A method, system and device for arranging light environment in a closed working space

Abstract

The present invention relates to the technical field of light environment layout of confined working space, and specifically to a method, system and equipment for light environment layout of confined working space. The present invention combines ergonomics theory with a learning algorithm to construct a multi-dimensional nonlinear prediction model, systematically analyzes the complex relationship between light environment layout and comprehensive efficiency, and ensures the best match between light environment layout and comprehensive efficiency in the confined working space. It can not only accurately predict the comprehensive efficiency of the existing light environment layout, but also obtain the optimal solution of the light environment layout of the confined working space through a deep neural network, providing a scientific and reliable basis for the design under different application scenarios, thereby significantly improving the comprehensive efficiency of the confined working space and meeting the actual light environment layout requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of light environment layout of a closed working space, and in particular to a light environment layout method, system and equipment for a closed working space. Background Art

[0002] With the rapid development of underwater space and manned space capsules, the design and research of confined space has become extremely important. Especially with the increase in aerospace and navigation missions and the exploration of extreme environments such as the deep sea and the polar regions, human work and life in confined spaces have become increasingly frequent and important; working in such a confined environment for a long time may cause problems such as reduced comfort, increased fatigue, and distraction. In severe cases, major accidents may even be caused by human errors. In aerospace and navigation missions, any mistakes may have irreversible consequences. Reasonable confined space layout design can not only improve work efficiency and comfort, but also significantly improve safety and reduce the accident rate, thereby increasing the success rate of missions and ensuring personnel safety.

[0003] As an important part of the space environment, the light environment not only directly affects physiological rhythms and emotional regulation, but also affects visual comfort and task efficiency through lighting intensity and light source distribution. Although the lighting intensity, light source distribution and color rendering index are adjusted to simulate the sunlight environment, a more natural working environment is provided for personnel in confined spaces. However, this method relies too much on experience and is difficult to accurately deal with the complex coupling relationship between multi-dimensional light environment parameters, especially in the extreme environment of aerospace and navigation, and cannot fully meet the actual work needs. Moreover, the impact of the light environment on the comprehensive performance of personnel is not considered, and there is a lack of systematic model support and scientific verification, resulting in the layout design being not ideal in improving the comprehensive performance. Therefore, it cannot meet the light environment layout requirements of confined working spaces. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system and equipment for the layout of the light environment in a confined working space. By combining ergonomics theory with a learning algorithm, a multi-dimensional nonlinear prediction model is constructed to solve the technical problem that the current light environment layout in a confined working space has a poor effect in improving the overall efficiency.

[0005] The present invention solves the above-mentioned technical problems by:

[0006] A method for arranging a light environment in a confined working space comprises the following steps:

[0007] S1. Determine the array element variables that affect the light environment layout of the confined working space;

[0008] S2. Analyze the influence of array element variables on comprehensive performance in the light environment layout, and extract the normalized influence factor of array element variables;

[0009] S3. Establishing an array element variable influence factor model based on the normalized influence factors of the array element variables;

[0010] S4, randomly generating a variety of light environment layouts, collecting parameters of each light environment layout and corresponding comprehensive performance, obtaining an initial data set, and establishing a multi-dimensional comprehensive performance evaluation model based on the initial data set;

[0011] S5. Based on the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model, a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance is constructed;

[0012] S6. Based on the multi-dimensional comprehensive performance evaluation model, use deep neural networks to optimize the light environment layout to maximize the comprehensive performance, determine the best matching solution between the light environment layout and the comprehensive performance, and verify the multi-dimensional nonlinear prediction model.

[0013] It is further defined that the step S1 is specifically as follows:

[0014] The array element variables that affect the light environment layout of the confined working space are determined through experimental measurement, theoretical derivation and scene simulation analysis;

[0015] The array element variables include the number of light sources, the height of the light source, the color temperature of the light source, the illumination of the light source, the angle of the light source and the position of the light source.

[0016] It is further defined that step S2 comprises:

[0017] S2.1. Build a closed working space and create an adjustable light environment layout;

[0018] S2.2, adjusting the parameters of a single variable in the array element variables to obtain a variety of single variable light environment layouts, obtaining the degree of influence of different single variable light environment layouts on the subjects, obtaining a comprehensive efficacy score, and constructing a model of the comprehensive efficacy score and the single variable light environment layout;

[0019] S2.3. The comprehensive performance score is fitted with the model of the single variable light environment layout by the least squares method to obtain the normalized influence factor of the array element variable.

[0020] It is further defined that step S3 comprises:

[0021] S3.1, converting the Cartesian coordinates of the light source position into polar coordinates;

[0022] S3.2. Construct an array element variable influence factor model based on the normalized influence factor of the array element variable and the polar coordinates of the light source position.

[0023] It is further defined that the parameters of the comprehensive efficacy include physiological efficacy parameters, psychological efficacy parameters and cognitive efficacy parameters;

[0024] The step S4 comprises:

[0025] S4.1, by changing the array element variables, a variety of light environment layouts are randomly obtained;

[0026] S4.2. Obtaining the physiological performance parameters of the subjects under different light environment layouts;

[0027] S4.3, obtaining the psychological efficacy parameters of the subjects under different light environment layouts;

[0028] S4.4. Obtain the cognitive performance parameters of the subjects under different light environment layouts;

[0029] S4.5. Obtain parameters of different light environment layouts and corresponding comprehensive performance to obtain an initial data set, and establish a multi-dimensional comprehensive performance evaluation model based on the initial data set.

[0030] It is further defined that step S5 comprises:

[0031] S5.1. Randomly generate light environment layouts of multiple light environments, and divide the obtained light environment layouts into a training set and a test set;

[0032] S5.2. Based on the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model, a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance is constructed. The obtained multi-dimensional non-linear prediction model is optimized through training sets and test sets to improve its prediction accuracy.

[0033] It is further defined that step S6 comprises:

[0034] S6.1, using the light environment layout obtained in step S5.1, obtaining an experimental data set of light environment layout and corresponding comprehensive efficiency evaluation value according to a multi-dimensional comprehensive efficiency evaluation model, and constructing an efficiency prediction model through a deep neural network based on the experimental data set;

[0035] S6.2. Use genetic algorithms to solve the optimal efficiency value and optimal layout parameters of the efficiency prediction model, and configure the light environment of the enclosed working space according to the optimal layout parameters;

[0036] S6.3. Input the optimal layout parameters into the multidimensional nonlinear prediction model obtained in step S5.2 to obtain the optimal performance prediction value, and determine whether the optimal performance prediction value is equal to the optimal performance prediction value. If so, end; if not, re-optimize the obtained multidimensional nonlinear prediction model through the training set and the test set and execute step S6.3.

