the condition of meeting the requirements of light environment comfort and minimum energy consumption, transmitting the final light environment scheme to the existing building, performing real-time control, and taking corresponding light environment measures. The strategy and mode of the light environment can be adjusted at any time through positive feedback and negative feedback.
In the framework of a programming model of the subsequent computer deep learning, through the deep learning and the addition of influence factors, the building light environment neural network system can combine comprehensive information such as seasonal factors, weather forecast, other types of sensor devices and the like to realize high intelligence:
scene 1, intelligently adjusting reflection information and color information of skin materials of a region under the distribution condition of snow parts in a post-snow service environment; the light environment energy consumption simulation data are distribution conditions of snow parts, the light environment comparison result is that the ground surface materials of places with more snow and places with less snow are different in color reflection information, and the outdoor lamp is controlled by connecting an operation platform after determining that the light environment scheme is that the outdoor lamp has illumination intensity at night.
Scene 2, the environmental model of the deciduous tree changes along with the seasonal change, and the transmissivity and color information of the deciduous tree in winter and summer are intelligently adjusted; the light environment energy consumption simulation data are the light transmittance under different fallen leaves in winter and summer, the light environment comparison result is the light transmittance in winter and summer, the light environment scheme is determined to be a winter light supplementing scheme, the adjustment of different illumination of the winter and summer lamps is distinguished, and the control is carried out by connecting a common operation platform.
The scene 3 can be used as special condition output information when the external light environment is frequently and greatly changed in cloudy weather, and an applicable execution scheme is made according to different requirements and application scenes; the light environment energy consumption simulation data are different in inverse color information of ground surface materials of places with more snow and places with less snow as a result of the light environment comparison, and the light environment scheme is determined to be outdoor illumination intensity after night snow, and the control is carried out through connecting an operation platform.
Scenario 4 may intelligently exclude situations where the architectural light environment neural network system is not applicable: for example, when the collector is only installed on the top of a super high-rise building, special conditions exist that the external light environments of the upper half of the building are different from each other outside the cloud layer and the lower half of the building is inside the cloud layer. The method is solved by adding a distributed auxiliary collector. The light environment energy consumption simulation data are obtained by respectively comparing indoor light environment distribution conditions below the cloud and fog layer, the light environment comparison result is whether indoor light below the cloud and fog layer meets the light environment requirement, and the light environment scheme is determined by adopting a local illumination and light supplementing mode and is controlled by connecting an operation platform.
The unpredictable light environment factors such as light pollution generated by the glass of the running or parking vehicle of the scene 5 and incidence on the collector can be eliminated when setting the boundary conditions of the model. The light environment energy consumption simulation data are used for eliminating light pollution incident light generated by driving or parking vehicle glass, the light environment comparison result is a stable light environment result obtained by removing unstable factors, and the light environment scheme is determined to be an indoor light environment treatment mode, such as windowing, curtain opening and the like, and is controlled by connecting an operation platform.
In extreme cases, intelligent judgment and emergency measures should be adopted for the possibility of the abnormal condition of the acquisition system. And outputting the filtered and processed execution information according to the application characteristics of different execution terminals, so as to realize the purpose of protecting the execution mechanism and the terminals.
The invention has at least the following outstanding beneficial effects:
through setting up the collector at the building highest point, and then make the light environment gather more accurate to realize better control. By adopting the BIM model to simulate, the optimal scheme with the lowest energy consumption under the condition of meeting the requirements of light environment comfort level can be found, and the method can be applied to the following fields:
1) Room light environment regulation: under the condition that an independent sensor in a living room is not needed, the indoor artificial illumination can be automatically started or regulated in real time according to the change of an external light ring; adjusting the sun-shading system to improve glare, adjust natural illuminance and light characteristics; the regular light pollution is automatically shielded.
2) Building thermal performance adjustment: the sunshade measures can be adjusted to form a building self-adaptive sunshade system which can improve the thermal performance of the outer protective structure of the building in summer; the device of participating in the heat collecting wall adjusts, realizes heat collecting wall intelligent regulation purpose.
By adopting the building light environment neural network technology, a customizable inlet can be reserved for the building light environment neural network system, different customizable options are reserved, and different execution schemes are formulated aiming at the characteristics of intelligent public buildings and intelligent living buildings so as to ensure different requirements of different functional spaces on light environments.
The information can be output in a standard format, and the requirements of users in each independent functional space on building finer neural networks and executing mechanisms by themselves are met.
The building light environment neural network system can record user customization data and data for assisting in intervention and adjustment by a user in the use process, and master rules of the user customization data and the data, so that the use habit of the user is gradually reflected, and the frequency of manual adjustment and intervention by the user is reduced.
