Design method, system and equipment for indoor illumination of transformer substation and storage medium

Abstract

The invention relates to the technical field of illumination design, and discloses a transformer substation indoor illumination design method, system and device and a storage medium, and the method comprises the steps: obtaining real-time data, spatial feature data and meteorological data of a transformer substation, inputting the data into an illumination prediction model, and obtaining the illumination demand value of each region of the transformer substation; inputting the illumination demand value and the environmental parameters into a lamp selection model to obtain a lamp layout initial scheme; establishing a three-dimensional simulation model of the transformer substation according to the transformer substation design drawing and the lamp layout initial scheme, and performing iterative simulation to obtain a lamp layout optimization scheme; performing clustering analysis on the lamp layout optimization scheme by adopting a clustering algorithm to obtain a plurality of lamp groups, and setting an independent power distribution loop for each lamp group; and generating a control strategy corresponding to each lamp group according to the illumination demand value. According to the invention, the design precision, the energy efficiency level and the operation and maintenance convenience of the lighting system are remarkably improved by fusing dynamic prediction, multi-objective optimization and intelligent control technologies.

Description

Technical Field

[0001] The present invention relates to the technical field of lighting design, and particularly to a method, a system, a device and a storage medium for indoor lighting design of a substation. Background Art

[0002] Currently, the indoor lighting system of a substation is mainly designed by using the method of estimating with empirical formulas in combination with two-dimensional plane layout according to industry standards. The limitations of these conventional methods are as follows. On the one hand, the design accuracy is insufficient. The existing technology relies on simplified illuminance calculation formulas and does not fully consider the response of spatial heterogeneity, and its lighting indicators are difficult to meet the requirements of the high-precision light environment of the substation. On the other hand, the verification of key indicators is missing. There is a lack of quantitative simulation verification means for the key parameters affecting visual comfort, and the design results cannot be verified by visualization tools, which easily causes problems such as excessive glare or color distortion in the actual scene. In addition, the existing methods also have problems of low energy efficiency and low intelligence level. The layout of indoor lamps in the substation in the existing methods mostly relies on manual experience, which easily leads to exceeding the lighting power density standard, and lacks a dynamic regulation mechanism and cannot optimize the lamp regulation according to the actual environment, thus causing energy waste and an increase in operation and maintenance costs. Summary of the Invention

[0003] In order to solve the above technical problems, the present invention provides a method, a system, a device and a storage medium for indoor lighting design of a substation, so as to solve the problems of low accuracy and high energy consumption existing in the existing design method of the indoor lighting system of the substation, and achieve the effects of improving the indoor lighting quality of the substation and reducing the energy consumption of the substation.

[0004] In a first aspect, the present invention provides a method for indoor lighting design of a substation, and the method includes: Obtain the real-time data, spatial feature data and meteorological data of the substation, and input them into a preset illuminance prediction model to obtain the illuminance requirement values of each area of the substation. The real-time data includes ambient light data, personnel activity data, and energy consumption data, and the illuminance prediction model is constructed based on a neural network model; Input the illuminance requirement values and environmental parameters into a preset lamp selection model to obtain an initial lamp layout plan. The lamp selection model is constructed based on a random forest algorithm, and the initial lamp layout plan includes the initial lamp type and initial lamp parameters; According to the substation design drawings and the initial lamp layout plan, establish a three-dimensional simulation model of the substation, and perform iterative simulation on the three-dimensional simulation model according to preset standards to obtain an optimized lamp layout plan; Use a clustering algorithm to perform clustering analysis on the optimized lamp layout plan to obtain several lamp groups, and set independent power distribution circuits for each lamp group; According to the illumination requirement value, fuzzy inference is used to generate a control strategy corresponding to each lamp group.

[0005] Further, the step of iteratively simulating the three-dimensional simulation model according to a preset standard to obtain a lamp layout optimization scheme includes: Taking the minimization of the sum of squared errors between each simulation standard and the corresponding preset standard as the objective function, and taking the lamp layout parameters as optimization variables, the three-dimensional simulation model is iteratively simulated to obtain a lamp layout optimization scheme, where the lamp layout parameters include lamp spacing, installation height, and the number of lamps; Among them, the following formula is used to represent the objective function: In the formula, E sim represents the average illumination simulation value, E std represents the average illumination standard value, UGR sim represents the glare simulation value, UGR std represents the glare standard value, U sim represents the illumination uniformity simulation value, U std represents the illumination uniformity standard value, LPD sim represents the lighting power density simulation value, LPD std represents the lighting power density standard value.

[0006] Further, the step of using a clustering algorithm to perform clustering analysis on the lamp layout optimization scheme to obtain several lamp groups and setting independent power distribution circuits for each lamp group includes: Taking each lamp as a node, calculating the edge weights between adjacent nodes according to power correlation, spatial distance, and usage frequency synchronization, and constructing a lamp structure diagram according to the nodes and the edge weights; Using the modularity optimization algorithm to perform community division on the lamp structure diagram to obtain several first communities; According to the community power difference and community spatial distance, merging the first communities to obtain several second communities, and obtaining the corresponding lamp groups according to the second communities; According to the total power of the lamps in each lamp group and a preset capacity safety margin, determining the circuit capacity of the lamp group, and using a path planning algorithm to perform cable path planning to obtain the lamp cable routing layout of the lamp group.