[0037] A closed work space light environment layout system, used to implement the closed work space light environment layout method, comprising:

[0038] An array element variable determination unit, used to determine array element variables that affect the light environment layout of a confined working space;

[0039] An array element variable normalized influence factor extraction unit is used to analyze the influence of array element variables on the comprehensive performance in the light environment layout and extract the normalized influence factor of the array element variables;

[0040] An array element variable influence factor model building unit is used to build an array element variable influence factor model based on the normalized influence factor of the array element variable;

[0041] A multi-dimensional comprehensive performance evaluation model establishment unit is used to randomly generate a variety of light environment layouts, collect parameters of each light environment layout and corresponding comprehensive performance, obtain an initial data set, and establish a multi-dimensional comprehensive performance evaluation model based on the initial data set;

[0042] A multi-dimensional nonlinear prediction model building unit is used to build a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance according to the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model;

[0043] The best matching scheme determination unit is used to optimize the light environment layout to maximize the comprehensive efficiency based on the multi-dimensional comprehensive efficiency evaluation model and utilize the deep neural network to determine the best matching scheme between the light environment layout and the comprehensive efficiency.

[0044] It is further defined that the comprehensive efficacy parameters include physiological efficacy parameters, psychological efficacy parameters and cognitive efficacy parameters; and the multi-dimensional comprehensive efficacy evaluation model establishment unit includes:

[0045] A light environment layout creation module is used to randomly obtain a variety of light environment layouts by changing array element variables;

[0046] A physiological performance parameter acquisition module is used to obtain the physiological performance parameters of the subjects under different light environment layouts;

[0047] A psychological efficacy parameter acquisition module is used to obtain the psychological efficacy parameters of the subjects under different light environment layouts;

[0048] A cognitive performance parameter acquisition module is used to obtain the cognitive performance parameters of the subjects under different light environment layouts;

[0049] The multi-source data fusion processing module is used to obtain the parameters of different light environment layouts and corresponding comprehensive performance to obtain the initial data set, and establish a multi-dimensional comprehensive performance evaluation model based on the initial data set;

[0050] The enclosed working space light environment layout system also includes:

[0051] A confined work space simulation unit is used to simulate a confined work space and also to simulate the light environment layout;

[0052] The layout scheme display unit is used to display the enclosed working space, the light environment layout, the comprehensive efficiency of the corresponding light environment layout, and the parameters of the corresponding light environment layout.

[0053] A device for arranging a light environment in a closed working space comprises a non-volatile storage medium and a central processing unit. The non-volatile storage medium stores executable code. When the central processing unit executes the executable code, the above-mentioned method for arranging a light environment in a closed working space is implemented.

[0054] The beneficial effects of the present invention are:

[0055] 1. The present invention combines ergonomics theory with learning algorithms to construct a multi-dimensional nonlinear prediction model, systematically analyzes the complex relationship between light environment layout and comprehensive efficiency, and ensures the best match between light environment layout and comprehensive efficiency in a confined working space. It can not only accurately predict the comprehensive efficiency of the existing light environment layout, but also use deep neural networks to obtain the optimal solution for the light environment layout of the confined working space, providing a scientific and reliable basis for the design in different application scenarios, thereby significantly improving the comprehensive efficiency of the confined working space and meeting the actual light environment layout requirements.

[0056] 2. The present invention utilizes a multi-dimensional nonlinear prediction model to accurately predict the specific impact of different light environment layouts on personnel's physiological efficiency, psychological efficiency and cognitive efficiency, providing a reliable scientific basis for layout design; breaking through the limitations of traditional empirical design, using deep neural networks to solve the optimal light environment layout plan, achieving the best match between layout parameters and comprehensive efficiency, and significantly improving the accuracy and reliability of the design. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a step diagram of a method for arranging a light environment for a confined work space according to Embodiment 1 of the present invention;

[0058] Figure 2 This is a schematic diagram of the light environment layout structure of a closed working space according to Example 1 of the present invention;

[0059] Figure 3 This is a schematic diagram of the light environment layout system for a closed working space according to Embodiment 2 of the present invention;

[0060] Figure 4 This is a schematic diagram of the light environment layout equipment for a closed working space according to Example 3 of the present invention. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] refer to Figure 1 The present invention provides a method for arranging a light environment in a closed working space, comprising the following steps:

[0063] S1. Determine the array element variables that affect the light environment layout of the confined working space;

[0064] S2. Analyze the influence of array element variables on comprehensive performance in the light environment layout, and extract the normalized influence factor of array element variables;

[0065] S3. Establishing an array element variable influence factor model based on the normalized influence factors of the array element variables;

[0066] S4, randomly generating a variety of light environment layouts, collecting parameters of each light environment layout and corresponding comprehensive performance, obtaining an initial data set, and establishing a multi-dimensional comprehensive performance evaluation model based on the initial data set;

[0067] S5. Based on the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model, a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance is constructed;

[0068] S6. Based on the multi-dimensional comprehensive performance evaluation model, use deep neural networks to optimize the light environment layout to maximize the comprehensive performance, determine the best matching solution between the light environment layout and the comprehensive performance, and verify the multi-dimensional nonlinear prediction model.

[0069] Example 1

[0070] This embodiment provides a method for arranging a light environment in a confined work space, wherein step S1 specifically comprises:

[0071] Specifically, it is preferred to determine the array element variables of the light environment layout of the enclosed working space by means of laboratory measurement, theoretical deduction and scene simulation.

[0072] Through laboratory measurements:

[0073] First, a closed laboratory was built to ensure that it was highly airtight and opaque. Silver-gray opaque curtains were used to protect the interior of the closed laboratory to ensure consistency between experimental groups and to reduce the potential impact of environmental variables on the operator's comfort in the lighting environment.

[0074] Specifically, ensure that the temperature, humidity, noise and wind speed of the closed laboratory environment drop to the set values, such as the temperature is 26°C, the humidity is 65%, the environmental noise is 45db, and the wind speed is 0.5m / s. The experimental area is equipped with a HVAC system to ensure that participants can complete the task normally without causing physical discomfort.

[0075] refer to Figure 2 , 16 LED (Light Emitting Diode) lamps can be optionally installed on the top of the closed laboratory to form a light source, and the 16 LED lamps are distributed in a 4×4 array.

[0076] A spectrometer was used to measure the illumination distribution, illumination angle and diffusion characteristics of the light source, and the illumination value and diffusion radius were recorded when the light source height was 2.5 meters and the illumination angle was 45°. The illumination value and diffusion radius were then recorded when the light source height was 2 meters and when the light source height was 1.8 meters. It was obtained that the brightness and uniformity changed under the arrangement conditions of different light source heights and different light source angles.

[0077] Through theoretical derivation:

[0078] Based on the CIE (Commission Internationale de I'Eclairage, International Commission on Illumination) standard observer model, the optimal color temperature range of the lighting source is calculated. Combined with the human visual fatigue model, the appropriate brightness for tasks lasting more than 1 hour is derived. At the same time, the optical reflection formula is used to calculate the reflected light intensity of the light source at a specific angle to ensure that the balance point between visual clarity and comfort is met, thereby obtaining changes in visual clarity and comfort under different light source color temperature arrangements.

[0079] Through scenario simulation:

[0080] Using lighting design software, a three-dimensional geometric model of a confined working space is established, which is 2 meters long, 2 meters wide and 2.5 meters high. The optical parameters of the light source are imported to simulate the lighting coverage under the conditions of different light source quantity arrangements, different light source positions and different light source illumination conditions, and different lighting uniformity indicators can be obtained.

[0081] Therefore, the array element variables that affect the light environment layout of the confined working space include six single variables: the number of light sources, light source height, light source color temperature, light source illumination, light source angle, and light source position.

[0082] Further explanation, step S2 includes the following steps:

[0083] S2.1. Build a closed working space and create an adjustable light environment layout;

[0084] Specifically, a confined working space is built so that the height adjustment range of the light source in the confined working space is 1.5 meters to 2.5 meters, and the adjustment interval is 0.5 meters; the adjustment range of the light source color temperature is 3000K to 7000K, and the adjustment interval is 1000K; the illumination range of the light source is 300lx to 900lx, and the adjustment interval is 200lx; the adjustment range of the light source angle is 0° to 45°, and the adjustment interval is 15°; the adjustment of the light source position is to randomly light up the set number of light sources in the 4×4 array layout, and the number of light sources is illustrated by taking 3 as an example, to complete the creation of the adjustable lighting environment layout of the confined working space.

[0085] S2.2, adjusting the parameters of a single variable in the array element variables to obtain a variety of single variable light environment layouts, obtaining the degree of influence of different single variable light environment layouts on the subjects, obtaining a comprehensive efficacy score, and constructing a model of the comprehensive efficacy score and the single variable light environment layout;

[0086] Specifically, in order to extract the influencing factors of a single variable on the comprehensive performance, the adjustable light environment layout created in step S2.1 is used. Each light environment layout only adjusts the parameters of one of the single variables, and the other single variables remain consistent, thereby obtaining a single variable light environment layout under multiple single variable parameter adjustments; for example, only the light source height is adjusted, and the other variables are all initial values, and three light environment layouts with light source heights of 1.5 meters to 2.5 meters and adjustment intervals of 0.5 meters can be obtained; next, only the light source color temperature is adjusted, and the other variables are all initial values, and five light environment layouts with light source color temperature adjustment ranges of 3000K to 7000K and adjustment intervals of 1000K are obtained again.

[0087] By adjusting the parameters of a single variable, that is, adjusting the number of light sources, light source height, light source color temperature, light source illumination, light source angle or light source position, a new single variable light environment layout will be formed. Different single variable light environment layouts will have a significant effect on people's physiology, psychology and cognition; specifically, the light source height and light source angle affect the lighting coverage of key areas, the light source color temperature and light source illumination affect emotions and fatigue, and the number of light sources and light source position are directly related to the uniformity of lighting and spatial perception ability.

[0088] Therefore, the subjects experienced different light environment layouts and made judgments on whether there was any impact.

[0089] The Likert scale was used to evaluate the impact of the light environment layout on the subjects under the adjustment of a single variable parameter. Specifically, the corresponding comprehensive effectiveness was quantified by collecting the subjects' comprehensive effectiveness scores for each of the above light environment layouts, where the score ranged from 1 to 5, with 1 representing "very little impact" and 5 representing "very much impact."

[0090] The questionnaire contains three modules, corresponding to the physiological dimension, psychological dimension and cognitive dimension. The physiological dimension includes indicators such as fatigue and discomfort perceived by the subjects; the psychological efficacy includes indicators such as negative emotions and stress perception perceived by the subjects; the cognitive dimension includes indicators such as the subjects' inattention; the score of each module is calculated by the mean to represent the efficacy level of that dimension; the comprehensive efficacy score corresponding to the light environment layout is calculated by the mean of the scores of each module.

[0091] Subsequently, multivariate regression analysis was used to correlate the comprehensive efficacy score with the changes in a single variable under different parameter settings, and a model of the comprehensive efficacy score and the single variable light environment layout was constructed to reflect the score relationship between the single variable light environment layout and the comprehensive efficacy.

[0092]

[0093] in, is the intercept, For the The regression coefficients of the variables, For the overall performance rating, For the The parameter value of the variable, is a random error, is the number of variables.

[0094] For the sake of convenience, the first single variable in the example is the number of light sources, the second single variable is the height of the light source, the third single variable is the color temperature of the light source, the fourth single variable is the illumination of the light source, the fifth single variable is the angle of the light source, and the sixth single variable is the position of the light source. .

[0095] S2.3. The comprehensive performance score is fitted with the model of the single variable light environment layout by the least squares method to obtain the normalized influence factor of the array element variable:

[0096]

[0097] in, .

[0098] Specifically, the validation shows that its fit is Reaching 0.85 can accurately reflect the contribution of light environment layout variables to the overall performance.

[0099] To further illustrate, step S3 includes the following steps:

[0100] S3.1, converting the Cartesian coordinates of the light source position into polar coordinates;

[0101] Specifically, the number of light sources is 16 as an example. According to the size limit of the confined working space, the area where the light source can be arranged is clarified. Based on the plane coordinate system, 16 light source arrangement points are divided. The plane coordinates of the light source are expressed in a matrix as follows:

[0102]

[0103] Each matrix element represents a light source arrangement point Coordinates; row representation Coordinates (changes along the height direction), columns represent Coordinates (changes along the width direction); coordinate origin Defined as the lower left corner of the confined work space.

[0104] The light source point is arranged at the top, and its Cartesian coordinates are ( ), the Cartesian coordinate matrix of the light source point layout array is as follows:

[0105]

[0106] In order to facilitate the analysis, it is preferred to convert the Cartesian coordinates of the light source position into polar coordinates. At this time, the origin of the polar coordinates is defined as the exact center of the desktop close to the wall on one side of the confined workspace. The desktop is centered relative to the wall on this side. The desktop height is 0.8m, the width is 0.5m, and the length is 1m. The position of the polar coordinate origin in the Cartesian coordinate system is ( ) = ( ).

[0107] Next, the Cartesian coordinates of each light source ( ) to polar coordinates ( ),in, is the radial distance, which indicates the distance from the light source to the origin of the polar coordinates; is the pitch angle, which represents the angle between the line connecting the light source and the origin of the polar coordinates and the vertical direction; is the azimuth, which indicates the azimuth of the light source on the horizontal plane. Axis as reference:

[0108]

[0109]

[0110]

[0111] Finally, the polar coordinate matrix of the light source transformation is expressed as:

[0112]

[0113] S3.2, constructing an array element variable influence factor model according to the normalized influence factor of the array element variable and the polar coordinates of the light source position;

[0114] Specifically, the impact of the light environment layout on the comprehensive efficiency of operators includes the number of light sources. , Light source height , Light source color temperature , light source illumination , light source angle And the light source position , and the influence of each variable is determined by its own parameters, so the array element variable influencing factor model is established:

[0115]

[0116] Where R is the number of light sources, is the polar coordinate of the rth light source.