The intelligent building light environment control system based on BIM combines the neural network control and the fuzzy control, realizes the self-adaption and autonomous learning of the light environment control system, lays a solid foundation for the combination of follow-up light environment intelligent furniture and intelligent manufacturing, and provides a new approach for light environment intelligent control.
Drawings
FIG. 1 is a schematic diagram of the present invention;
FIG. 2 is a schematic diagram of the operation of setting a "daylight path" in the autodesk revit software;
FIG. 3 is a schematic illustration of the operation of setting "building space type" in an autodesk revit software functional area;
FIG. 4 is a schematic diagram of the operation of selecting "list/quantity" in the autodesk revit software;
FIG. 5 is a schematic view of a collector collecting sunlight;
fig. 6 is a schematic structural diagram of the collector.
Detailed Description
The invention will now be described in further detail with reference to the drawings and to specific examples.
As shown in fig. 1, a building light environment neural network control method based on BIM comprises the following specific steps: (1) building light environment information acquisition; (2) building a BIM environment; (3) Building a BIM model by using a revit, inputting collected surrounding environment information into the software by using urban thermal environment system software Envi-met, and simulating the influence of the change of the surrounding environment data on a building; (4) And determining a light environment scheme mode according to light environment prediction and deduction analysis, continuously feeding back and continuously adjusting.
(1) Building light environment information acquisition: and planning a light environment acquisition range. The light environment collecting area is divided by adopting a 3*3 square grid within the range of 1 km, and the light environment information collectors are distributed on roofs of building groups to collect lighting coefficients and illumination.
(2) BIM environment creation.
A. Building a basic building model by using an Autodesk Revit.
And acquiring a BIM model of the building group, wherein planning information of the building group can be acquired, the Revit software is used as a carrier of the building information model, the BIM building model is built according to the planning information, or the BIM model is acquired according to a Revit forward design result, and the building group model comprises a plurality of surrounding buildings arranged within a certain distance. The size of the spacing between the surrounding building and the object building directly influences the lighting effect of the building.
The environment for establishing projects by utilizing the autodesk revit comprises a series of objects with fixed characteristics, such as a surrounding building model, a construction model, a ground and road, a model greening and landscape model and the like; while each model is numbered differently.
Wherein, there is specific differentiation and requirement to model precision, specific requirement is as follows:
the shape accuracy of the service environment model: the skin model is generally only created on the principle that the closer to the service ontology the higher the accuracy is. An internal model is created for the transparent structural environment.
Information accuracy of service environment model: in addition to the light shielding information generated by the model form, determining a potential influencing object, adding necessary information to the building model that influences the external light environment of the service body, including: the material information of the surface of the body comprises the material information of the line reflection characteristic, the color characteristic and the transmission material of the photo-thermal radiation.
Mainly uses buildings, structures, ground, roads, landscapes and greening near the service body and objects which can generate light pollution to the service body in a long distance.
When the service environment range is larger, the service environment model can acquire basic information based on the existing regional GIS system, select potential influencing objects, such as a high-rise glass curtain wall building which is far beyond the service ontology by a few kilometers, finely adjust light environment data of the high-rise glass curtain wall building, and bring the light environment data into the service environment BIM model creation range. And adopting a function gradient grid in software outside a range of 1 km, connecting the Bezier curve grid with a gradient formula, and increasing the grid spacing.
B. And building a sunlight model by utilizing the sunlight data information acquired by the Autodesk Revit collector.
And accurately positioning the geographical position information of the building, selecting a 'sunlight setting' from the 'sunlight path' of the autodesk device, and selecting a day to simulate the sunlight path. The "daylight path" is set to simulate changes in sunlight throughout the day, as shown in fig. 2.
The lighting load of the main room is determined by defining the building type according to different building functions and properties. If a living area project belongs to a residential building, parameters of the living building are filled in the setting related to illumination according to the requirements of GB50034 building illumination design Standard, and then a house is attached in the building in the later period. Areas with other properties such as a machine room are independently modified, and the workload of repeated definition of each area is saved. In the Revit model function area 'management' - 'MEP setting' -setting in building space type, automatically acquiring information of different rooms in a building through space functions in the Revit: area, volume, etc., as shown in fig. 3.
The illumination requirements are specified for areas of different functions within the building according to the requirements of the specification building illumination design criteria, as shown in fig. 4, the space type and the required illumination level are selected in the "analysis" - "list/number" attribute dialog "field". According to the actual condition of the project, adding rows for the detail table, and inputting the space type and corresponding illumination required to be used in the project.