[0007] Further, the following formula is used to represent the edge weight: In the formula, w ij represents the edge weight between node i and node j, P i and Pj Denote the power of node i and node j respectively, D ij represents the Euclidean distance between node i and node j, f i and f j Represent the usage frequency of node i and node j respectively, P max Indicates the maximum power, D max Indicates the maximum distance, f max represents the maximum frequency, α represents the power weight, β represents the distance weight, and γ represents the frequency weight.

[0008] Furthermore, the step of using a path planning algorithm to perform cable path planning to obtain a lamp cable routing layout of the lamp group includes: According to the edge weight of the second community corresponding to each lighting group, the actual path cost function is constructed; Construct a heuristic function based on the Euclidean distance between nodes in the second community corresponding to each lighting group and obstacle data around the nodes; Obtaining a total cost function according to the actual path cost function and the heuristic function; According to the total cost function, the A-star algorithm is used to perform iterative calculations to obtain the shortest path corresponding to each lamp group; Obtaining a lamp cable routing layout for each lamp group based on the shortest path; The actual path cost function is expressed by the following formula: Where g(n) represents the actual path cost function of the current node n, w ij Represents the edge weight between node i and node j, Path(start→n) represents the path from the starting node start to the current node n; The heuristic function is expressed by the following formula: Where h(n) represents the heuristic function of the current node n, x n and y n Indicates the coordinate value of the current node n, x g and y g represents the coordinate value of the target node g, λ represents the obstacle coefficient, and ρ(n) represents the obstacle density around the current node n; The total cost function is expressed as follows: Where f(n) represents the total cost function of the current node n.

[0009] Furthermore, the step of generating a control strategy corresponding to each lamp group using fuzzy reasoning according to the illumination requirement value includes: Generate a meteorological disturbance factor according to meteorological data, and generate a control matrix according to the real-time data of the substation, the illumination demand value and the meteorological disturbance factor; The control matrix is input into a preset fuzzy rule library for fuzzy matching to obtain an output value, and the output value is converted into a dimming level and a switching instruction.

[0010] Furthermore, after the step of converting the output value into a dimming level and a switching instruction, the method further includes: According to preset conditions, the scene mode of the substation is determined, and the corresponding scene control strategy is selected according to the scene mode, wherein the scene mode includes daily mode, energy-saving mode, emergency mode and night mode; The dimming level and the switching instruction are modified based on the scene control strategy as a constraint condition.

[0011] In a second aspect, the present invention provides a substation indoor lighting design system, the system comprising: An illumination prediction module is used to obtain real-time data, spatial feature data, and meteorological data of the substation, and input a preset illumination prediction model to obtain the illumination demand value of each area of the substation. The real-time data includes ambient light data, personnel activity data, and energy consumption data. The illumination prediction model is constructed based on a neural network model; a layout design module, configured to input the illumination requirement value and environmental parameters into a preset lamp selection model to obtain an initial lamp layout plan, wherein the lamp selection model is constructed based on a random forest algorithm, and the initial lamp layout plan includes initial lamp types and initial lamp parameters; a layout optimization module, configured to establish a three-dimensional simulation model of the substation based on the substation design drawings and the initial lighting layout plan, and iteratively simulate the three-dimensional simulation model according to preset standards to obtain an optimized lighting layout plan; A circuit design module is used to perform cluster analysis on the lighting fixture layout optimization scheme using a clustering algorithm to obtain a number of lighting fixture groups, and set an independent power distribution circuit for each lighting fixture group; The strategy generation module is used to generate a control strategy corresponding to each lamp group based on the illumination requirement value by using fuzzy reasoning.

[0012] In a third aspect, an embodiment of the present invention further provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0013] Fourthly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0014] The present invention provides a substation indoor lighting design method, system, device and storage medium. Through the collaborative simulation of multiple parameters and dynamic demand prediction, the present invention can achieve high-precision light environment design. Through intelligent grouping control and load balancing optimization, energy conservation and consumption reduction can be achieved. By using the improved A* algorithm to plan the cable path and integrating fuzzy logic control to generate an intelligent control strategy, the adaptive dynamic regulation of the lighting system can be achieved. By integrating dynamic prediction, multi-objective optimization and intelligent control technologies, the present invention can significantly improve the design accuracy, energy efficiency level and operation and maintenance convenience of the lighting system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic flowchart of the substation indoor lighting design method in an embodiment of the present invention; Figure 2 is a schematic structural diagram of the substation indoor lighting design system in an embodiment of the present invention; Figure 3 is an internal structural diagram of a computer device in an embodiment of the present invention; Reference Signs: 10, illuminance prediction module; 20, layout design module; 30, layout optimization module; 40, circuit design module; 50, strategy generation module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Please refer to Figure 1 , a substation indoor lighting design method proposed in the first embodiment of the present invention includes steps S10 to S50: Step S10, obtain the real-time data, spatial feature data and meteorological data of the substation, and input them into a preset illuminance prediction model to obtain the illuminance demand values of each area of the substation. The real-time data includes ambient light data, personnel activity data, and energy consumption data. The illuminance prediction model is constructed based on a neural network model; Step S20: Input the illuminance requirement value and environmental parameters into a preset lamp selection model to obtain an initial lamp layout plan. The lamp selection model is constructed based on the random forest algorithm. The initial lamp layout plan includes the initial lamp type and initial lamp parameters. Step S30: Establish a 3D simulation model of the substation according to the substation design drawings and the initial lamp layout plan, and perform iterative simulation on the 3D simulation model according to preset standards to obtain an optimized lamp layout plan. Step S40: Use a clustering algorithm to perform clustering analysis on the optimized lamp layout plan to obtain several lamp groups, and set independent power distribution circuits for each lamp group. Step S50: Generate a control strategy corresponding to each lamp group by using fuzzy inference according to the illuminance requirement value.