[0117] Further explanation, step S4 includes the following steps:

[0118] S4.1, by changing the array element variables, a variety of light environment layouts are randomly obtained;

[0119] Specifically, by changing the parameters of any variable, a variety of different light environment layouts are generated in the closed working space, and the number of light environment layouts is set to W; the adjustment range and adjustment interval of each variable shown in step S2.1 are used as an example for explanation.

[0120] S4.2. Obtaining the physiological performance parameters of the subjects under different light environment layouts;

[0121] Specifically, the parameters of comprehensive effectiveness include physiological effectiveness parameters, psychological effectiveness parameters and cognitive effectiveness parameters.

[0122] Physiological performance is monitored mainly through the subjects' heart rate HR (heartrate), heart rate variability RMSSD (root mean square of successive differences), skin conductivity SCL (Skin Conductance Level) and skin conductivity response SCR (Skin Conductance Response) indicators.

[0123] Therefore, in actual operation, the HR and other parameters of the subject can be collected through a wearable heart rate sensor; among them, the wearable heart rate sensor consists of a heart rate belt and a built-in chip. The built-in chip needs to be firmly clamped on the heart rate belt and worn on the chest close to the heart. The length of the heart rate belt should be adjusted as needed to avoid weakening of the data signal. The electrode area of ​​the heart rate belt should be moistened with water or electrode gel. The wearable heart rate sensor will automatically turn on after detecting the electrical signal.

[0124] Parameters such as SCL are collected through a skin conductivity meter, which can be selected as a medical-grade wearable device.

[0125] Based on statistical analysis methods, the expanded physiological index parameter values ​​can be obtained:

[0126] The parameter value of HR is the average heart rate in 1 minute, which is calculated by the following formula:

[0127]

[0128] in, It represents the average heart rate in 1 minute. Indicates the heart rate value recorded for the tth time within 1 minute. Indicates the number of heart rate records within 1 minute. N Indicates the number of heart rate records within 1 minute.

[0129] RMSSD is the root mean square of the difference between the RR intervals of adjacent ventricular beats. It mainly reflects the amplitude of short-term fluctuations in heart rate. It is usually regulated by the vagus nerve (parasympathetic nerve) and is often used for short-term HRV (Heart Rate Variability) assessment. It is very sensitive to the activity of the parasympathetic nerve and belongs to the field of view analysis method of HRV. The calculation formula is as follows:

[0130]

[0131] in, Indicates The time interval from one heartbeat to the next heartbeat (in milliseconds), Indicates The time interval from one heartbeat to the next heartbeat, S represents the number of time intervals between heartbeats.

[0132] SCL is a physiological indicator that reflects the activity of skin sweat glands. It is a parameter value suitable for evaluating emotions, stress, and arousal levels. The calculation formula is as follows:

[0133]

[0134] in Represents the average value of skin conductance level within 1 minute. Indicates the first p The skin conductance level values ​​recorded, P It is the number of times the skin conductance level is recorded within 1 minute.

[0135] The data collected by the skin conductance meter is observed to see if there are "spikes" or sudden increases in the skin conductance signal, which usually represent SCR.

[0136] Analyze the occurrence time and intensity of the "spike" to evaluate the intensity of the individual's physiological response: compare the baseline data before the test and the changes during the test, calculate the SCR reaction amplitude, reaction time and recovery time, where the reaction time is the delay time from the start of the test to the change of skin conductance, and the recovery time is the length of time it takes to recover from the peak to the baseline; at the same time, count the duration of the skin electrical response and the frequency of reaction per minute; during the test, record and analyze at a frequency of once per minute.

[0137] S4.3, obtaining the psychological efficacy parameters of the subjects under different light environment layouts;

[0138] Specifically, the emotional scale PANAS (Positive and Negative Affect Schedule) was used to test and evaluate the psychological comfort and emotional state of the subjects in simulation scenarios and experimental scenarios respectively. Different light environment layouts have different effects on spatial perception. For example, the sense of openness and comfort of the space, as well as the uniformity of light, will all have an impact on psychological comfort.

[0139] PANAS is a standardized tool for measuring an individual's emotional state. It consists of two subscales: positive emotion PA (Positive Affect) and negative emotion NA (Negative Affect). Each subscale contains 10 items. PA includes positive emotions such as happiness, excitement, and self-confidence, corresponding to 10 positive emotion factors. NA includes negative emotions such as tension, fatigue, and depression, corresponding to 10 negative emotion factors.

[0140] Before the subjects experienced the simulation scene and the experimental scene respectively, they were required to complete the initial assessment of the emotion scale and record their baseline emotional state; when the subjects completed the experience in each simulated light environment of the simulation scene and in each light environment layout environment of the experimental scene, they filled out the emotion scale again to evaluate the impact of different light environment layouts on their emotional state.

[0141] Each emotional state score ranges from 1 to 5, with 1 being completely absent and 5 being very obvious. The cumulative score is used to calculate the PA and NA indexes, and the difference between PA and NA is used as a direct indicator of psychological comfort:

[0142]

[0143] in, Indicates psychological comfort, Indicates a The score of the positive emotion factor, is the total number of positive emotion factors, Indicates The score of the negative emotion factor, is the total number of negative emotion factors.

[0144] The above formula can be used to calculate the psychological comfort of the subjects after experiencing different light environment layouts of the simulation scene. The psychological comfort of the subjects after experiencing different light environment layouts in the experimental scene .

[0145] To obtain a comprehensive psychological comfort score, the psychological comfort scores of the simulation scenario and the experimental scenario were combined and the following formula was used:

[0146]

[0147] in, Indicates the overall psychological comfort. Indicates the psychological comfort of the simulation scene, Indicates the psychological comfort of the experimental scene, is the weight coefficient of psychological comfort of the simulation scene, is the weight coefficient of psychological comfort of the experimental scene, The value range and The value range is 0.1~0.9 and can be set according to specific needs, reflecting the relative importance of simulation data and experimental data.

[0148] For example, in order to improve the efficiency of test evaluation, it is mainly carried out through simulation scenarios. At this time, the importance of simulation data is high, and the psychological comfort weight coefficient of simulation scenarios is greater than that of experimental scenarios. On the contrary, in order to ensure the real experience of the subjects, test evaluation is mainly carried out by building experimental scenarios. At this time, the importance of experimental data is high, and the psychological comfort weight coefficient of experimental scenarios is greater than that of simulation scenarios.

[0149] Before the subjects experienced the simulation scene and the experimental scene respectively, they were required to complete the initial assessment of the emotion scale and record their baseline emotional state; when the subjects completed the experience in each simulated light environment of the simulation scene and in each light environment layout environment of the experimental scene, they filled out the emotion scale again to evaluate the impact of different light environment layouts on their emotional state.

[0150] After completing the simulation and experimental scene experiences, the subjects obtained psychological comfort scores for different light environment layouts based on their recorded baseline emotional states:

[0151]

[0152] in, To score the psychological comfort of the subjects before experiencing the simulation scene and the experimental scene, Score the psychological comfort of the subjects after experiencing the simulation scenario and the experimental scenario.

[0153] S4.4. Obtain the cognitive performance parameters of the subjects under different light environment layouts;

[0154] Specifically, cognitive behavioral data were obtained by conducting cognitive stimulation simulation experiments, focusing on evaluating changes in two indicators: reaction time and task completion accuracy.

[0155] The cognitive stimulation simulation experiment includes:

[0156] Target recognition requires the subject to quickly identify a designated target from a series of displayed targets within a specified time. The purpose is to test whether the subject can concentrate. The target can be an image, symbol or letter, and the designated target appears at a different time in each experiment.

[0157] The subjects’ accuracy and reaction time in the target recognition task were then obtained.

[0158] Color matching requires the subject to quickly select the same color as the given color template from a variety of colors; the purpose is to test the subject's ability to remember colors, especially to remember and recognize the target color in a short period of time.

[0159] The subjects' accuracy and reaction time in the color matching task were then obtained.

[0160] Graphic tracking requires the subject to follow a dynamically changing graphic, keep their eyes focused on the graphic's movement path, and respond quickly. It aims to test the subject's reaction speed and ability to perform tasks when processing dynamic information, especially in rapidly changing situations.

[0161] The subjects' accuracy and reaction time in the figure tracking task were then obtained.

[0162] The required image content of the cognitive stimulation simulation experiment was presented through a computer interface, and the stimulation time interval and order were randomized to avoid predictive interference. The cognitive stimulation simulation experiment environment was a closed working space with different light environment layouts. The reaction time of the subjects from stimulus presentation to response was recorded by professional equipment. The reaction time was measured in milliseconds, and the accuracy of the task completion was counted. The accuracy rate was the ratio of the number of correctly completed tasks to the total number of tasks, and the accuracy rate range was [0, 1].

[0163] In data analysis, reaction time is used to evaluate the cognitive agility and task completion speed of the subjects, accuracy is used to measure the correctness of task judgment and execution accuracy, and cognitive efficiency value is calculated by the following formula:

[0164]

[0165] in, Represents the comprehensive evaluation value of cognitive effectiveness, Indicates the total number of test tasks. Indicates The accuracy of the task, Indicates The reaction time of the task.

[0166] S4.5. Obtain parameters of different light environment layouts and corresponding comprehensive performance to obtain an initial data set, and establish a multi-dimensional comprehensive performance evaluation model based on the initial data set;

[0167] Specifically, different light environment layouts and corresponding comprehensive efficacy parameters are obtained to obtain the initial data set. Since the units of psychological efficacy parameters, physiological efficacy parameters, and cognitive efficacy parameters are different, a standardized method is used to process each parameter of comprehensive efficacy separately:

[0168]

[0169] in, are the parameters of each comprehensive performance after standardization. are the parameters of each comprehensive performance in the initial data set (including ,RMSSD, , as well as ), is the average value of the corresponding comprehensive performance parameter, is the standard deviation of the corresponding comprehensive performance parameter.

[0170] In order to evaluate the impact of different performance dimensions on the comprehensive performance score, the data set matrix is ​​constructed as follows:

[0171]

[0172] Among them, the physiological efficacy dimension After standardization , RMSSD and Composition, psychological efficacy dimension After standardization Composition, cognitive efficacy dimension After standardization constitute, is the comprehensive effectiveness questionnaire score, and W is the number of light environment layouts generated in step S4.1.

[0173] Overall effectiveness questionnaire score Obtaining:

[0174] In order to systematically evaluate the comprehensive impact of the light environment layout on the subjects' physiological performance, psychological performance and cognitive performance, the questionnaire is divided into three modules, corresponding to the performance evaluation of the physiological dimension, psychological dimension and cognitive dimension respectively. The physiological dimension includes indicators such as perceived fatigue and discomfort; the psychological dimension includes indicators such as perceived negative emotions and pressure perception; the cognitive dimension includes indicators such as inattention; the Likert scale of 1 to 5 is used for scoring, with 1 representing very little and 5 representing very much, to ensure the quantification of data and the convenience of analysis; the score of each module is calculated by the mean to represent the corresponding performance dimension, and the scores of each module are integrated to generate a comprehensive performance score of the corresponding light environment layout. , as the experimental target value; in this way, the comprehensive performance score corresponding to each light environment layout is obtained .

[0175] The questionnaire design was verified for reliability and validity before implementation. The results showed that the questionnaire structure was reasonable and the data was reliable. In order to reduce the impact of individual scoring differences, all scoring data were standardized, laying the foundation for subsequent correlation analysis and weight calculation.

[0176] The Spearman rank correlation coefficient was used to calculate the correlation between each efficacy dimension and the comprehensive efficacy questionnaire score. The results can be expressed as the physiological efficacy correlation , psychological efficacy correlation Correlation with cognitive efficacy , the higher the correlation, the greater the effectiveness impact of this effectiveness dimension.

[0177] Next, The absolute value of The absolute value and The absolute value of is converted into a weight to ensure that the sum of the weights of the three is 1, thus obtaining the impact factor of each performance dimension:

[0178]

[0179] in, The correlation between different efficacy and comprehensive efficacy questionnaire scores, j = , or .

[0180] Finally, the multi-dimensional comprehensive performance evaluation model is expressed as:

[0181]

[0182] in, is the normalized value of physiological efficacy, is the standardized value of psychological efficacy, It is a standardized value for cognitive effectiveness to ensure unit consistency between different dimensions.

[0183] To further illustrate, step S5 includes the following steps:

[0184] S5.1. Randomly generate light environment layouts of multiple light environments, and divide the obtained light environment layouts into a training set and a test set;

[0185] Specifically, under different light environment layouts, the light environment layout variable influencing factors collected in step S3.2 are And the multi-dimensional comprehensive performance evaluation value S collected by step S4.5 is expressed as: ,in Indicates the value of the light environment layout variable influencing factor corresponding to different light environment layouts, Represents the multi-dimensional comprehensive performance evaluation value corresponding to different light environment layouts.

[0186] To ensure the representativeness and universality of the model, 200 groups of different light environment layouts are randomly generated. In each group, any three positions in the 4×4 array layout are selected to place the light source, thus covering a variety of layout combinations. At this time, E=200.

[0187] Then, the 200 sets of light source environment layouts were divided into 160 training sets and 40 test sets. The test sets were used to evaluate the model performance and avoid overfitting.

[0188] S5.2. Based on the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model, a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance is constructed. The obtained multi-dimensional non-linear prediction model is optimized through training sets and test sets to improve its prediction accuracy.

[0189] Due to the influence of light environment layout variables The complex relationship between the multi-dimensional comprehensive effectiveness evaluation value S is preferably based on the training set and fitted by a polynomial regression model to establish a multi-dimensional nonlinear prediction model:

[0190]

[0191] in, represents the highest power of the polynomial, and the optimal value is determined by cross-validation. They all represent the regression parameters to be optimized.

[0192] Using mean square error MSE as the loss function, the prediction error of the multi-dimensional comprehensive performance evaluation prediction model is evaluated based on the test set, specifically:

[0193]

[0194] in, represents the actual comprehensive performance value of the bth sample in the test set, represents the comprehensive performance value of the b-th sample prediction in the test set, represents the number of samples in the test set, Represents the light environment layout variable influencing factor of the bth sample.

[0195] Preferably, the gradient descent algorithm is used to adjust the regression parameters to be optimized in the objective function. , in order to minimize MSE and reduce the misprediction error.

[0196] In addition, using the coefficient of determination Evaluate the goodness of fit of the objective function, and the calculation formula is:

[0197]

[0198] in Represents the average value of the actual comprehensive performance values ​​of all samples in the test set; A value close to 1 indicates that the model fit is good.

[0199] After verification, a multi-dimensional nonlinear prediction model was finally established .

[0200] In practical applications, input new light environment layout influencing factors , through multi-dimensional nonlinear prediction model Predict the corresponding comprehensive efficacy value .

[0201] To further illustrate, step S6 includes the following steps:

[0202] S6.1, using the light environment layout obtained in step S5.1, obtaining an experimental data set of light environment layout and corresponding comprehensive efficiency evaluation value according to a multi-dimensional comprehensive efficiency evaluation model, and constructing an efficiency prediction model through a deep neural network based on the experimental data set;

[0203] Specifically, based on the 200 sets of light environment layouts obtained in step S5.1, the parameters of the comprehensive performance of the subjects under different light environment layouts are recorded, and the corresponding multi-dimensional comprehensive performance evaluation values ​​are obtained through step S4.5. , and record the layout parameters corresponding to different light environment layouts , construct the experimental data set, { }.

[0204] A deep neural network is used to build an efficiency prediction model, where the input layer contains the layout parameters of the light environment layout The hidden layer consists of three layers, with 128, 64, and 32 neurons in each layer, respectively. The activation function is ReLU (Rectified Linear Unit), and dropout (random inactivation layer) is added to prevent overfitting. The output layer is a single neuron, the activation function is linear, and the output performance prediction value is .

[0205] The training set obtained in step S5.1 is used for model training.

[0206] Set the loss function to mean square error MSE:

[0207]

[0208] in, is the actual multi-dimensional comprehensive performance evaluation value of the cth light environment layout group among the 200 light environment layout groups, It is the performance prediction value of the cth lighting environment layout group among the 200 lighting environment layout groups.

[0209] The Adam (Adaptive Moment Estimation) optimizer is used for weight update, the learning rate is 0.001, and the early stopping mechanism is used to determine the optimal training round to avoid overfitting.

[0210] Based on the trained performance prediction model, the performance value As the objective function:

[0211]

[0212] in, is the prediction function of the deep learning model, and LP is the layout parameter of the light environment.

[0213] S6.2. Use genetic algorithms to solve the optimal efficiency value and optimal layout parameters of the efficiency prediction model, and configure the light environment of the enclosed working space according to the optimal layout parameters.

[0214] The genetic algorithm is used to optimize the layout parameters, which includes initializing the population, that is, randomly generating the initial light environment layout group. ; Fitness function, that is, using deep learning models to predict performance values ; Genetic operations, including selection, crossover and mutation, to update the population of layout parameters and optimize them generation by generation; Convergence condition, when the performance prediction value The optimization is terminated when convergence or the preset maximum performance value is reached.

[0215] Finally, the optimal performance value is obtained and optimal layout parameters ; In practical applications, use the optimal layout parameters Configure the light environment.

[0216] S6.3, input the optimal layout parameters into the multi-dimensional nonlinear prediction model obtained in step S5.2 to obtain the optimal performance prediction value, and determine whether the deviation between the optimal performance prediction value and the optimal performance prediction value is within the set range. If so, the process ends; if not, the multi-dimensional nonlinear prediction model is optimized again through the training set and the test set, and then step S6.3 is executed;

[0217] Specifically, in order to verify whether the prediction results of the multi-dimensional nonlinear prediction model obtained in step S5.2 are accurate, the Input the multi-dimensional nonlinear prediction model obtained in step S5.2 to obtain the optimal performance prediction value S 预测 ,judge and S 预测 Whether the deviation belongs to the set range, if so, the process ends; if not, it is necessary to re-optimize the multidimensional nonlinear prediction model obtained in step S5.2, so it is necessary to re-execute step S5.2. After the optimization is completed, step S6.3 is executed again to predict the multidimensional nonlinear prediction model again until the end.

[0218] For example, and S 预测 The deviation is calculated as:

[0219]

[0220] The setting range is 0~5%, so when and S 预测 When the deviation does not exceed 5%, the optimization of the multidimensional nonlinear prediction model is considered complete.

[0221] Example 2

[0222] Based on the closed work space light environment layout method provided in Example 1, this embodiment provides a closed work space light environment layout system, including:

[0223] An array element variable determination unit, used to determine array element variables that affect the light environment layout of a confined working space;

[0224] An array element variable normalized influence factor extraction unit is used to analyze the influence of array element variables on the comprehensive performance in the light environment layout and extract the normalized influence factor of the array element variables;

[0225] An array element variable influence factor model building unit is used to build an array element variable influence factor model based on the normalized influence factor of the array element variable;

[0226] A multi-dimensional comprehensive performance evaluation model establishment unit is used to randomly generate a variety of light environment layouts, collect parameters of each light environment layout and corresponding comprehensive performance, obtain an initial data set, and establish a multi-dimensional comprehensive performance evaluation model based on the initial data set;

[0227] A multi-dimensional nonlinear prediction model building unit is used to build a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance according to the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model;

[0228] The best matching scheme determination unit is used to optimize the light environment layout to maximize the comprehensive efficiency based on the multi-dimensional comprehensive efficiency evaluation model and utilize the deep neural network to determine the best matching scheme between the light environment layout and the comprehensive efficiency.

[0229] Among them, the parameters of comprehensive efficacy include physiological efficacy parameters, psychological efficacy parameters and cognitive efficacy parameters;

[0230] Further explanation: the multi-dimensional comprehensive performance evaluation model establishment unit includes:

[0231] A light environment layout creation module is used to randomly obtain a variety of light environment layouts by changing array element variables;

[0232] A physiological performance parameter acquisition module is used to obtain the physiological performance parameters of the subjects under different light environment layouts;

[0233] The psychological efficacy parameter acquisition module is used to obtain the psychological efficacy parameters of the subjects under different light environment layouts;

[0234] A cognitive performance parameter acquisition module is used to obtain the cognitive performance parameters of the subjects under different light environment layouts;

[0235] The multi-source data fusion processing module is used to obtain the parameters of different light environment layouts and corresponding comprehensive performance to obtain an initial data set, and to establish a multi-dimensional comprehensive performance evaluation model based on the initial data set.

[0236] refer to Figure 3 , it is further explained that the confined work space light environment layout system also includes a confined work space simulation unit and a layout scheme display unit;

[0237] Among them, the confined working space simulation unit is used to simulate the confined working space and also used to simulate the light environment layout;

[0238] Specifically, the confined working space simulation unit can not only simulate different confined working spaces according to needs, but also can simulate different light environment layouts in the corresponding confined working space according to the input array element variables or adjusting the number of light sources, light source height, light source color temperature, light source illumination, light source angle and light source position of the corresponding confined working space, in conjunction with the simulated confined working space.

[0239] Therefore, the closed working space simulation unit can cooperate with the array variable determination unit and the psychological performance parameter collection module to run the lighting design software.

[0240] A layout scheme display unit, used to display the enclosed working space, display the light environment layout, display the comprehensive efficiency of the corresponding light environment layout, and display the parameters of the corresponding light environment layout;

[0241] Specifically, the layout scheme display unit displays the parameters of the light environment layout input into the confined working space simulation unit. The layout scheme display unit also displays the confined working space simulated by the confined working space simulation unit and the simulated light environment layout, so that designers or subjects can experience it intuitively.

[0242] Therefore, the layout scheme display unit can simulate and display a confined working space of 2 meters long, 2 meters wide and 2.5 meters high through the confined working space simulation unit and in cooperation with the array element variable determination unit. It can also simulate and display the lighting coverage under the conditions of different light source arrangement quantities, different light source positions and different light source illumination conditions.

[0243] The layout scheme display unit can test and evaluate the psychological comfort and emotional state of the subjects in the simulation scenario when cooperating with the psychological performance parameter acquisition module through the closed working space simulation unit.

[0244] The layout scheme display unit can also obtain the best matching scheme of the light environment layout and the comprehensive performance obtained by the best matching scheme determination unit, and display it after simulation by the closed working space simulation unit.

[0245] The layout scheme display unit can also serve as an input end to send parameters of the light environment layout and parameters of the enclosed working space to the enclosed working space simulation unit.

[0246] Example 3

[0247] refer to Figure 4 Based on the enclosed work space light environment layout method provided in Example 1, this embodiment provides a enclosed work space light environment layout device, including a central processing unit, a non-volatile storage medium, an input device, a network interface, a memory, a data storage device and a sensor device, wherein the non-volatile storage medium stores an operating system and a computer program; the central processing unit is respectively communicated with the non-volatile storage medium, the input device, the network interface, the memory, the data storage device and the sensor device through a system bus.

[0248] The central processing unit is responsible for executing computer programs, coordinating and completing the operation of the entire computer operating system and external devices; the operating system is used to provide underlying system language support and peripheral interface management, and coordinate the operation of hardware resources and computer programs.

[0249] Specifically, the input devices include a mouse and a keyboard. For example, the mouse and keyboard can be used together to input the parameters of the enclosed working space that need to be simulated, the parameters of the light environment layout that need to be simulated, the comprehensive performance scores of different light environment layouts, the emotion scale scores of the subjects in different light environment layouts, and the accuracy of the subjects in the cognitive stimulation simulation experiment.

[0250] The network interface is connected to the external network to achieve high-speed data transmission, thereby enabling remote control via the network.

[0251] The data storage device is used to store the relevant data input by the input device, the relevant data obtained by the sensor device and the optimal layout parameters solved by the layout solving module, and is used to generate valid data readable by the computer program from the received data.

[0252] The sensor equipment is used to collect the original physiological performance parameters such as the heart rate HR and skin conductivity SCL rate of the subjects under different light environment layouts, and to collect the reaction time of the subjects in the cognitive stimulation simulation experiment.

[0253] Specifically, the computer program includes a user interface module, a data import module, a data processing module, an environmental simulation module I, a layout solution module, an environmental simulation module II, an efficiency prediction module, a data storage module and a system integration module.

[0254] The user interface module is used to display the light environment layout of the enclosed working space simulated by the environmental simulation module I and the environmental simulation module II, and can also display the optimal layout parameters and the optimal efficiency value. It can also be used as an input device to input the parameters of the light environment layout and the parameters of the enclosed working space.

[0255] The data import module is used to read the valid data generated by the data storage device to ensure the integrity and standardization of the input data.

[0256] The data processing module is used to convert and preprocess the data obtained by the data import module to provide a high-quality data foundation for subsequent calculations; for example, the comprehensive performance score is associated with the changes of a single variable under different parameter settings, and a model of the comprehensive performance score and the single variable light environment layout is constructed; the model of the comprehensive performance score and the single variable light environment layout is fitted by the least squares method to obtain the normalized influence factor of the array element variable; the Cartesian coordinates of the light source position are converted into polar coordinates; according to the normalized influence factor of the array element variable and the polar coordinates of the light source position, an array element variable influence factor model is constructed; by changing the array element variable, a variety of light environment layouts are randomly obtained; according to the original physiological performance parameters, the physiological performance parameters of the subjects under different light environment layouts are obtained; according to the emotional scale score, the psychological comfort score of different light environment layouts is calculated; according to the reaction time and accuracy of the subjects in the cognitive stimulation simulation experiment, the cognitive performance value is calculated; according to the initial data set, a multi-dimensional comprehensive performance evaluation model is established; according to the light environment layout, the corresponding light environment layout variable influence factor is calculated. And the multi-dimensional comprehensive performance evaluation value S and the performance prediction model constructed by deep neural network according to the light environment layout.

[0257] The environment simulation module I is used to simulate the enclosed working space according to the parameters of the enclosed working space, and to simulate the light environment layout corresponding to the enclosed working space according to the parameters of the light environment layout.

[0258] Layout solving module, used to calculate the light environment layout variable influencing factors corresponding to different light environment layouts A multidimensional nonlinear prediction model is established based on the multidimensional comprehensive performance evaluation value S, the gradient descent algorithm is used to reduce the prediction error of the multidimensional nonlinear prediction model, and the genetic algorithm is used to solve the performance prediction model to obtain the optimal performance value and optimal layout parameters.

[0259] Environmental simulation module II simulates the optimal light environment layout of the corresponding confined working space according to the optimal layout parameters to verify the feasibility and effectiveness of the design scheme.

[0260] The performance prediction module is used to predict the comprehensive performance value of the corresponding light environment layout according to the input parameters of the light environment layout.

[0261] The data storage module is used to store experimental data and save the data obtained by the data processing module, layout solution module and performance prediction module to ensure data security and traceability;

[0262] The system integration module integrates the functions of each module and coordinates the efficient operation between modules.

[0263] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the present invention.

Claims

1. A method for arranging the light environment of a closed working space, characterized in that: The following steps are involved: S1. Determine array element variables that affect the layout of the light environment of the enclosed work space; the array element variables include the number of light sources, light source height, light source color temperature, light source illumination, light source angle, and light source position; S2. Analyze the influence of array element variables on comprehensive performance in the light environment layout, and extract the normalized influence factor of array element variables; S3. Establishing an array element variable influence factor model based on the normalized influence factors of the array element variables; S4, randomly generating a variety of light environment layouts, collecting parameters of each light environment layout and corresponding comprehensive performance, obtaining an initial data set, and establishing a multi-dimensional comprehensive performance evaluation model based on the initial data set; S5. Based on the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model, a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance is constructed; S6. Based on the multi-dimensional comprehensive performance evaluation model, use deep neural networks to optimize the light environment layout to maximize the comprehensive performance, determine the best matching solution between the light environment layout and the comprehensive performance, and verify the multi-dimensional nonlinear prediction model; The step S3 comprises: S3.1, converting the Cartesian coordinates of the light source position into polar coordinates; S3.2, constructing an array element variable influence factor model according to the normalized influence factor of the array element variable and the polar coordinates of the light source position; The parameters of the comprehensive efficacy include physiological efficacy parameters, psychological efficacy parameters and cognitive efficacy parameters; The step S4 comprises: S4.1, by changing the array element variables, a variety of light environment layouts are randomly obtained; S4.

2. Obtaining the physiological performance parameters of the subjects under different light environment layouts; S4.3, obtaining the psychological efficacy parameters of the subjects under different light environment layouts; S4.

4. Obtain the cognitive performance parameters of the subjects under different light environment layouts; S4.

5. Obtain parameters of different light environment layouts and corresponding comprehensive performance to obtain an initial data set, and establish a multi-dimensional comprehensive performance evaluation model based on the initial data set.

2. The method for arranging the light environment of a confined working space according to claim 1, characterized in that: The step S1 is specifically as follows: The array element variables that affect the light environment layout of confined working spaces are determined through experimental measurement, theoretical deduction and scene simulation analysis.

3. The method for arranging the light environment of a confined working space according to claim 2, characterized in that: The step S2 comprises: S2.

1. Build a closed working space and create an adjustable light environment layout; S2.2, adjusting the parameters of a single variable in the array element variables to obtain a variety of single variable light environment layouts, obtaining the degree of influence of different single variable light environment layouts on the subjects, obtaining a comprehensive efficacy score, and constructing a model of the comprehensive efficacy score and the single variable light environment layout; S2.

3. The comprehensive performance score is fitted with the model of the single variable light environment layout by the least squares method to obtain the normalized influence factor of the array element variable.

4. The method for arranging the light environment of a confined working space according to claim 1, characterized in that: The step S5 comprises: S5.

1. Randomly generate light environment layouts of multiple light environments, and divide the obtained light environment layouts into a training set and a test set; S5.

2. Based on the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model, a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance is constructed. The obtained multi-dimensional non-linear prediction model is optimized through training sets and test sets to improve its prediction accuracy.

5. The method for arranging the light environment of a confined work space according to claim 4, characterized in that: The step S6 comprises: S6.1, using the light environment layout obtained in step S5.1, obtaining an experimental data set of light environment layout and corresponding comprehensive efficiency evaluation value according to a multi-dimensional comprehensive efficiency evaluation model, and constructing an efficiency prediction model through a deep neural network based on the experimental data set; S6.

2. Use genetic algorithms to solve the optimal efficiency value and optimal layout parameters of the efficiency prediction model, and configure the light environment of the enclosed working space according to the optimal layout parameters; S6.

3. Input the optimal layout parameters into the multidimensional nonlinear prediction model obtained in step S5.2 to obtain the optimal performance prediction value, and determine whether the optimal performance prediction value is equal to the optimal performance prediction value. If so, end; if not, re-optimize the obtained multidimensional nonlinear prediction model through the training set and the test set and execute step S6.

3.

6. A closed working space light environment layout system, characterized in that: A method for implementing the light environment layout of a confined work space as described in any one of claims 1 to 5, comprising: An array element variable determination unit is used to determine array element variables that affect the layout of the light environment of the enclosed working space; the parameters of the comprehensive performance include physiological performance parameters, psychological performance parameters and cognitive performance parameters; An array element variable normalized influence factor extraction unit is used to analyze the influence of array element variables on the comprehensive performance in the light environment layout and extract the normalized influence factor of the array element variables; An array element variable influence factor model building unit is used to build an array element variable influence factor model based on the normalized influence factor of the array element variable; A multi-dimensional comprehensive performance evaluation model establishment unit is used to randomly generate a variety of light environment layouts, collect parameters of each light environment layout and corresponding comprehensive performance, obtain an initial data set, and establish a multi-dimensional comprehensive performance evaluation model based on the initial data set; A multi-dimensional nonlinear prediction model building unit is used to build a multi-dimensional nonlinear prediction model between the light environment layout and the comprehensive performance according to the array element variable influencing factor model and the multi-dimensional comprehensive performance evaluation model; The best matching scheme determination unit is used to optimize the light environment layout to maximize the comprehensive efficiency based on the multi-dimensional comprehensive efficiency evaluation model and to determine the best matching scheme between the light environment layout and the comprehensive efficiency by using a deep neural network; The multi-dimensional comprehensive effectiveness evaluation model establishment unit includes: A light environment layout creation module is used to randomly obtain a variety of light environment layouts by changing array element variables; A physiological performance parameter acquisition module is used to obtain the physiological performance parameters of the subjects under different light environment layouts; The psychological efficacy parameter acquisition module is used to obtain the psychological efficacy parameters of the subjects under different light environment layouts; A cognitive performance parameter acquisition module is used to obtain the cognitive performance parameters of the subjects under different light environment layouts; The multi-source data fusion processing module is used to obtain the parameters of different light environment layouts and corresponding comprehensive performance to obtain an initial data set, and to establish a multi-dimensional comprehensive performance evaluation model based on the initial data set.

7. The closed work space light environment layout system according to claim 6, characterized in that: The enclosed working space light environment layout system also includes: A confined work space simulation unit is used to simulate a confined work space and also to simulate the light environment layout; The layout scheme display unit is used to display the enclosed working space, the light environment layout, the comprehensive efficiency of the corresponding light environment layout, and the parameters of the corresponding light environment layout.

8. A device for arranging the light environment of a closed working space, characterized in that: It comprises a non-volatile storage medium and a central processing unit, wherein the non-volatile storage medium stores executable code, and when the central processing unit executes the executable code, the method for arranging the light environment of a confined working space as described in any one of claims 1 to 5 is implemented.

Images