After the parameter settings and the illumination requirements are finished, the lighting fixture arrangement is performed by creating a space illumination analysis list,
3. the microclimate model is used for calculating the software Envi-met, collected surrounding environment information (especially plants) is input into the software, and the influence of the change of the surrounding environment data on the environments such as temperature, humidity, wind speed and the like in the building is simulated.
Short and long wave radiant fluxes are calculated by comprehensively considering various shielding, surfaces, buildings and vegetation of the building and multiple reflections. Advanced modeling of the radiation process, including scattering and diffuse reflection, is performed in the plant canopy; the transpiration and heat flux of the plants were determined, including a comprehensive simulation of all plant physical parameters (e.g., photosynthetic rate). The feedback process between soil moisture and plant moisture was simulated. The moisture and heat exchange within the soil system was simulated. Three-dimensional heat transfer simulation was performed based on the floor material and moisture content. Advanced calculations include hydraulic moisture exchange in the soil, including root absorption and plant water supply, and micro-environmental climate research for buildings.
And (5) establishing a model. And acquiring construction geographic satellite information, and converting the acquired image format into an identifiable BMP format base map, and establishing a simulation model of a research area in ENVI-met software according to the base map, wherein the provided maximum grid number of the calculation area is 250 multiplied by 30, enough space is reserved in the horizontal direction, and the height in the vertical direction is more than or equal to 2 times of the highest construction height in the simulation area.
ENVI-met modeling in addition to modeling of building space information, vegetation, underlying surfaces, etc. all need to be set and modeled. And then calculating to check parameters such as wind speed, pressure, temperature, humidity, PM2.5 and the like.
The obtained ENVI-met outdoor microenvironment data is linked to a BIM model in real time, and the Design Builder software is utilized to perform indoor heat energy simulation calculation on the light environment structure with the surrounding landscape and the building, so that light environment energy consumption simulation data is obtained. And analyzing the light environment simulation data by using SPSS data analysis software to obtain a light environment comparison result.
And comparing the heat conduction through the light-transmitting enclosure structure, obtaining heat by insolation and obtaining heat, and evaluating the heat preservation and insulation performance of the window in winter and the daylighting and sunshade performance in summer through the two indexes of the heat transfer coefficient and the photo-thermal ratio.
And creating illumination requirements, and designating the illumination requirements for areas with different functions in a building according to the requirements of the specification building illumination design standard.
After the space illumination analysis is completed, each parameter setting and the illumination requirement assignment are completed, the data simulated in the BIM of the specific model can be simultaneously compared by creating a space illumination analysis detail table, so that an energy consumption comparison result is obtained, and the data are analyzed. Because the energy consumption comparison and the shading requirement are in a coupling relation, for example, the energy consumption is low, the comfort of the possible light environment is poor, and the BIM energy consumption is utilized to simulate a window, so that the optimal energy consumption balance scheme of the 3 groups of light environments is compared.
The building envelope performance can be comprehensively evaluated by using two parameters of heat preservation and insulation performance K in winter and daylighting and sun shading performance LSG in summer, whether a window is opened and the opening angle is regulated by the size of the K value, whether a window curtain adopts a semitransparent form or a full transparent form, whether an air conditioner is required to be opened indoors, intelligent household setting is carried out, indoor light is ensured to be sufficient, if insufficient, a local light source is automatically opened for intelligent control, the internal daylighting requirement is ensured, and the energy consumption requirement is reduced.
According to different climatic regions, different requirements are adopted, and taking a winter-cooling-summer-heating region as an example, a scheme with an adjustable heat transfer coefficient and a photo-thermal ratio index in a simulation model is the optimal choice. According to the building sensitivity, a light environment strategy and a light environment combination scheme, such as a sensitivity high-area superposition recombination comprehensive light environment scheme, for example, artificial light environment and natural light environment unified consideration, and spatial light environment unified consideration are established. The corresponding low sensitivity zone adopts a basic light environment scheme.
(4) The coupling relationship exists between the light environment, the light comfort, the heat environment, the heat comfort and the building energy consumption.
