Real-time prediction method and system for dynamic boundary of indoor photosensitive area

Abstract

The invention relates to an indoor photosensitive area dynamic boundary real-time prediction method and system, and the method comprises the steps: obtaining a key building factor and a level thereof, carrying out an orthogonal test, carrying out the working condition division according to a test result, and selecting a target working condition; acquiring a plurality of parameter groups for the target working condition, and constructing an indoor photosensitive partition database based on the plurality of parameter groups; training a machine learning model by using the indoor photosensitive partition database to obtain a photosensitive region dynamic boundary prediction model; collecting static design parameters and dynamic scene parameters of a scene to be predicted in real time, and outputting an indoor photosensitive area dynamic boundary by using the photosensitive area dynamic boundary prediction model; the system provided by the invention is used for implementing the method. Compared with the prior art, the building photosensitive dynamic boundary identification method provided by the invention can overcome geometric constraints of a traditional simulation method, and has the advantages of universality, high efficiency and climate self-adaption.

Description

Technical Field

[0001] This invention relates to the field of energy conservation and environmental protection technology, and in particular to a method and system for real-time prediction of dynamic boundaries of indoor light-sensitive areas. Background Technology

[0002] Making full use of natural light is a core technology in modern green building and energy-saving design. It not only significantly reduces the energy consumption of artificial lighting systems, aligning with the low-carbon development concept, but also effectively improves the quality of the indoor light environment, enhances the comfort of living and working environments, and ultimately optimizes work efficiency, possessing significant energy-saving and environmental protection value and engineering application significance. However, in the practice of promoting the full utilization of daylight, a key technical challenge is faced: how to efficiently introduce beneficial natural light while precisely avoiding the negative impacts of direct sunlight, such as glare interference and excessive heat accumulation indoors. This technical challenge naturally leads to two functionally differentiated areas within the interior space: one is the interior depth area where soft, diffused light can be directly utilized, and the other is the window-side light-sensitive area where direct sunlight requires targeted control. Because the area of ​​the photosensitive zone dynamically changes with factors such as solar altitude angle, azimuth angle, and weather conditions, how to accurately define this dynamic photosensitive zone in real time is an urgent problem to be solved. Only through precise zoning can we provide a scientific basis for the targeted and coordinated regulation of shading systems, daylighting control systems, etc., effectively suppressing the negative effects of direct sunlight while maximizing the retention of beneficial natural light resources in deeper indoor areas, ultimately achieving the optimal balance between building energy efficiency and indoor living comfort. Existing research on indoor lighting mostly focuses on indoor light comfort. For example, Chinese patent application CN110334387A provides an indoor lighting prediction method based on a BP neural network algorithm. While this method predicts indoor lighting using a BP neural network algorithm and combines it with LED luminous flux and luminous flux transfer function matrix models, achieving efficient utilization of power resources and improved lighting uniformity in different seasons, it still cannot identify the dynamic boundary of the photosensitive zone. While some methods can identify light-controlled and uncomfortable lighting zones, the typical process involves: creating a computer model of the building with fixed geometric dimensions and specific materials, such as using SketchUp or Revit; performing hourly or minute-by-minute light environment simulations using specialized simulation software (e.g., Radiance, Daysim) combined with year-round meteorological data (TMY files) for the specific region; analyzing the simulation results for indicators such as vertical eye illuminance (Ev), glare probability (DGP), and illuminance at various points indoors; and finally, manually judging and generating the results. Although these methods can identify areas, their simulation results heavily rely on initially set fixed geometric parameters. If any parameter such as building orientation, window-to-wall ratio, or room depth changes, remodeling and resimulating the area requires significant time. Therefore, the conclusions drawn from this method lack universality and cannot form a set of universal engineering design guidelines or scales that can guide different projects, nor can they be easily integrated into building automation systems that require rapid response.

[0003] Therefore, providing a method for identifying building light-sensitive dynamic boundaries that can overcome the geometric constraints of traditional simulation methods and is universal, efficient, and climate-adaptive is a technical problem that needs to be solved. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a method and system for real-time prediction of the dynamic boundary of indoor light-sensitive areas. It aims to achieve accurate dynamic definition of indoor light-sensitive areas through systematic simulation and intelligent algorithms.

[0005] The objective of this invention can be achieved through the following technical solutions: According to a first aspect of the present invention, a method for real-time prediction of dynamic boundaries of indoor light-sensitive areas is provided, the method comprising: Key building factors and their levels are obtained, orthogonal experiments are conducted based on the key building factors and their levels, working conditions are divided according to the experimental results, and target working conditions are selected. For the target operating condition, multiple parameter sets are obtained, each of which includes static design parameters and dynamic scenario parameters. Based on these multiple parameter sets, the following operations are performed to construct an indoor photosensitive zoning database, which includes multiple data sets: Indoor lighting environment simulation is performed based on the aforementioned static design parameters and dynamic scenario parameters; In a simulated light environment, multiple judgment points are selected, and for each judgment point, the probability of sunlight glare and the illuminance of the working surface are calculated hourly. Based on the aforementioned probability of sunlight glare and illuminance of the working surface, sensitive and non-sensitive locations are determined, and a set of sensitive locations is constructed. Select a target sensitive point from the set of sensitive points, take the distance from the window to the target sensitive point as the photosensitive area depth of the current parameter group, and construct a data group with the photosensitive area depth and the corresponding static design parameters and dynamic scenario parameters. A machine learning model was trained using the aforementioned indoor photosensitive zone database to obtain a dynamic boundary prediction model for the photosensitive zone. The static design parameters and dynamic scenario parameters of the scene to be predicted are collected in real time, and the dynamic boundary of the indoor light-sensitive area is output using the aforementioned dynamic boundary prediction model of the light-sensitive area.

[0006] As a preferred technical solution, the key building factors include typical window orientation, typical window-to-wall ratio, room height, room width, room depth, visible light transmittance of glass, and reflectivity of walls, floors and ceilings. The static design parameters include: key building factors and the angle between the line of sight and the window normal; The dynamic scenario parameters are variables that change with time or environment, including date, time, location, and the sun's altitude and azimuth angles; the altitude and azimuth angles are obtained based on the date and time.

[0007] As a preferred technical solution, the method for selecting the judgment points is as follows: select a series of discrete judgment points along the direction perpendicular to the window.

[0008] As a preferred technical solution, the conditions for determining the sensitive and non-sensitive points are as follows: if the probability of sunlight glare is greater than the uncomfortable sunlight glare probability threshold, or the illuminance of the working surface is greater than the uncomfortable working surface illuminance threshold, then it is determined to be a sensitive point; otherwise, it is a non-sensitive point.

[0009] As a preferred technical solution, the target sensitive point is the sensitive point with the largest distance value from the window in the set of sensitive points.

[0010] According to a second aspect of the present invention, a real-time prediction system for the dynamic boundary of an indoor light-sensitive area is provided, the system comprising: Data acquisition module: This module is used to acquire parameter sets and collect static design parameters and dynamic scenario parameters of the scenario to be predicted in real time; Orthogonal test module: This module conducts orthogonal tests based on the acquired key building factors and their levels, and divides the work conditions according to the test results to select the target work condition; Database Construction Module: This module performs the following operations on each of the multiple parameter groups acquired under the target operating conditions to construct an indoor light-sensitive zoning database; the parameter groups include static design parameters and dynamic scenario parameters, and the indoor light-sensitive zoning database includes multiple data groups: Indoor lighting environment simulation is performed based on the aforementioned static design parameters and dynamic scenario parameters; In a simulated light environment, multiple judgment points are selected, and for each judgment point, the probability of sunlight glare and the illuminance of the working surface are calculated hourly. Based on the aforementioned probability of sunlight glare and illuminance of the working surface, sensitive and non-sensitive locations are determined, and a set of sensitive locations is constructed. Select a target sensitive point from the set of sensitive points, take the distance from the window to the target sensitive point as the photosensitive area depth of the current parameter group, and construct a data group with the photosensitive area depth and the corresponding static design parameters and dynamic scenario parameters. Model training module: This module uses the indoor light-sensitive zone database to train a machine learning model and obtain a dynamic boundary prediction model for the light-sensitive zone; Dynamic boundary prediction module: Based on the static design parameters and dynamic scenario parameters of the scene to be predicted collected in real time, this module uses the aforementioned dynamic boundary prediction model for the photosensitive area to output the dynamic boundary of the indoor photosensitive area.

[0011] Compared with existing technologies, this invention provides a real-time prediction method for the dynamic boundary of indoor light-sensitive areas. It utilizes orthogonal experiments based on key building factors and their corresponding levels, and selects the target working condition to be analyzed based on the working condition table constructed from the orthogonal experiments. Environmental scene simulation is then performed, systematically exploring the design parameter space with minimal simulation iterations. This avoids repetitive and brute-force simulations while quantitatively evaluating the impact of various key building factors on the depth of light-sensitive zones under different working conditions. Furthermore, this invention calculates hourly solar glare probability and work surface illuminance under simulated scenarios, constructing hourly training datasets for the corresponding working conditions based on the hourly calculation results. This enables the model to generalize and learn the instantaneous dynamic boundary of the light-sensitive area at each time interval under different working conditions. In actual inference, only static design parameters and corresponding dynamic scenario parameters need to be input to obtain the prediction result for the current scene, eliminating the need to remodel the scene to be predicted. This simplification greatly enhances the robustness and engineering applicability of the model. Attached Figure Description

[0012] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a flowchart of the model training process of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] Example 1 To address the problems existing in the prior art, this invention provides a real-time prediction method for the dynamic boundary of indoor light-sensitive areas. This method constructs a machine learning model trained based on orthogonal experimental results to solve the problems in the prior art. The process is as follows: Figure 1 As shown, it includes the following steps: S1. Obtain key building factors and their levels, conduct orthogonal experiments based on key building factors and their levels, classify working conditions according to the experimental results, and select target working conditions.

[0015] Specifically, in this embodiment, key building factors include typical window orientation, typical window-to-wall ratio, room height, room width, room depth, visible light transmittance of glass, and reflectance of walls, floor and ceiling; the level here refers to the value range or boundary of the corresponding key building factor. For example, for the window-to-wall ratio, it is set to include three levels, namely 0.3, 0.5 and 0.7.

[0016] In this step, a target type orthogonal table is selected from various existing orthogonal tables. In this embodiment, a 64-row orthogonal table is selected. The key building factors collected above and their corresponding levels are filled into the selected target type orthogonal table. According to the orthogonal table, it can be divided into 64 working conditions. From the obtained 64 working conditions, the working condition that needs to be simulated is selected as the target working condition. The design parameter space is systematically explored with the fewest number of simulations, so that the analysis results can transcend any single geometric model and obtain universal laws.

[0017] The method provided by this invention aims to achieve precise dynamic definition of light-sensitive areas inside buildings through systematic simulation and intelligent algorithms. This process requires first training the intelligent algorithm model, and the steps are as follows: Figure 2 As shown, this corresponds to steps S2 to S3.

[0018] S2. For the target working condition, obtain multiple parameter sets, and perform the following operations based on the multiple parameter sets to construct an indoor light-sensitive partition database.

[0019] In this step, for the target working condition, a random sampling strategy with space-filling characteristics is used to sample the parameter space, generating multiple parameter sets with multidimensional uniform distribution characteristics. Each parameter set includes static design parameters and dynamic scenario parameters. The static design parameters include key building factors and the angle between the line of sight and the window normal. The dynamic scenario parameters are variables that change with time or environment, including date, time, location, and the solar altitude and azimuth angles. The altitude and azimuth angles are obtained based on the date and time.

[0020] S21. Simulate the indoor lighting environment based on static design parameters and dynamic scenario parameters.

[0021] S22. In the simulated light environment, select multiple judgment points along the direction perpendicular to the window, and calculate the probability of sunlight glare and the illuminance of the working surface hourly for each judgment point.

[0022] S23. Determine sensitive and non-sensitive locations based on the probability of sunlight glare and the illuminance of the working surface, and construct a set of sensitive locations.

[0023] In this step, a visual discomfort threshold is first defined: the uncomfortable sunlight glare probability threshold is 0.35, and the uncomfortable work surface illuminance threshold is 2000 Lux. If the sunlight glare probability is greater than the uncomfortable sunlight glare probability threshold, or the work surface illuminance is greater than the uncomfortable work surface illuminance threshold, it is determined to be a sensitive location; otherwise, it is a non-sensitive location.

[0024] S24. Select the target sensitive point from the set of sensitive points, take the distance from the window to the target sensitive point as the photosensitive zone depth of the current parameter group, and construct a data group with the photosensitive zone depth and the corresponding static design parameters and dynamic scenario parameters. Multiple data groups are constructed into an indoor photosensitive zone database.

[0025] Furthermore, in this step, the target sensitive point is the sensitive point in the set of sensitive points that has the largest distance value from the window.

[0026] S3. Use the indoor light-sensitive zone database to train a machine learning model to obtain a dynamic boundary prediction model for the light-sensitive zone.

[0027] Machine learning is employed using an indoor light-sensitive zone database, which contains multiple datasets including light-sensitive zone depths determined based on visual comfort indices such as DGP (Distance Gauge Per Dimension) and work surface illuminance. Building upon this, a machine learning model (such as XGBoost or a neural network) is trained to directly learn and master the end-to-end mapping relationship from input parameters to the final light-sensitive zone depth; after training, a fixed model is obtained.

[0028] S4. Real-time acquisition of static design parameters and dynamic scenario parameters of the scene to be predicted, and output of the dynamic boundary of the indoor light-sensitive area using the dynamic boundary prediction model of the light-sensitive area.

[0029] Finally, once deployed, the dynamic boundary prediction model for the photosensitive zone can be put into practical use. Users only need to input the static design parameters of the current building and the real-time dynamic scenario parameters, and the model can instantly calculate and output an accurate prediction value, namely the depth of the photosensitive zone where control measures should be taken at the current moment or time period. This result can be directly used to guide building design or drive building automation systems, such as automatic curtains and intelligent lighting.

[0030] Example 2 To illustrate the practicality of this invention, this embodiment assumes that a smart shading control system is designed and deployed in a south-facing office.

[0031] In the initial design phase, architects need to input the static design parameters of the office and the project location into the dynamic boundary prediction model for light-sensitive zones provided by this invention. By changing different viewing angles in the dynamic boundary prediction model, the model outputs light-sensitive depths at different angles. Based on the predicted light-sensitive depths, conservative design suggestions for the office layout are obtained. That is, once all parameters are input, the dynamic boundary prediction model can output light-sensitive depth prediction results. These depth prediction results are obtained by analyzing year-round data to identify the optimal indoor lighting conditions. or illuminance The value is determined by the furthest distance that the effect can reach when the conditions occur.

[0032] The final design of the shading system is based on the forecast results. For example, to ensure visual comfort for most of the year, it is recommended that the shading system cover an area of ​​at least 3.5 meters from the window.

[0033] Example 3 To illustrate the practicality of this invention, in this embodiment it is assumed that the dynamic boundary prediction model of the photosensitive area provided by Benfubu is integrated into the building automation system (BAS) of an office in a certain city.

[0034] The data collected is for 2:00 PM on a certain afternoon. The BAS inputs the dynamic scene parameters of this office, along with the fixed static design parameters, into the dynamic boundary prediction model for the light-sensitive area. This model eliminates the need for complex and time-consuming light environment simulations; it can instantly predict the boundary between the light-sensitive and non-light-sensitive areas with a single, efficient calculation. If the model output indicates that, under the current sun position, the depth of the area satisfying the boundary conditions between the light-sensitive and non-sensitive areas is 3 meters, the BAS system automatically focuses on the DGP (Distance Gauge Gradient) and work surface illuminance of the personnel within this 3-meter depth range. If either indicator exceeds the comfort threshold, it immediately sends a command to the motorized curtains to adjust their position.

[0035] This approach transforms the simulation process, which previously required enormous computing power, into a lightweight algorithm call, enabling low-cost, refined, and intelligent environmental control. It ensures visual comfort, avoids excessive shading, and maximizes the use of natural light.

[0036] Example 4 This invention provides a real-time prediction system for dynamic boundaries of indoor light-sensitive areas, the system comprising: Data acquisition module: This module is used to acquire parameter sets and collect static design parameters and dynamic scenario parameters of the scenario to be predicted in real time.

[0037] Orthogonal test module: This module conducts orthogonal tests based on the acquired key building factors and their levels, and divides the work conditions according to the test results to select the target work condition.

[0038] Database Construction Module: This module performs the following operations on each of the multiple parameter groups obtained under the target operating conditions to construct an indoor light-sensitive zoning database; the parameter groups include static design parameters and dynamic scenario parameters, and the indoor light-sensitive zoning database includes multiple data groups: Indoor lighting environment simulation based on static design parameters and dynamic scenario parameters; In a simulated light environment, multiple judgment points are selected, and the probability of sunlight glare and the illuminance of the working surface are calculated hourly for each judgment point. Sensitive and non-sensitive locations are determined based on the probability of sunlight glare and the illuminance of the working surface, and a set of sensitive locations is constructed. Select the target sensitive point from the set of sensitive points, take the distance from the window to the target sensitive point as the photosensitive area depth of the current parameter group, and construct a data group with the photosensitive area depth and the corresponding static design parameters and dynamic scenario parameters.

[0039] Model training module: This module uses an indoor light-sensitive zone database to train a machine learning model and obtain a dynamic boundary prediction model for the light-sensitive zone.

[0040] Dynamic Boundary Prediction Module: Based on the static design parameters and dynamic scenario parameters of the scene to be predicted collected in real time, this module uses a dynamic boundary prediction model for the photosensitive area to output the dynamic boundary of the indoor photosensitive area.

[0041] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0042] Furthermore, the present invention provides an electronic device including a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). Various programs and data required for device operation can also be stored in the RAM. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0043] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0044] The processing unit executes the various methods and processes described above, such as methods S1 to S4. For example, in some embodiments, methods S1 to S4 may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of methods S1 to S4 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S4 by any other suitable means (e.g., by means of firmware).

[0045] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0046] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0047] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0048] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for real-time prediction of dynamic boundaries of indoor light-sensitive areas, characterized in that, The methods include: Key building factors and their levels are obtained, orthogonal experiments are conducted based on the key building factors and their levels, working conditions are divided according to the experimental results, and target working conditions are selected. For the target operating condition, multiple parameter sets are obtained, each of which includes static design parameters and dynamic scenario parameters. Based on these multiple parameter sets, the following operations are performed to construct an indoor photosensitive zoning database, which includes multiple data sets: Indoor lighting environment simulation is performed based on the aforementioned static design parameters and dynamic scenario parameters; In a simulated light environment, multiple judgment points are selected, and for each judgment point, the probability of sunlight glare and the illuminance of the working surface are calculated hourly. Based on the aforementioned probability of sunlight glare and illuminance of the working surface, sensitive and non-sensitive locations are determined, and a set of sensitive locations is constructed. Select a target sensitive point from the set of sensitive points, take the distance from the window to the target sensitive point as the photosensitive area depth of the current parameter group, and construct a data group with the photosensitive area depth and the corresponding static design parameters and dynamic scenario parameters. A machine learning model was trained using the aforementioned indoor photosensitive zone database to obtain a dynamic boundary prediction model for the photosensitive zone. The static design parameters and dynamic scenario parameters of the scene to be predicted are collected in real time, and the dynamic boundary of the indoor light-sensitive area is output using the aforementioned dynamic boundary prediction model of the light-sensitive area.

2. The method for real-time prediction of dynamic boundaries of indoor light-sensitive areas according to claim 1, characterized in that, The key building factors mentioned include typical window orientation, typical window-to-wall ratio, room height, room width, room depth, visible light transmittance of glass, and reflectivity of walls, floors and ceilings; The static design parameters include: key building factors and the angle between the line of sight and the window normal; The dynamic scenario parameters are variables that change with time or environment, including date, time, location, and the sun's altitude and azimuth angles; the altitude and azimuth angles are obtained based on the date and time.

3. The method for real-time prediction of dynamic boundaries of indoor light-sensitive areas according to claim 1, characterized in that, The method for selecting the judgment points is as follows: select a series of discrete judgment points along the direction perpendicular to the window.

4. The method for real-time prediction of dynamic boundaries of indoor light-sensitive areas according to claim 1, characterized in that, The conditions for determining the sensitive and non-sensitive locations are as follows: if the probability of sunlight glare is greater than the uncomfortable sunlight glare probability threshold, or the illuminance of the working surface is greater than the uncomfortable working surface illuminance threshold, then it is determined to be a sensitive location; otherwise, it is a non-sensitive location.

5. The method for real-time prediction of dynamic boundaries of indoor light-sensitive areas according to claim 1, characterized in that, The target sensitive point is the sensitive point with the largest distance value from the window in the set of sensitive points.

6. A real-time prediction system for dynamic boundaries of indoor light-sensitive areas, characterized in that, The system includes: Data acquisition module: This module is used to acquire parameter sets and collect static design parameters and dynamic scenario parameters of the scenario to be predicted in real time; Orthogonal test module: This module conducts orthogonal tests based on the acquired key building factors and their levels, and divides the work conditions according to the test results to select the target work condition; Database Construction Module: This module performs the following operations on each of the multiple parameter groups acquired under the target operating conditions to construct an indoor light-sensitive zoning database; the parameter groups include static design parameters and dynamic scenario parameters, and the indoor light-sensitive zoning database includes multiple data groups: Indoor lighting environment simulation is performed based on the aforementioned static design parameters and dynamic scenario parameters; In a simulated light environment, multiple judgment points are selected, and for each judgment point, the probability of sunlight glare and the illuminance of the working surface are calculated hourly. Based on the aforementioned probability of sunlight glare and illuminance of the working surface, sensitive and non-sensitive locations are determined, and a set of sensitive locations is constructed. Select a target sensitive point from the set of sensitive points, take the distance from the window to the target sensitive point as the photosensitive area depth of the current parameter group, and construct a data group with the photosensitive area depth and the corresponding static design parameters and dynamic scenario parameters. Model training module: This module uses the indoor light-sensitive zone database to train a machine learning model and obtain a dynamic boundary prediction model for the light-sensitive zone; Dynamic boundary prediction module: Based on the static design parameters and dynamic scenario parameters of the scene to be predicted collected in real time, this module uses the aforementioned dynamic boundary prediction model for the photosensitive area to output the dynamic boundary of the indoor photosensitive area.

7. The real-time prediction system for dynamic boundaries of indoor light-sensitive areas according to claim 6, characterized in that, The key building factors mentioned include typical window orientation, typical window-to-wall ratio, room height, room width, room depth, visible light transmittance of glass, and reflectivity of walls, floors and ceilings; The static design parameters include: key building factors and the angle between the line of sight and the window normal; The dynamic scenario parameters are variables that change with time or environment, including date, time, location, and the sun's altitude and azimuth angles; the altitude and azimuth angles are obtained based on the date and time.

8. The real-time prediction system for dynamic boundaries of indoor light-sensitive areas according to claim 6, characterized in that, The database construction module selects the judgment points by selecting a series of discrete judgment points along a direction perpendicular to the window.

9. The real-time prediction system for dynamic boundaries of indoor light-sensitive areas according to claim 6, characterized in that, The conditions for determining the sensitive and non-sensitive locations are as follows: if the probability of sunlight glare is greater than the uncomfortable sunlight glare probability threshold, or the illuminance of the working surface is greater than the uncomfortable working surface illuminance threshold, then it is determined to be a sensitive location; otherwise, it is a non-sensitive location.

10. A real-time prediction system for dynamic boundaries of indoor light-sensitive areas according to claim 6, characterized in that, The target sensitive point is the sensitive point with the largest distance value from the window in the set of sensitive points.

Images