[0018] In the present invention, first, the lighting brightness requirements of different areas of the substation are analyzed and predicted. The used illuminance prediction model is constructed based on a neural network model. Preferably, the illuminance prediction model is constructed based on a long short-term memory neural network (LSTM) or a Transformer model. Among them, the input data of the illuminance prediction model includes the real-time data of the substation, spatial feature data, and meteorological data. The real-time data includes environmental light data, personnel activity data, energy consumption data, etc. The meteorological data includes temperature, humidity, natural light intensity, etc. The output data of the model is the illuminance requirement value of each area of the substation in the future time period. Each area of the substation refers to the functional areas obtained by dividing the substation into areas according to different functions, including the control room, power distribution room, etc. It should be noted that the structure and training process of the illuminance prediction model in this embodiment can refer to the structure and training method of a conventional neural network model, and will not be limited here too much.

[0019] After predicting the illuminance demand values of different areas of the substation through the illuminance prediction model, according to the illuminance demand values and the actual environmental parameters of the substation, the type and parameters of the lamps in each area are selected through the lamp selection model, so as to obtain the initial lamp layout scheme. Specifically, in this embodiment, the random forest algorithm is preferably used to construct the lamp selection model. The input features of this model include the illuminance demand value and environmental parameters. The environmental parameters include the functional area, the reflection coefficient and spatial area of each area, etc. The random forest model randomly samples the original data set to form multiple different sample data sets, and then builds multiple different decision tree models according to these sample data sets. Finally, the final result is obtained according to the average value or voting situation of these decision tree models. In this embodiment, by establishing multiple decision trees and adopting the integrated voting method, the initial lamp type and lamp parameters are voted and selected from the results of multiple decision trees. Here, the lamp parameters include luminous flux, installation height, lamp spacing, number of lamps and maintenance factor, etc. Finally, according to the initial lamp type and lamp parameters, the initial lamp layout scheme is determined.

[0020] Then, according to the initial lamp layout scheme and the two-dimensional design drawings of the substation, a three-dimensional simulation model is constructed. Specifically, first, use AutoCAD to process the two-dimensional floor plan of the substation, extract key information such as room size and door and window positions, and then establish a fine model through SketchUp / BIM, and convert the lamp parameters and environmental parameters into structured data recognizable by DIALux software. The lamp parameters include luminous flux, installation height, lamp spacing, number of lamps, maintenance factor, and the environmental parameters include the functional area, the reflection coefficient and spatial area of each area. Among them, the luminous flux refers to the total luminous amount of the lamp; the installation height refers to the height of the lamp from the ground; the lamp spacing refers to the horizontal distance between the lamps; the maintenance factor refers to the lamp light decay coefficient, which is used to correct the luminous flux attenuation after long-term use; the reflection coefficient includes the reflection coefficients of the ceiling, wall and floor on the light, and the spatial area is represented by the spatial dimensions (length, width, height). Then, these parameters are organized into a file structure recognizable by DIALux software such as JSON or CSV. And through API calls or plugins, these data files are imported into DIALux software to create a room model, and the lamps are arranged in batches according to the coordinates determined by the lamp spacing, and the lamp attributes are designed, including luminous flux, power and light distribution curve, etc. Then, the simulation environment configuration is carried out, including the light source model configuration and the grid accuracy configuration. For example, select the IES / LDT file to define the lamp light distribution curve, and set the grid density according to the preset specifications. The specific three-dimensional model construction steps can refer to the general construction steps of DIALux software, which will not be elaborated here one by one.

[0021] In a preferred embodiment, after constructing the three-dimensional simulation model, the three-dimensional simulation model is iteratively optimized according to a preset standard to obtain a lighting layout optimization scheme. The specific optimization steps are as follows: Taking the minimization of the sum of squared errors between each simulation standard and the corresponding preset standard as the objective function, and using the lighting layout parameters as the optimization variables, the three-dimensional simulation model is iteratively simulated to obtain a lighting layout optimization scheme. The lighting layout parameters include lamp spacing, installation height, and the number of lamps.

[0022] In this embodiment, first, the preset standard values of the key indicators are set according to industry standards. The key indicators here include average illuminance, glare value, illuminance uniformity, and lighting power density, and the standard values of these key indicators are set. Then, the simulation values of the key indicators of the three-dimensional simulation model are calculated. Among them, the average illuminance is calculated using the luminous flux method, and the average illuminance = total luminous flux of the light source * utilization factor * maintenance factor / area of the region. The illuminance uniformity is calculated using the point illuminance method, that is, the illuminance value is calculated point by point, and then the illuminance uniformity is calculated through the ratio of the minimum illuminance to the average illuminance; the glare value can be calculated through the CIE glare value formula.

[0023] Based on the sum of squared errors between the preset standard values and the simulation values, an objective function is established: In the formula, E sim represents the simulation value of the average illuminance, E std represents the standard value of the average illuminance, UGR sim represents the simulation value of the glare, UGR std represents the standard value of the glare, U sim represents the simulation value of the illuminance uniformity, U std represents the standard value of the illuminance uniformity, LPD sim represents the simulation value of the lighting power density, LPD std represents the standard value of the lighting power density.

[0024] According to the above objective function, using the lamp spacing, installation height, and the number of lamps as the optimization variables, the three-dimensional simulation model is iteratively optimized. When performing iteration, the variable step size can be adjusted using the gradient descent method for iteration, or the exhaustive method can be used to traverse all possible parameter combinations within a reasonable range until the objective function is less than the threshold or the maximum number of iterations is reached, to obtain the optimized lamp type and optimized lamp parameters, thereby determining the lighting layout optimization scheme.

[0025] For the lighting layout optimization scheme, the present invention divides the lamps into multiple logically related lamp groups through cluster analysis, and then sets independent power distribution circuits for each lamp group. The power distribution circuit includes circuit capacity and cable routing.

[0026] In a preferred embodiment, the lamp power, usage frequency, and position coordinates are used as the lamp feature matrix, and the K-Means algorithm is employed to perform clustering analysis on the lamp feature matrix to obtain each lamp group. Then, based on the total power of the lamps in each lamp group, the capacity of the power distribution circuit is designed, and based on the coordinate positions of the lamps in each lamp group, with the minimization of the cable length as the objective function, the cable routing scheme for the lamps in each lamp group is calculated. Specifically, the total power P of the lamps within the lamp group is calculated, and the total power P of the lamps is expanded according to a preset capacity safety margin, such as 1.2 times, to obtain the power distribution circuit capacity P`, that is, P` = P * 1.2. Then, based on the lamp position coordinates, the cable routing is planned according to the minimization of the total cable length, the minimum spanning tree algorithm is used to obtain the shortest path number connecting all the lamps, and finally, the spanning tree is traversed and the branch paths are merged to generate the cable routing scheme.

[0027] Since the K-Means algorithm has the problem of relying on the initial centroid, if the initial centroid is not properly selected, the algorithm may fall into a local optimal solution and thus fail to find a better global solution. Therefore, in another preferred embodiment, the present invention uses an improved community discovery algorithm based on graph theory for lamp grouping and uses a path planning algorithm to determine the cable routing. The specific steps include: Regarding each lamp as a node, the edge weights between adjacent nodes are calculated according to power correlation, spatial distance, and usage frequency synchronization. Based on the nodes and the edge weights, a lamp structure diagram is constructed; The modularity optimization algorithm is used to perform community partitioning on the lamp structure diagram to obtain several first communities; Based on the community power difference and community spatial distance, the first communities are merged to obtain several second communities, and corresponding lamp groups are obtained according to the second communities; Based on the total power of the lamps in each lamp group and the preset capacity safety margin, the circuit capacity of the lamp group is determined, and a path planning algorithm is used to perform cable path planning to obtain the cable routing layout of the lamps in the lamp group.

[0028] In this embodiment, first, graph construction and edge weight definition are performed. Specifically, each lamp is regarded as a node, and the edge weights are defined by comprehensively considering the power correlation, spatial distance, and usage frequency synchronization between adjacent nodes. The formula for calculating the edge weights is as follows: where w ij represents the edge weight between node i and node j, P i and P j respectively represent the powers of node i and node j, D ij represents the Euclidean distance between node i and node j, fi and f j represent the usage frequencies of node i and node j respectively, and P max represents the maximum power, and D max represents the maximum distance, and f max represents the maximum frequency. α represents the power weight, β represents the distance weight, γ represents the frequency weight, and α + β + γ = 1 is satisfied.

[0029] In the above formula, the meaning of the power weight is that power difference dominates the grouping. For example, it can make high - power lamps be supplied with power centrally. The meaning of the distance weight is that spatial distance dominates the grouping. For example, neighboring lamps are divided into the same circuit. The meaning of the frequency weight is that frequency synchronization dominates the grouping. For example, lamps with high usage frequency are controlled collaboratively.

[0030] According to the node and edge weights, a lamp structure diagram is established, and then the modularity optimization algorithm is used to partition the nodes in the lamp structure diagram into communities. Modularity is the core index in complex network analysis for evaluating the quality of community partitioning. Its core idea is to compare the difference between the number of internal edges in the community in the actual network and the expected value in the random case. The larger the modularity value, the closer the internal connection in the community and the sparser the external connection. In this embodiment, based on the defined modularity Q, the modularity optimization algorithm (Louvain algorithm) is used for iterative optimization. The Louvain algorithm divides nodes into different communities through iteration and optimizes the modularity score in each iteration to find the optimal community partitioning. The specific steps of the Louvain algorithm are divided into two stages: the local optimization stage and the community merging stage. In the local optimization stage, first, each node is initialized as an independent community, and then all nodes are traversed. Try to move the node to an adjacent community and select the move that maximizes the Q gain, and repeat until Q no longer improves. In the community merging stage, each community is merged into a super - node, and the edge weight is the sum of the edge weights between the original communities. These two stages are repeatedly iterated until no further merging is possible, thereby obtaining the community partitioning.