According to the requirements of the light environment and the energy consumption of the building, a balanced intermediate value is obtained, the comfort requirement of the light environment is met, the energy consumption of the building is reduced as much as possible, a real-time light environment scheme mode is determined, and the feedback and the regulation are continuously carried out. And (3) performing real-time control according to the optimal scheme analyzed and evaluated in the step (3), and taking corresponding light environment measures. And feeding back the light environment effect to the building BIM model in real time through the sensor. And determining a light environment scheme mode according to the light environment prediction and deduction analysis. And (3) linking the obtained data into a BIM model in real time, using the BIM to simulate the influence of the sunlight intensity change of the micro environment on the energy consumption of the building, comparing the energy consumption simulation data of different light environment combination schemes, simultaneously comparing the data to obtain an energy consumption comparison result, determining the light environment scheme mode, transmitting the energy consumption comparison result to the existing building, performing real-time control, and taking corresponding light environment measures. And the light environment effect is fed back to the building BIM model in real time through the sensor, and the strategy and mode of the light environment are adjusted at any time through positive feedback and negative feedback.
Although the invention has been described herein with reference to the above-described illustrative embodiments thereof, the above-described embodiments are merely preferred embodiments of the present invention, and the embodiments of the present invention are not limited by the above-described embodiments, it should be understood that numerous other modifications and embodiments can be devised by those skilled in the art that will fall within the scope and spirit of the principles of this disclosure.
Claims (7)
1. The building light environment neural network control method based on BIM is characterized by comprising the following steps:
step one, building light environment information acquisition;
step two, building a BIM environment;
thirdly, building a BIM model by using a revit, inputting collected surrounding environment information into software by using urban thermal environment system software Envi-met, and simulating the influence of the change of the surrounding environment data on a building;
and step four, determining a light environment scheme mode according to light environment analysis, and carrying out refined control.
2. The building light environment neural network control method based on BIM according to claim 1, wherein the method comprises the following steps:
in the first step, a 3 m-3 m square grid is adopted to divide a light environment collection area within a range of 1 km, and light environment information is collected through light environment information collectors distributed on roofs of building groups.
3. The building light environment neural network control method based on BIM according to claim 2, wherein:
the light environment information includes lighting coefficients and illuminance.
4. A building light environment neural network control method based on BIM according to claim 2 or 3, wherein:
the second step specifically comprises:
1) Building a basic building model by using an Autodesk Revit;
2) And building a sunlight model by combining the Autodesk Revit with the daily light environment information data acquired by the light environment information acquisition device.
5. The building light environment neural network control method based on BIM according to claim 4, wherein the method comprises the following steps:
the third step specifically comprises:
the obtained data are linked to a BIM model in real time, and indoor heat energy simulation calculation is carried out on the light environment structure with the surrounding landscape and the building by using Design Builder software, so that light environment energy consumption simulation data are obtained; analyzing the light environment simulation data by using SPSS data analysis software to obtain a light environment comparison result;
wherein LSG represents photo-thermal ratio, T lum Indicating visible light transmittance, and TIR indicating near infrared transmittance.
6. The building light environment neural network control method based on BIM according to claim 5, wherein the method comprises the following steps:
the fourth step comprises:
according to the optimal scheme analyzed and evaluated in the step three, real-time control is carried out, and corresponding light environment measures are adopted; feeding back the light environment effect to the building BIM model in real time through a sensor; determining a plurality of light environment combination schemes according to the light environment comfort level requirement; the obtained data are linked to a BIM model in real time, the influence of the change of the sunlight intensity of the micro environment on the building energy consumption is simulated by using the BIM, the energy consumption simulation data of different light environment combination schemes are compared, the data are compared at the same time, an energy consumption comparison result is obtained, the final light environment scheme is determined under the condition that the requirements of light environment comfort and the minimum energy consumption are met at the same time, the final light environment scheme is transmitted to the existing building, real-time control is carried out, and corresponding light environment measures are adopted; the strategy and mode of the light environment can be adjusted at any time through positive feedback and negative feedback.
