CN-121997716-A - Real-time prediction method and system for dynamic boundary of indoor light sensitive area

CN121997716ACN 121997716 ACN121997716 ACN 121997716ACN-121997716-A

Abstract

The invention relates to a real-time prediction method and a real-time prediction system for a dynamic boundary of an indoor photosensitive area, wherein the method comprises the steps of obtaining key building factors and levels thereof, carrying out orthogonal experiments, carrying out working condition division according to experimental results, selecting target working conditions, obtaining a plurality of parameter sets for the target working conditions, constructing an indoor photosensitive partition database based on the plurality of parameter sets, training a machine learning model by utilizing the indoor photosensitive partition database to obtain a dynamic boundary prediction model of the photosensitive area, collecting static design parameters and dynamic scene parameters of a scene to be predicted in real time, and outputting the dynamic boundary of the indoor photosensitive area by utilizing the dynamic boundary prediction model of the photosensitive area. Compared with the prior art, the invention provides the method for identifying the photosensitive dynamic boundary of the building, which can overcome the geometric constraint of the traditional simulation method and has the advantages of universality, high efficiency and weather adaptation.

Inventors

ZHU HAN
DENG ZHIQING
LI ZHENGRONG

Assignees

同济大学

Dates

Publication Date: 20260508
Application Date: 20251229

Claims (10)

1. A real-time prediction method for dynamic boundary of indoor light sensitive area is characterized in that the method comprises the following steps: Acquiring a key building factor and a level thereof, performing an orthogonal test based on the key building factor and the level thereof, dividing working conditions according to test results, and selecting target working conditions; For a target working condition, acquiring a plurality of parameter sets, wherein each parameter set comprises a static design parameter and a dynamic scene parameter, and constructing an indoor photosensitive partition database based on the plurality of parameter sets, wherein the indoor photosensitive partition database comprises a plurality of data sets: performing indoor light environment simulation based on the static design parameters and the dynamic scene parameters; Selecting a plurality of judgment points in the simulated light environment, and calculating sunlight glare probability and working surface illuminance for each judgment point; Judging sensitive points and non-sensitive points based on the sunlight glare probability and the working surface illuminance, and constructing a sensitive point set; selecting a target sensitive point from the sensitive point set, taking the distance from a window to the target sensitive point as the depth of a photosensitive area of a current parameter set, and constructing the depth of the photosensitive area, a corresponding static design parameter and a corresponding dynamic scene parameter into a data set; training a machine learning model by using the indoor photosensitive partition database to obtain a photosensitive region dynamic boundary prediction model; And acquiring static design parameters and dynamic scene parameters of a scene to be predicted in real time, and outputting the dynamic boundary of the indoor light sensitive area by using the dynamic boundary prediction model of the light sensitive area.
2. The method of claim 1, wherein the key building factors include typical window orientation, typical window to wall ratio, room height, room width, room depth, glass visible light transmittance, and reflectivity of walls, floors and ceilings; the static design parameters comprise key building factors and angles between the line of sight and the normal line of the window; The dynamic scene parameters are variables which change with time or environment, and comprise date, time, place and altitude and azimuth angle of the sun, wherein the altitude and azimuth angle are obtained based on the date and the time.
3. The method for predicting dynamic boundary of indoor photosensitive area in real time according to claim 1, wherein the method for selecting the judgment point location is to select a series of discrete judgment point locations along the direction perpendicular to the window.
4. The method for predicting the dynamic boundary of the indoor light sensitive area in real time according to claim 1, wherein the condition for judging the sensitive point location and the non-sensitive point location is that if the sunlight glare probability is larger than the uncomfortable sunlight glare probability threshold value or the working surface illuminance is larger than the uncomfortable working surface illuminance threshold value, the sensitive point location is judged, and otherwise, the non-sensitive point location is judged.
5. The method for predicting the dynamic boundary of the indoor light sensitive area in real time according to claim 1, wherein the target sensitive points are sensitive points with the largest distance value from a window in the sensitive point set.
6. A real-time prediction system for dynamic boundaries of an indoor light-sensitive area, the system comprising: the data acquisition module is used for acquiring a parameter set and acquiring static design parameters and dynamic scene parameters of a scene to be predicted in real time; The orthogonal test module is used for carrying out orthogonal test according to the acquired key building factors and the level thereof, carrying out working condition division according to test results and selecting target working conditions; The database construction module is used for constructing an indoor photosensitive partition database according to the obtained parameter sets under the target working condition, wherein the parameter sets comprise static design parameters and dynamic scene parameters, and the indoor photosensitive partition database comprises a plurality of data sets: performing indoor light environment simulation based on the static design parameters and the dynamic scene parameters; Selecting a plurality of judgment points in the simulated light environment, and calculating sunlight glare probability and working surface illuminance for each judgment point; Judging sensitive points and non-sensitive points based on the sunlight glare probability and the working surface illuminance, and constructing a sensitive point set; selecting a target sensitive point from the sensitive point set, taking the distance from a window to the target sensitive point as the depth of a photosensitive area of a current parameter set, and constructing the depth of the photosensitive area, a corresponding static design parameter and a corresponding dynamic scene parameter into a data set; the model training module is used for training a machine learning model by utilizing the indoor photosensitive partition database to obtain a photosensitive region dynamic boundary prediction model; And the dynamic boundary prediction module is used for outputting the dynamic boundary of the indoor light sensitive area by utilizing the dynamic boundary prediction model of the light sensitive area based on the static design parameters and the dynamic scene parameters of the scene to be predicted, which are acquired in real time.
7. The system of claim 6, wherein the key building factors include typical window orientation, typical wall to window ratio, room height, room width, room depth, glass visible light transmittance, and reflectivity of walls, floors and ceilings; the static design parameters comprise key building factors and angles between the line of sight and the normal line of the window; The dynamic scene parameters are variables which change with time or environment, and comprise date, time, place and altitude and azimuth angle of the sun, wherein the altitude and azimuth angle are obtained based on the date and the time.
8. The system for predicting dynamic boundary of indoor light sensitive area in real time as set forth in claim 6, wherein said database construction module selects said judgment points by selecting a series of discrete judgment points along a direction perpendicular to the window.
9. The system for predicting dynamic boundary of indoor light sensitive area in real time as set forth in claim 6, wherein the condition for determining said sensitive point location and said non-sensitive point location is that if said daylight glare probability is greater than said uncomfortable daylight glare probability threshold or said working surface illuminance is greater than said uncomfortable working surface illuminance threshold, then determining said sensitive point location, otherwise, determining said non-sensitive point location.
10. The system for predicting dynamic boundary of indoor light sensitive area in real time as set forth in claim 6, wherein said target sensitive points are sensitive points with the largest distance value from window in said sensitive point set.