[0031] To avoid waste of loop capacity caused by over - partitioning and improve the engineering practicability, in this embodiment, after obtaining multiple first - level communities through the Louvain algorithm, it is also necessary to perform dynamic community merging on each first - level community. The merging conditions include community power difference constraint and community spatial distance constraint. Specifically, first, calculate the total lamp power difference between two communities c1 and c2, and determine whether the total lamp power difference is less than the difference threshold. Among them, the total lamp power difference P c has the following expression: ‌ In the formula, P c1 represents the total lamp power of community c1, and P c2 represents the total lamp power of community c2. Represents the maximum value function.

[0032] Then calculate the centroid distance between two communities and determine whether the centroid distance is less than the distance threshold.

[0033] If the total power difference between two communities is less than the difference threshold and the centroid distance is less than the distance threshold, then merge the two communities, recalculate the modularity Q, and continue iterative optimization based on the Louvain algorithm until the final community grouping is obtained, with each community corresponding to a lighting fixture group.

[0034] Then design independent distribution circuits and optimize the cable routing based on the community division results. For the design of distribution circuits, first calculate the total power of the lighting fixtures in each lighting fixture group, and then expand the total power of the lighting fixtures according to the preset capacity safety margin to obtain the circuit capacity. For cable routing, in a preferred embodiment, the A* algorithm is used for cable path planning. The specific steps include: Construct an actual path cost function according to the edge weights of the second community corresponding to each lighting fixture group; Construct a heuristic function according to the Euclidean distance between the nodes and the obstacle data around the nodes in the second community corresponding to each lighting fixture group; Obtain the total cost function according to the actual path cost function and the heuristic function; Perform iterative calculations using the A* algorithm according to the total cost function to obtain the shortest path corresponding to each lighting fixture group; Obtain the lighting cable routing layout of each lighting fixture group according to the shortest path.

[0035] In this embodiment, for each lighting fixture group, first establish a total cost function. In the A* algorithm, the total cost function consists of two parts: the actual path cost function and the heuristic function. Among them, the actual path cost refers to the actual cumulative path cost from the starting node to the current node n. In this embodiment, the actual path cost is represented by the sum of all edge weights in the second community corresponding to the lighting fixture group: In the formula, g(n) represents the actual path cost function of the current node n, w ij represents the edge weight between node i and node j, and Path(start→n) represents the path from the starting node start to the current node n.

[0036] The heuristic function represents the estimated minimum cost from the current node n to the target node. In this embodiment, the estimated minimum cost is represented by the Euclidean distance between the current node n and the target node g, that is, the end point g, and an obstacle density factor is introduced to dynamically correct the estimated minimum cost. Its formula is expressed as: Wherein, h(n) represents the heuristic function of the current node n, and x n and y n represent the coordinate values of the current node n, and x g and y g represent the coordinate values of the target node g, λ represents the obstacle coefficient, and ρ(n) represents the obstacle density around the current node n.

[0037] In the heuristic function, the Euclidean distance is calculated based on the node coordinate values, and at the same time, an obstacle influence factor is introduced. The obstacle influence factor includes an obstacle influence coefficient and an obstacle density. Among them, the obstacle influence coefficient is a preset value. The larger this value is, the greater the influence of the obstacle on the path planning, and the more inclined to bypass the obstacle. The obstacle density represents the number of obstacles per unit area around the current node n. The obstacle density can be obtained through grid map statistics, that is, by counting the number of obstacles within a range with a radius of 1 meter centered on node n, and dividing the number of obstacles by the circular area to obtain the obstacle density.

[0038] Adding the above two formulas can obtain the total cost function f(n): Finally, the A* algorithm is used to search and solve the total cost function, so as to obtain the shortest path connecting all nodes within the lamp group, and based on the shortest path, the cable routing path is planned to obtain the cable routing layout of the lamps. It should be noted that the steps of using the A* algorithm for search and solution can refer to the conventional solution steps and will not be repeated here.

[0039] In this embodiment, the total cost function of the conventional A* algorithm is improved, and spatial, electrical and frequency characteristics are integrated into the actual path cost function, which can realize the collaborative optimization of path length, load balance and operation and maintenance cost. For the heuristic function, on the basis of the conventional geometric distance, dynamic obstacle perception is carried out by introducing an obstacle influence factor, so that the cable routing preferentially bypasses high-risk areas, enhancing the safety and adaptability in complex environments.

[0040] In a preferred embodiment, to enhance the safety of path planning, a distance safety margin constraint is added, that is, an inflation radius is set, and in the original obstacle grid map, the obstacle boundary is expanded outward by the inflation radius to generate a safety map. Then, path planning is performed based on the safety map to ensure that the actual distance between the cable and the obstacle is greater than or equal to the inflation radius. After the path planning is completed, the distance between each node on the planned path and the nearest obstacle is checked. If this distance is less than the inflation radius, the path is considered invalid and needs to be replanned. In this way, the path can be forced to maintain a safe distance from the obstacle, thus avoiding operation and maintenance risks.

[0041] After designing an independent power distribution circuit for each lighting group, according to the predicted illuminance demand value, a fuzzy inference algorithm is used to generate the control strategy for each lighting group. The specific steps include: Generate a meteorological disturbance factor according to the meteorological data, and generate a control matrix according to the real-time data of the substation, the illuminance demand value, and the meteorological disturbance factor; Input the control matrix into a preset fuzzy rule base for fuzzy matching to obtain an output value, and convert the output value into a dimming level and a switch command.