7. The building light environment neural network control method based on BIM according to claim 6, wherein the method comprises the following steps:
the real-time control is realized by constructing a fuzzy neural network control systemThe fuzzy neural network control system comprises a neural network prediction model and a fuzzy controller; the neural network prediction model takes the illumination intensity of an actual collector, the illumination intensity simulated by BIM, the visible light transmittance and the heat transfer coefficient as input vectors, takes the actual illumination intensity after corresponding sun-shading measures as output vectors, and predicts the effectiveness of the sun-shading measures; the fuzzy controller takes the deviation e of the actual illumination intensity and the expected illumination intensity after taking the corresponding sun-shading measures, and the deviation change rate e c As input quantity, taking the sunshade area difference delta q and the indoor temperature difference delta t as output quantity, thereby obtaining a sunshade area deviation delta q, adding the sunshade area deviation delta q and the sunshade area q to obtain a final sunshade area q+delta q, and taking the final sunshade area q+delta q as a sunshade area value input in the next cycle of the neural network prediction model;
the neural network prediction model adopts a BP neural network and is of a three-layer structure, the first layer is an input layer, the number of nodes is 4, and the nodes are respectively the illumination intensity of an actual collector, the illumination intensity of BIM simulation, the visible light transmittance and the heat transfer coefficient; the second layer is an hidden layer, and the number of nodes is 4; the third layer is an output layer, the number of nodes is 1, namely the actual illumination intensity after the corresponding sun-shading measures are taken;
the hidden layer transfer function of the neural network prediction model adopts an S-shaped transfer function logsig; the TRAINGDM with momentum is triggered by a training function TRAINGDM; the training algorithm adopts a quick-Newtonmethod; the maximum training frequency is set to 10 3 Secondary times; training object of 10 -2 The method comprises the steps of carrying out a first treatment on the surface of the The quasi-Newton training direction optimization parameters are calculated for the first time, and then an appropriate learning rate is searched;
the BP algorithm takes the square of the network error as an objective function, and adopts a gradient descent method to calculate the minimum value of the objective function;
a database is established, sensing data acquired by a sensor of the building roof is stored in the database, and a neural network prediction model selects representative data from the sensing data and continuously carries out autonomous learning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311460071.5A CN117574490A (en) | 2023-11-03 | 2023-11-03 | BIM-based building light environment neural network control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311460071.5A CN117574490A (en) | 2023-11-03 | 2023-11-03 | BIM-based building light environment neural network control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117574490A true CN117574490A (en) | 2024-02-20 |
Family
ID=89883465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311460071.5A Pending CN117574490A (en) | 2023-11-03 | 2023-11-03 | BIM-based building light environment neural network control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117574490A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118171580A (en) * | 2024-04-09 | 2024-06-11 | 哈尔滨工业大学 | Photovoltaic roller shutter system for residential building in severe cold region and dynamic intelligent regulation and control method thereof |
-
2023
- 2023-11-03 CN CN202311460071.5A patent/CN117574490A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118171580A (en) * | 2024-04-09 | 2024-06-11 | 哈尔滨工业大学 | Photovoltaic roller shutter system for residential building in severe cold region and dynamic intelligent regulation and control method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Du et al. | A review of multi-scale modelling, assessment, and improvement methods of the urban thermal and wind environment | |
| Santamouris | Natural ventilation in buildings: a design handbook | |
| Okeil | A holistic approach to energy efficient building forms | |
| Erell et al. | Urban microclimate: designing the spaces between buildings | |
| CN103593530B (en) | A kind of integrated approach that adopts digital simulation to carry out energy saving building design | |
| Xu et al. | Performance-driven optimization of urban open space configuration in the cold-winter and hot-summer region of China | |
| Qiu et al. | Identification of passive solar design determinants in office building envelopes in hot and humid climates using data mining techniques | |
| Zhang et al. | A novel productive double skin façades for residential buildings: Concept, design and daylighting performance investigation | |
| KR20130114965A (en) | Design system of building having high energy efficiency using simulation and design process thereof | |
| CN115795638B (en) | Residential area layout optimization design method, system and storage medium | |
| CN116258239B (en) | Intelligent control energy carbon-grid management integrated system and carbon-grid management method | |
| Hong et al. | Optimal tree design for sunshine and ventilation in residential district using geometrical models and numerical simulation | |
| CN112329113A (en) | Building energy-saving design method, system, device and storage medium based on BIM | |
| Lee et al. | Assessing the performance potential of climate adaptive greenhouse shells | |
| KR20230150484A (en) | A Design Method for Environmentally Friendly Buildings Reflecting Energy Management System | |
| Tablada et al. | Vertical farming on facades: transforming building skins for urban food security | |
| CN117574490A (en) | BIM-based building light environment neural network control method | |
| KR102179250B1 (en) | Effect analysis system of temperature reduction activities for temperature controlling | |
| CN103823976B (en) | Heliogreenhouse light thermal environment computational methods | |
| Pan et al. | Simulating the impact of deciduous trees on energy, daylight, and visual comfort: impact analysis and a practical framework for implementation | |
| CN116861524A (en) | Green building design method based on BIM | |
| Thayer et al. | Modeling the effects of street trees on the performance of solar and conventional houses in Sacramento, California | |
| Maksoud et al. | Generating optimum form for vertical farms skyscrapers in UAE | |
| CN112883660A (en) | Functional facility configuration method based on cell microenvironment and outdoor comfort simulation | |
| Bhiwapurkar et al. | Street geometry and energy conservation of urban buildings in Chicago |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |