Smart city lighting system adaptive brightness adjustment method, medium and system

Abstract

The invention provides a smart city lighting system adaptive brightness adjustment method, medium and system, and belongs to the technical field of lighting brightness adjustment processing, and the method comprises the steps: obtaining the partition information of a smart city lighting system, building a lighting system matrix, and obtaining the partition information of the smart city lighting system; and the ambient illuminance, the people flow density, the traffic flow density and the weather condition data of the lighting units of all the partitions are collected in real time. Similarity analysis is carried out based on real-time data, and a multi-objective optimization model is established, including illumination demand satisfaction, energy consumption minimization and illumination uniformity. And performing optimization calculation on the partitions by using the optimization model to obtain an optimized partition scheme, and constructing a second multi-objective optimization model including energy consumption minimization, illumination uniformity maximization and partition number minimization. And establishing a brightness adaptive model comprising a fuzzy controller for the optimized minimum partition, and calculating the target brightness of the illumination unit in each partition according to the real-time data, thereby solving the problem that the brightness cannot be adaptively adjusted according to the environment in the prior art.

Description

Technical Field

[0001] The present invention belongs to the technical field of lighting brightness adjustment processing, and specifically, relates to a method, medium and system for adaptive brightness adjustment of a smart city lighting system. Background Art

[0002] With the continuous advancement of urbanization, the construction of smart cities has become one of the important directions of current social and economic development. As an important part of smart city construction, the smart lighting system is the core element of urban nighttime lighting. Its automation and intelligence directly affect the quality of urban nighttime lighting and energy consumption efficiency. At present, most of the existing smart city lighting systems use timing control or simple light intensity detection feedback mechanism to adjust the lighting brightness. Although this method can achieve a certain degree of automatic control, it cannot respond quickly to complex urban environmental factors such as dynamic changes in pedestrian density, vehicle density, etc., and often cannot meet the lighting needs of different areas and different time periods, resulting in poor overall lighting effects and high energy consumption. Summary of the invention

[0003] In view of this, the present invention provides a method, medium and system for adaptive brightness adjustment of a smart city lighting system, which can solve the technical problem that the existing automatic adjustment of urban lighting brightness cannot respond quickly to complex urban environmental factors such as dynamic changing factors such as pedestrian density and vehicle density.

[0004] The present invention is achieved in that:

[0005] A first aspect of the present invention provides a method for adaptively adjusting brightness of a smart city lighting system, comprising the following steps:

[0006] S10, obtaining partition information of the smart city lighting system, wherein each partition includes a plurality of lighting units;

[0007] S20. Establish a smart city lighting system matrix;

[0008] S30, collecting real-time operating data of all lighting units in each of the partitions;

[0009] S40, performing similarity analysis on each lighting unit in each of the partitions based on the real-time operation data, and establishing a first multi-objective optimization model including lighting demand satisfaction, energy consumption minimization, and lighting uniformity;

[0010] S50, using the first multi-objective optimization model to perform optimization calculation on each of the partitions, calculate the control similarity matrix between the lighting units in the partition, construct lighting unit group division constraint conditions based on the control similarity matrix, and obtain an optimized partitioning scheme;

[0011] S60. According to the optimized partitioning scheme, a second multi-objective optimization model including minimizing the energy consumption of the lighting unit, maximizing the lighting uniformity, and minimizing the number of partitions is established, and a plurality of extremely small partitions are obtained by solving the model, wherein the control similarity of the lighting units in the extremely small partitions is greater than a first preset threshold, and the control similarity with the adjacent lighting units is less than a second preset threshold;

[0012] S70, establishing and fitting a brightness adaptive model including a fuzzy controller for each of the extremely small partitions, setting the input variables of the fuzzy controller to the ambient light illumination, crowd density, vehicle density, and weather conditions, and setting the output variable to the target brightness of the lighting unit; the brightness adaptive model includes a fuzzy controller, a brightness adaptive equation group, and a lightweight neural network;

[0013] S80. Calculate the target brightness value of the lighting unit in each of the extremely small partitions based on the brightness adaptive model, and send a brightness adjustment instruction to all the lighting units in each of the extremely small partitions to adjust the actual brightness of the lighting units to the target brightness value.

[0014] On the basis of the above technical solution, the adaptive brightness adjustment method of a smart city lighting system of the present invention can also be improved as follows:

[0015] Among them, the lighting unit refers to an intelligent street lamp monomer composed of a controller, an LED light source, and a sensor. The sensor includes a light sensor, a pedestrian flow detection sensor, and a vehicle flow detection sensor.

[0016] Furthermore, the elements of the smart city lighting system matrix include lighting unit identification, lighting unit geographic location information, lighting unit brightness parameters, lighting unit energy consumption parameters, and lighting unit operating status parameters.

[0017] Furthermore, the real-time operation data includes ambient light illumination data, pedestrian density data, vehicle density data, and weather condition data.

[0018] Furthermore, the fuzzy controller is used to realize the mapping relationship from environmental factors to lighting control strategies. The input is ambient light illumination data, pedestrian density data, vehicle density data, and weather condition data. The output is a reference brightness value of the lighting unit. The structure adopts a two-layer fuzzy reasoning system. The specific structure is that the first-layer fuzzy system processes the ambient light illumination and weather condition data to obtain an environmental evaluation value, and the second-layer fuzzy system processes the pedestrian density, vehicle density data and the environmental evaluation value to obtain a reference brightness value.

[0019] Furthermore, the brightness adaptive equation group includes a lighting compensation equation, a pedestrian flow response equation, a vehicle flow response equation and an energy consumption balance equation.

[0020] Furthermore, the illumination compensation equation is used to dynamically compensate the reference brightness value, the input is the ambient light illumination data and the reference brightness value, and the output is the compensated target brightness value;

[0021] The crowd response equation is used to adjust the lighting brightness according to the crowd density, the input is the crowd density data and the target brightness value after compensation, and the output is the target brightness value after the crowd response;

[0022] The traffic flow response equation is used to adjust the lighting brightness according to the traffic flow density, the input is the traffic flow density data and the target brightness value after the pedestrian flow response, and the output is the target brightness value after the traffic flow response;

[0023] The energy consumption balance equation is used to balance the lighting effect and energy consumption. The input is the target brightness value after the traffic flow response and the energy consumption parameter of the lighting unit, and the output is the final target brightness value.

[0024] Furthermore, the lightweight neural network is used to optimize the parameters of the brightness adaptation model. The input is historical operating data and current operating data, and the output is the optimization parameters of the brightness adaptation equation group. The structure adopts a three-layer feedforward neural network. The specific structure is that the number of input layer nodes is 8, the number of hidden layer nodes is 16, the number of output layer nodes is 4, and the ReLU activation function is adopted.

[0025] Furthermore, the step of obtaining the first preset threshold value specifically includes: calculating the average control similarity of the lighting units in the extremely small partition based on the historical operation data, selecting the extremely small partition that meets the lighting uniformity requirements and has a stable control response, and calculating the mean value of the average control similarity as the first preset threshold value; the default value is 0.85;

[0026] The step of obtaining the second preset threshold specifically includes: calculating the lighting unit control similarity between adjacent minimal partitions based on historical operation data, selecting a partition scheme with good lighting effect and low energy consumption, and calculating the average of the control similarity between adjacent partitions as the second preset threshold; the default value is 0.3;

[0027] The smart city lighting system matrix is ​​specifically expressed as follows:

[0028]

[0029] In the formula, m ij represents the parameter vector of the jth lighting unit in the ith partition, k represents the number of partitions, and n represents the number of lighting units in each partition; m ij =[ID ij ,x ij ,y ij ,l ij ,P ij ,S ij ], where IDij is the lighting unit identification code, x ij ,y ij is the geographic coordinate, L ij is the brightness value, P ij is the energy consumption value, S ij The running status value.

[0030] The similarity analysis is specifically expressed as follows:

[0031]

[0032] In the formula, sim ij is the control similarity between lighting units i and j; d ij is the geographical distance between the two lighting units; d0 is the distance attenuation coefficient, the default value is 100 meters; θ ij is the azimuth difference between the two lighting units; L i ,L j is the brightness value; P i ,P j is the energy consumption value; α1, α2, α3, α4 are weight coefficients, and satisfy The default values ​​are 0.3, 0.2, 0.3, 0.2 respectively.

[0033] The first multi-objective optimization model is specifically expressed as follows:

[0034] minF1=(f1,f2,f3);

[0035]

[0036] Where f1 is the lighting demand satisfaction objective function; f2 is the energy consumption objective function; f3 is the lighting uniformity objective function; L i is the brightness value of the i-th lighting unit; L req is the lighting demand value; P i is the energy consumption value of the i-th lighting unit; N i is the neighborhood set of the i-th lighting unit.

[0037] The second multi-objective optimization model is specifically expressed as follows:

[0038] minF2=(g1,g2,g3);

[0039]

[0040] g3=K;

[0041] Where g1 is the objective function for minimizing the energy consumption of the partition; g2 is the objective function for the uniformity of lighting within the partition; g3 is the objective function for minimizing the number of partitions; K is the total number of partitions; Ck is the set of lighting units of the kth partition.

[0042] The illumination compensation equation is specifically expressed as follows:

[0043]

[0044] Where, L c is the target brightness value after compensation; L b is the reference brightness value; E is the ambient light illumination; E0 is the reference light illumination, the default value is 10000 lux; β1, β2 are compensation coefficients; ω is the angular frequency of day and night changes; t is time; φ is the phase difference; ∈1 is the compensation error term.

[0045] The crowd flow response equation is specifically expressed as follows:

[0046]

[0047] Where, L p is the target brightness value after the flow of people responds; D p is the flow density; D p0 is the base crowd density, the default value is 10 people / 100 square meters; γ1, γ2 are response coefficients; is the change rate of human flow density; ∈2 is the response error term.

[0048] The vehicle flow response equation is specifically expressed as follows:

[0049]

[0050] Where, L v is the target brightness value after the traffic flow responds; D v is the traffic density; D v0 is the benchmark traffic density, the default value is 5 vehicles / 100 meters; δ1, δ2, δ3 are response coefficients; is the rate of change of traffic density; θ v is the main direction angle of the traffic flow; ∈3 is the response error term.

[0051] The energy consumption balance equation is specifically expressed as follows:

[0052]

[0053] Where, L f is the final target brightness value; P is the current energy consumption; P0 is the benchmark energy consumption; η1, η2, η3 are balance coefficients; is the energy consumption change rate of the i-th lighting unit; ∈4 is the balance error term.

[0054] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are run in a computer, they are used to execute the above-mentioned method for adaptive brightness adjustment of a smart city lighting system.

[0055] A third aspect of the present invention provides a smart city lighting system adaptive brightness adjustment system, which includes the above-mentioned computer-readable storage medium.

[0056] Compared with the prior art, the method, medium and system for adaptive brightness adjustment of a smart city lighting system provided by the present invention can make full use of lighting equipment in various areas of the city, combine real-time perception data such as ambient light intensity, crowd density, and vehicle density, and use a multi-objective optimization algorithm to perform partition optimization and adaptive brightness control, so as to meet the lighting needs of different areas and different time periods to the greatest extent, and balance the lighting effect and energy consumption, thus realizing adaptive control of the smart city lighting system. The beneficial effects of the present invention are:

[0057] 1. Make full use of lighting equipment in various areas of the city, combine real-time perceived environmental factor data, and use multi-objective optimization algorithms for zoning and brightness control, so as to quickly respond to dynamic changes in the urban environment and meet the lighting needs of different areas and time periods.

[0058] 2. The adaptive brightness adjustment model that combines fuzzy control, brightness adaptive equations and lightweight neural network can accurately grasp the complex relationship between environmental factors and lighting needs, realize targeted adaptive dimming control, and improve lighting effects and energy utilization efficiency.

[0059] 3. By further optimizing the zoning scheme, the lighting units within the zones are ensured to be highly coordinated, while the number of zones is minimized, further improving the overall energy efficiency of the system.

[0060] In summary, the present invention solves the technical problem that the existing automatic adjustment of urban lighting brightness cannot quickly respond to complex urban environmental factors such as dynamic changing factors such as pedestrian density and vehicle density. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flow chart of the method provided by the present invention;

[0062] Figure 2 This is the result diagram of the similarity analysis of the lighting unit;

[0063] Figure 3 Performance analysis diagram for multi-objective optimization models;

[0064] Figure 4 It is the response surface diagram of the adaptive brightness adjustment model;

[0065] Figure 5 Optimize the result graph for very small partitions;

[0066] Figure 6 Dynamic response curve diagram for lighting control. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0068] like Figure 1 FIG. 1 is a flow chart of a method for adaptive brightness adjustment of a smart city lighting system provided by the present invention. The method comprises the following steps:

[0069] S10, obtaining partition information of the smart city lighting system, where each partition includes multiple lighting units;

[0070] S20. Establish a smart city lighting system matrix;

[0071] S30, collecting real-time operating data of all lighting units in each partition;

[0072] S40, performing similarity analysis on each lighting unit in each partition based on real-time operation data, and establishing a first multi-objective optimization model including lighting demand satisfaction, energy consumption minimization, and lighting uniformity;

[0073] S50, using the first multi-objective optimization model to perform optimization calculation on each partition, calculate the control similarity matrix between the lighting units in the partition, construct the lighting unit group division constraint conditions based on the control similarity matrix, and obtain an optimized partition scheme;

[0074] S60, according to the optimized partitioning scheme, establish a second multi-objective optimization model including minimizing the energy consumption of the lighting unit, maximizing the lighting uniformity, and minimizing the number of partitions, and solve to obtain a plurality of extremely small partitions, wherein the control similarity of the lighting units in the extremely small partitions is greater than a first preset threshold, and the control similarity with the adjacent lighting units is less than a second preset threshold;

[0075] S70, establishing and fitting a brightness adaptive model including a fuzzy controller for each extremely small partition, setting the input variables of the fuzzy controller to the ambient light illumination, crowd density, vehicle density, and weather conditions, and setting the output variable to the target brightness of the lighting unit; the brightness adaptive model includes a fuzzy controller, a brightness adaptive equation group, and a lightweight neural network;

[0076] S80. Calculate the target brightness value of the lighting unit in each extremely small partition based on the brightness adaptive model, and send a brightness adjustment instruction to all lighting units in each extremely small partition to adjust the actual brightness of the lighting unit to the target brightness value.

[0077] The specific implementation methods of the above steps are described in detail below:

[0078] The specific implementation of step S10 is as follows:

[0079] The system first obtains the partition information D = {d1, d2, ..., d K}, where K represents the total number of partitions. Each partition d k It includes a plurality of lighting units L = {l1,l2,…,l N}, where N represents the number of lighting units in each partition. Each lighting unit l n It consists of a controller, an LED light source and a sensor. The sensor includes a light sensor S l , Crowd Detection Sensor S p and traffic detection sensor S v By obtaining relevant attribute information of partitions and lighting units, a data basis is provided for subsequent optimization calculations and decision-making.

[0080] The specific implementation of step S20 is as follows:

[0081] According to the partition and lighting unit information obtained in step S10, the smart city lighting system matrix M is established. The element m of the matrix M is ij Contains the lighting unit's ID ij 、Geographic coordinates (x ij ,y ij ), brightness parameter L ij , Energy consumption parameter P ij and operating status parameter S ij , that is, m ij =[ID ij ,x ij ,y ij ,L ij ,P ij ,S ij ] Construct this matrix representation containing the key parameters of the lighting system Provide a data basis for subsequent optimization calculations and decision-making.

[0082] The specific implementation of step S30 is as follows:

[0083] The system collects the operating data of all lighting units in each partition in real time, including ambient light illumination data E, crowd density data D p , Traffic density data Dv and weather condition data W. Where E represents the ambient light intensity, D p Denotes the density of people flow, D v represents the traffic density, and W represents the weather conditions. The key environmental factor data that drives the adaptive dimming of the lighting system is obtained to provide a basis for subsequent similarity analysis and optimization decisions.

[0084] The specific implementation of step S40 is as follows:

[0085] Based on the real-time operation data obtained in step S30, similarity analysis is performed on the lighting units in each partition. The similarity analysis uses the formula:

[0086]

[0087] Among them, d ij represents the geographical distance between lighting units i and j, θ ij represents the azimuth difference between lighting units i and j, L i ,L j Represents the brightness value of lighting unit i and j, P i ,P j Respectively represent the energy consumption values ​​of lighting units i and j, α1, α2, α3, α4 are weight coefficients, and satisfy The default values ​​are 0.3, 0.2, 0.3, 0.2 respectively.

[0088] Based on the similarity analysis results, the first multi-objective optimization model including lighting demand satisfaction, energy consumption minimization and lighting uniformity is established:

[0089] minF1=(f1,f2,f3);

[0090]

[0091] Among them, f1 is the lighting demand satisfaction objective function, f2 is the energy consumption objective function, f3 is the lighting uniformity objective function, L i is the brightness value of the i-th lighting unit, L req is the lighting demand value, P i is the energy consumption value of the i-th lighting unit, N i is the neighborhood set of the i-th lighting unit.

[0092] Evaluate and model the operating status of lighting units in each partition of the lighting system to provide a basis for the next optimization decision.

[0093] The specific implementation of step S50 is as follows:

[0094] Using the first multi-objective optimization model established in step S40, each partition is optimized and calculated to obtain a control similarity matrix between lighting units in each partition. Based on the control similarity matrix, the system constructs lighting unit group division constraints and obtains an optimized partitioning scheme.

[0095] Specifically, the element c of the control similarity matrix C ij Defined as:

[0096] c ij =sim ij ;

[0097] Among them, sim ij is the control similarity between lighting units i and j calculated in step S40.

[0098] The constraints are: when c ij >θ1, lighting units i and j belong to the same group; when c ij <θ2, lighting units i and j belong to different groups. θ1 and θ2 are the first preset threshold and the second preset threshold, respectively, and the default values ​​are 0.85 and 0.3.

[0099] According to the operating characteristics of the lighting unit, the system partitioning scheme is optimized to improve the coordination of lighting control.

[0100] The specific implementation of step S60 is as follows:

[0101] According to the optimized partitioning scheme obtained in step S50, the system establishes a second multi-objective optimization model including minimizing the energy consumption of the lighting unit, maximizing the lighting uniformity and minimizing the number of partitions:

[0102] minF2=(g1,g2,g3);

[0103]

[0104] g3=K;

[0105] Among them, g1 is the objective function for minimizing the energy consumption of the partition, g2 is the objective function for the uniformity of lighting within the partition, g3 is the objective function for minimizing the number of partitions, K is the total number of partitions, C k is the set of lighting units of the kth partition.

[0106] The system solves the optimization model and obtains multiple minimal partitions that meet the lighting needs. This step further optimizes the partition scheme to ensure high coordination of lighting units within the partition, while minimizing the number of partitions and improving system energy efficiency.

[0107] The specific implementation of step S70 is as follows:

[0108] For each tiny partition, the system builds and fits a brightness adaptive model including a fuzzy controller. The model includes a fuzzy controller, a brightness adaptive equation group and a lightweight neural network.

[0109] The input variables of the fuzzy controller are the ambient light intensity E, the crowd density D p , traffic density D v and weather conditions W, the output variable is the reference brightness value L of the lighting unit b The fuzzy controller adopts a two-layer fuzzy inference system. The first layer processes the ambient light illumination and weather condition data to obtain the environmental evaluation value E′, and the second layer processes the pedestrian density and vehicle density data and the environmental evaluation value to obtain the reference brightness value L b .

[0110] The brightness adaptation equations include:

[0111] Light compensation equation:

[0112] Crowd response equation:

[0113] Traffic flow response equation:

[0114] Energy balance equation:

[0115] Among them, E0 is the reference illuminance, D p0 ,D v0 are the benchmark pedestrian and vehicle flow densities, β1, β2, δ1, γ2, δ1, δ2, δ3, η1, η2, η3 are the corresponding response coefficients, and ∈1, ∈2, ∈3, ∈4 are error terms.