[0042] In this embodiment, a fuzzy logic control is used to generate the control strategy. First, a meteorological disturbance factor is generated according to the real-time meteorological data. The meteorological disturbance factor is used to dynamically correct the natural light. For example, it is set to 1.0 on sunny days and 0.6 on cloudy days. Then, a control matrix with dynamic inputs is generated according to the real-time data of the substation, the illuminance demand value, and the meteorological disturbance factor. Here, the real-time data includes ambient light data, personnel activity data, and energy consumption data. The input variables of the control matrix include illuminance deviation (i.e., the difference between the illuminance demand value and the ambient light data), personnel activity intensity (determined by the personnel density per unit area in the personnel activity data), and meteorological disturbance factor. That is, the control matrix uses the ambient light data and personnel activity data in the real-time data. Then, the input variables of the control matrix are fuzzified and converted into semantic variables that can be processed by the fuzzy logic system.

[0043] For the output variables of the fuzzy logic system, they are defined as the dimming level and the switch command. The dimming level is the percentage of the lamp brightness, and the switch command is the on / off state of the lighting group. At the same time, membership functions are defined. For example, the illuminance deviation is divided into 5 fuzzy sets, and triangular or trapezoidal functions are defined. The personnel activity intensity is divided into 3 fuzzy sets, and the dimming level is divided into 4 levels. The specific membership functions can be flexibly set according to the actual situation and are not limited here.

[0044] Based on the definitions of the above input and output variables and membership functions, logical rules between the input variables and output variables are defined with the input variables as conditions and the output variables as actions based on expert experience or historical data, and a fuzzy rule base is constructed.

[0045] Then, fuzzy inference is performed on the control matrix according to the logical rules in the fuzzy rule base to obtain the corresponding output value. Then, through defuzzification, the output value is converted into an accurate dimming signal and switch command. Specifically, the control matrix is input into the membership function, and the minimum value is taken as the triggering intensity of the rule. The triggering intensities of all rules are combined, and the maximum value is taken as the final fuzzy output, and the centroid method is used for defuzzification to obtain the accurate dimming level: In the formula, S represents the dimming level, n represents the number of rules, μ i represents the triggering intensity of the i-th rule, and S i represents the center value of the dimming level corresponding to the i-th rule.

[0046] Then, a switch command is generated according to the dimming level. For example, if the dimming level is greater than the level threshold, the switch command is on; otherwise, the switch command is off.

[0047] In a preferred embodiment, in order to be able to dynamically switch the control strategy based on multiple scenarios, the present invention dynamically adjusts the control strategy by presetting different scenario modes as constraint conditions for the control strategy. The specific steps include: According to the preset conditions, judge the scenario mode of the substation, and select the corresponding scenario control strategy according to the scenario mode. The scenario modes include daily mode, energy-saving mode, emergency mode, and night mode; Taking the scenario control strategy as a constraint condition, correct the dimming level and the switch command.

[0048] In this embodiment, first, multiple different scenario modes are set, and different scenario control strategies are set for each scenario mode. Among them, the scenario modes include daily mode, energy-saving mode, emergency mode, and night mode. The judgment of the scenario mode includes: determining the daily mode and night mode according to time, determining whether to enter the energy-saving mode according to the sensor data of the power system, and if the power system fails, entering the emergency mode. The priority of these modes is: emergency > night > energy-saving > daily, and the current scenario mode of the substation is determined according to the priority.

[0049] For the daily mode, there are no constraints on the dimming level and switch commands in its scene control strategy; for the energy-saving mode, the upper limit of the dimming level is restricted within the range of the scene setting value, such as the dimming level being less than 50%; for the emergency mode, all lamp groups are forced to be fully on and the dimming level is 100%; for the night mode, only the lamp groups for safety lighting are restricted to be turned on, and at the same time, the upper and lower limits of the dimming level are restricted.

[0050] Based on the above scene modes, when controlling the lamps, first judge the current scene mode of the substation, and select the corresponding scene control strategy according to the scene mode. Then judge whether the dimming level and switch commands obtained through the above fuzzy reasoning are within the scope of the scene control strategy. If so, control the lamp groups according to the dimming level and switch commands obtained through fuzzy reasoning. Otherwise, use the scene control strategy as a constraint condition to correct the dimming level and switch commands. For example, if the dimming level obtained through fuzzy reasoning is 70%, but in the energy-saving mode, the upper limit of the dimming level is 50%, so after correcting the dimming level to 50%, then control the lamp groups.

[0051] Through the control strategy generation method of this embodiment, it can not only ensure the stability of different scenes, but also retain the flexibility of real-time dynamic adjustment, thus achieving the balance of safety, energy saving and comfort of the substation lighting system.

[0052] A method for indoor lighting design of a substation provided in this embodiment realizes high-precision light environment design through the collaborative simulation of multiple parameters and dynamic demand prediction, realizes significant energy conservation and consumption reduction through intelligent grouping control and load balancing optimization, can visually verify key indicators through three-dimensional simulation, solves the defect of non-visualization of traditional two-dimensional design, and improves the efficiency of lighting design through the closed-loop optimization of prediction, simulation and iteration. At the same time, the present invention plans the cable path through an improved A* algorithm and integrates fuzzy logic control to generate an intelligent control strategy, realizing the adaptive dynamic regulation of the lighting system, thereby significantly improving the design accuracy, energy efficiency level and operation and maintenance convenience of the lighting system.