Description

Real-time prediction method and system for dynamic boundary of indoor light sensitive area Technical Field The invention relates to the technical field of energy conservation and environmental protection, in particular to a method and a system for predicting dynamic boundaries of indoor photosensitive areas in real time. Background Natural light is fully utilized as the core technical direction of modern green building and energy-saving design, so that the energy consumption of an artificial lighting system can be remarkably reduced, the low-carbon development concept is met, the indoor light environment quality can be effectively improved, the living and working comfort of personnel is improved, the working efficiency is further optimized, and the energy-saving and environment-friendly type energy-saving artificial lighting system has important energy-saving and environment-friendly value and engineering application significance. However, in the practical process of promoting the full utilization of sunlight illumination, a key technical problem is faced, namely, how to accurately avoid the negative effects of glare interference, indoor excessive heat accumulation and the like caused by direct sunlight while efficiently introducing beneficial natural light. The technical problem causes that two types of functional differentiation areas are naturally formed in the indoor space, namely an indoor depth area capable of directly utilizing soft diffuse light and a window leaning photosensitive area needing to conduct targeted control on direct sunlight. Because the range of the photosensitive area can dynamically change along with factors such as the sun altitude angle, azimuth angle, weather conditions and the like, how to realize real-time and accurate definition of the dynamic photosensitive area is an urgent problem to be solved, and scientific basis can be provided for targeted cooperative regulation and control of a sunshade system, a lighting regulation and control system and the like only through accurate partition, natural light resources beneficial to indoor deep areas are reserved to the maximum extent on the basis of effectively inhibiting the negative influence of direct sunlight, and finally the optimal balance of building energy-saving benefit and indoor human living comfort is achieved. In the prior art, researches on indoor illumination are mostly focused on researches on indoor light comfort, such as a Chinese patent application CN110334387A, which provides an indoor illumination prediction method based on a BP neural network algorithm, wherein indoor illumination is predicted through the BP neural network algorithm, and an LED luminous flux and luminous flux transfer function matrix model are combined, so that efficient utilization of power resources and improvement of illumination uniformity under different seasons are realized, but identification of dynamic boundaries of a photosensitive region is still not realized. Although some methods can realize the determination of the illumination regulation area and the illumination uncomfortable area, the common flow of the methods is generally to set up a building computer model with fixed geometric dimensions and specific materials, for example, using Sketchup, revit and other modeling, using professional simulation software (such as Radiance, daysim) and combining annual meteorological data (TMY file) of the specific area to perform time-by-time or minute light environment simulation, analyzing indexes such as vertical eye illuminance (Ev), glare probability (DGP), illuminance (Illuminance) and the like of each point in the room according to the simulation result, and finally performing manual determination to generate a result. The method can realize region identification, but the simulation result is seriously dependent on the fixed geometric parameters which are originally set, once any parameter such as the orientation of a building, the window wall ratio, the room depth and the like is changed, the method needs to be modeled again and takes a great amount of time to simulate again, so that the conclusion obtained by the method does not have universality, cannot form a set of general engineering design criteria or scales which can guide different projects, and is difficult to integrate in a building automatic control system which needs quick response. Therefore, the method for identifying the photosensitive dynamic boundary of the building, which can overcome the geometric constraint of the traditional simulation method and has the advantages of universality, high efficiency and self-adaption to the climate, is a technical problem to be solved. Disclosure of Invention The invention aims to overcome the defects of the prior art and provide a real-time prediction method and a real-time prediction system for the dynamic boundary of an indoor photosensitive area, which aim to realize accurate dynamic definition of the indoor photosensitive area o