[0053] Please refer to Figure 2 , based on the same inventive concept, a system for indoor lighting design of a substation proposed in the second embodiment of the present invention includes: An illuminance prediction module 10, configured to obtain real-time data, spatial feature data and meteorological data of the substation, and input them into a preset illuminance prediction model to obtain the illuminance demand values of each area of the substation. The real-time data includes ambient light data, personnel activity data, and energy consumption data. The illuminance prediction model is constructed based on a neural network model; The layout design module 20 is configured to input the illumination requirement value and environmental parameters into a preset lamp selection model to obtain an initial lamp layout plan. The lamp selection model is constructed based on the random forest algorithm. The initial lamp layout plan includes an initial lamp type and initial lamp parameters; The layout optimization module 30 is configured to establish a 3D simulation model of the substation according to the substation design drawings and the initial lamp layout plan, and perform iterative simulation on the 3D simulation model according to preset criteria to obtain a lamp layout optimization plan; The circuit design module 40 is configured to perform clustering analysis on the lamp layout optimization plan by using a clustering algorithm to obtain several lamp groups, and set independent power distribution circuits for each lamp group; The strategy generation module 50 is configured to generate a control strategy corresponding to each lamp group by using fuzzy inference according to the illumination requirement value.

[0054] The technical features and technical effects of the indoor lighting design system for substations proposed in the embodiments of the present invention are the same as those of the method proposed in the embodiments of the present invention, and will not be elaborated here. Each module in the above indoor lighting design system for substations can be implemented in whole or in part by software, hardware, and their combinations. The above-mentioned modules can be embedded in the processor of the computer device in the form of hardware or independent of it, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0055] In addition, an embodiment of the present invention also proposes a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0056] Please refer to Figure 3 , the internal structure diagram of the computer device in one embodiment. The computer device can specifically be a terminal or a server. The computer device includes a processor, a memory, a network interface, a display, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements the indoor lighting design method for substations. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0057] Those of ordinary skill in the art can understand that Figure 3 the structure shown in Figure 3 is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computing device may include more or fewer components than those shown in the figure, or combine some components, or have the same component arrangement.

[0058] In addition, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0059] In summary, an indoor lighting design method, system, device and storage medium proposed by an embodiment of the present invention. The method obtains real-time data, spatial feature data and meteorological data of a substation and inputs them into a preset illuminance prediction model to obtain the illuminance demand values of each area of the substation. The real-time data includes ambient light data, personnel activity data, and energy consumption data. The illuminance prediction model is constructed based on a neural network model; the illuminance demand values and environmental parameters are input into a preset lamp selection model to obtain an initial lamp layout plan. The lamp selection model is constructed based on a random forest algorithm. The initial lamp layout plan includes the initial lamp type and initial lamp parameters; according to the substation design drawings and the initial lamp layout plan, a three-dimensional simulation model of the substation is established, and the three-dimensional simulation model is iteratively simulated according to a preset standard to obtain an optimized lamp layout plan; a clustering algorithm is used to perform clustering analysis on the optimized lamp layout plan to obtain several lamp groups, and an independent power distribution circuit is set for each lamp group; according to the illuminance demand values, fuzzy reasoning is used to generate a control strategy corresponding to each lamp group. The present invention realizes high-precision light environment design through the collaborative simulation of multiple parameters and dynamic demand prediction, and realizes the adaptive dynamic regulation of lighting fixtures by improving the A* algorithm to plan the cable path and integrating fuzzy logic control to generate an intelligent control strategy. The present invention significantly improves the design accuracy, energy efficiency level and operation and maintenance convenience of the lighting system by integrating dynamic prediction, multi-objective optimization and intelligent control technologies.

[0060] Each embodiment in this specification is described in a progressive manner. For the parts that are the same or similar in each embodiment, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for the relevant parts, reference can be made to the description of the method embodiment. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as within the scope described in this specification.

[0061] The above embodiments only represent several preferred embodiments of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and substitutions can be made, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for indoor lighting design of a substation, characterized in that, Including: Obtain the real-time data, spatial feature data, and meteorological data of the substation, and input them into a preset illuminance prediction model to obtain the illuminance demand values of each area of the substation. The real-time data includes ambient light data, personnel activity data, and energy consumption data. The illuminance prediction model is constructed based on a neural network model; Input the illuminance demand values and environmental parameters into a preset lamp selection model to obtain an initial lamp layout plan. The lamp selection model is constructed based on a random forest algorithm. The initial lamp layout plan includes the initial lamp type and initial lamp parameters; According to the substation design drawings and the initial lamp layout plan, establish a 3D simulation model of the substation, and perform iterative simulation on the 3D simulation model according to preset standards to obtain an optimized lamp layout plan; Use a clustering algorithm to perform clustering analysis on the optimized lamp layout plan to obtain several lamp groups, and set independent power distribution circuits for each lamp group; According to the illuminance demand values, use fuzzy inference to generate a control strategy corresponding to each lamp group.

2. The indoor lighting design method for a substation according to claim 1, characterized in that, The step of performing iterative simulation on the 3D simulation model according to preset standards to obtain an optimized lamp layout plan includes: Taking the minimization of the sum of squared errors between each simulation standard and the corresponding preset standard as the objective function, and using the lamp layout parameters as optimization variables, perform iterative simulation on the 3D simulation model to obtain an optimized lamp layout plan. The lamp layout parameters include lamp spacing, installation height, and the number of lamps; Among them, the objective function is expressed by the following formula: Where, E sim represents the simulated value of average illuminance, and E std represents the standard value of average illuminance. UGR sim represents the simulated value of glare, and UGR std represents the standard value of glare. U sim represents the simulated value of illuminance uniformity, and U std represents the standard value of illuminance uniformity. LPD sim represents the simulated value of lighting power density, and LPD std represents the standard value of lighting power density.

3. The indoor lighting design method for a substation according to claim 1, characterized in that, The step of using a clustering algorithm to perform clustering analysis on the optimized lamp layout plan to obtain several lamp groups, and setting independent power distribution circuits for each lamp group includes: Taking each lamp as a node, calculate the edge weights between adjacent nodes according to power correlation, spatial distance, and usage frequency synchronization. According to the nodes and the edge weights, construct a lamp structure diagram; Use a modularity optimization algorithm to perform community division on the lamp structure diagram to obtain several first communities; According to the community power difference and community spatial distance, merge the first communities to obtain several second communities, and obtain the corresponding lamp groups according to the second communities; According to the total power of the lamps in each lamp group and the preset capacity safety margin, determine the circuit capacity of the lamp group, and use a path planning algorithm to perform cable path planning to obtain the lamp cable routing layout of the lamp group.

4. The indoor lighting design method for a substation according to claim 3, characterized in that The edge weights are expressed by the following formula: where w ij represents the edge weight between node i and node j, P i and P j represent the power of node i and node j respectively, D ij represents the Euclidean distance between node i and node j, f i and f j represent the usage frequencies of node i and node j respectively, P max represents the maximum power, D max represents the maximum distance, f max represents the maximum frequency, α represents the power weight, β represents the distance weight, and γ represents the frequency weight.

5. The indoor lighting design method for a substation according to claim 3, characterized in that, The step of using a path planning algorithm to perform cable path planning to obtain the lamp cable routing layout of the lamp group includes: Construct an actual path cost function according to the edge weights of the second community corresponding to each lamp group; Construct a heuristic function according to the Euclidean distance between the nodes of the second community corresponding to each lamp group and the obstacle data around the nodes; According to the actual path cost function and the heuristic function, obtain the total cost function; According to the total cost function, use the A* algorithm to perform iterative calculations to obtain the shortest path corresponding to each lamp group; According to the shortest path, obtain the lamp cable routing layout of each lamp group; Among them, the actual path cost function is represented by the following formula: where g(n) represents the actual path cost function of the current node n, and w ij represents the edge weight between node i and node j, and Path(start→n) represents the path from the start node start to the current node n; The heuristic function is represented by the following formula: Where h(n) represents the heuristic function of the current node n, x n and y n represent the coordinate values of the current node n, x g and y g represent the coordinate values of the target node g, λ represents the obstacle coefficient, and ρ(n) represents the obstacle density around the current node n; The total cost function is represented by the following formula: In the formula, f(n) represents the total cost function of the current node n.

6. The indoor lighting design method for a substation according to claim 1, characterized in that, The step of generating a control strategy corresponding to each lamp group by using fuzzy inference according to the illumination demand value includes: Generating a meteorological disturbance factor according to meteorological data, and generating a control matrix according to the real-time data of the substation, the illumination demand value, and the meteorological disturbance factor; Inputting the control matrix into a preset fuzzy rule base for fuzzy matching to obtain an output value, and converting the output value into a dimming level and a switching instruction.

7. The indoor lighting design method for a substation according to claim 6, characterized in that After the step of converting the output value into a dimming level and a switching instruction, it further includes: Judging the scene mode of the substation according to preset conditions, and selecting a corresponding scene control strategy according to the scene mode, where the scene mode includes a daily mode, an energy-saving mode, an emergency mode, and a night mode; Taking the scene control strategy as a constraint condition to correct the dimming level and the switching instruction.

8. An indoor lighting design system for a substation, characterized in that, It includes: An illumination prediction module, configured to obtain the real-time data, spatial feature data, and meteorological data of the substation, and input them into a preset illumination prediction model to obtain the illumination demand values of each area of the substation. The real-time data includes ambient light data, personnel activity data, and energy consumption data. The illumination prediction model is constructed based on a neural network model; A layout design module, configured to input the illumination demand value and environmental parameters into a preset lamp selection model to obtain an initial lamp layout plan. The lamp selection model is constructed based on a random forest algorithm. The initial lamp layout plan includes an initial lamp type and initial lamp parameters; A layout optimization module, configured to establish a three-dimensional simulation model of the substation according to the substation design drawing and the initial lamp layout plan, and perform iterative simulation on the three-dimensional simulation model according to a preset standard to obtain a lamp layout optimization plan; A circuit design module, configured to perform clustering analysis on the lamp layout optimization plan by using a clustering algorithm to obtain several lamp groups, and set independent power distribution circuits for each lamp group; A strategy generation module, configured to generate a control strategy corresponding to each lamp group by using fuzzy inference according to the illumination demand value.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 7.

Images