[0116] The illumination compensation equation mainly considers the impact of two factors on target brightness:

[0117] Ambient light intensity E: When the ambient light intensity is strong, the lighting system can appropriately reduce the output brightness to achieve energy saving. Here, an exponential decay function is used. To describe this relationship, E0 is the reference illuminance. When E is much smaller than E0, this term approaches 1, which means that not much compensation is needed when the ambient light is weak; when E is much larger than E0, this term approaches 0, which means that the output brightness needs to be greatly reduced when the ambient light is strong.

[0118] Day and night variation: Since sunlight varies from day to night, the lighting system needs to dynamically adjust the output brightness to adapt to this pattern. Here, the sine function sin(ωt+φ) is used to describe this periodic change, where ω is the angular frequency of the day and night variation and φ is the phase difference. This allows the system to reduce the brightness during the day and increase the brightness at night, thereby achieving better energy-saving effects.

[0119] In addition, the equation also contains an error term ∈1 to compensate for the error of the model itself.

[0120] The crowd response equation takes into account the crowd density D p Impact on target brightness:

[0121] Crowd density value: When the crowd density D p Less than the reference density D p0 When D p Greater than D p0 When the brightness is increased, it is necessary to appropriately increase the brightness to meet people's visual needs. Here, the hyperbolic tangent function is used To describe this nonlinear relationship, a smooth transition can be achieved.

[0122] Crowd density change rate: In addition to considering the current crowd density, the system also needs to adjust the brightness according to the changing trend of the crowd density. When the crowd density increases rapidly, the brightness needs to be increased in advance; when the crowd density decreases rapidly, the brightness needs to be reduced in time. The derivative term is used here To describe this dynamic change.

[0123] Similarly, the equation also contains an error term ∈2.

[0124] The traffic flow response equation takes into account the traffic density D v and the traffic direction θ v Impact on target brightness:

[0125] Traffic density value: Similar to the response of pedestrian flow, when the traffic density D v Less than the reference density D v0 When D v Greater than D v0 When the brightness is increased, it is necessary to appropriately increase the brightness to meet the visual needs of the driver. The hyperbolic tangent function is also used here. to describe this nonlinear relationship.

[0126] Traffic density change rate: It is also necessary to consider the dynamic changes of traffic density and use the derivative term to describe this trend.

[0127] Traffic flow direction: In addition to traffic density, traffic flow direction θ vIt will also affect the output of the lighting system. When the traffic is mainly concentrated in a certain direction, it is necessary to appropriately enhance the lighting in that direction to improve driving safety. Here we use the exponential function to describe this directional effect.

[0128] Similarly, the equation also contains an error term ∈3.

[0129] The energy balance equation takes into account the total energy consumption P of the system and the energy consumption change rate of each lighting unit. Impact on target brightness:

[0130] Total energy consumption: Although it is necessary to meet the lighting needs of pedestrian and vehicle traffic, energy consumption must also be taken into account. Here, the power function is used To describe the relationship between total energy consumption and target brightness, η2 is a suitable exponent. When p is much smaller than p0, this term approaches 0, which means that the brightness can be appropriately increased; when p is much larger than p0, this term approaches 1, which means that the brightness needs to be appropriately reduced.

[0131] Energy consumption change rate: In addition to the total energy consumption, the energy consumption change rate of each lighting unit also needs to be taken into account. When the unit energy consumption changes dramatically, it may cause an imbalance in the overall energy consumption, and the brightness needs to be adjusted appropriately. Here, the square sum term is used To describe this dynamic change feature.

[0132] Similarly, the equation also contains an error term ∈4.

[0133] In general, these four equations fully consider the impact of many factors such as ambient light, crowd density, traffic density, traffic direction, and energy consumption on the output brightness of the lighting system. By establishing such a multi-objective optimization model, it is possible to meet lighting needs, minimize energy consumption, and maximize lighting uniformity. Compared with traditional fixed value or simple feedback control, this adaptive adjustment method is more intelligent and efficient, and can greatly improve the performance of smart city lighting systems.

[0134] The lightweight neural network is used to optimize the parameters of the brightness adaptive model. Its structure is a three-layer feedforward neural network with 8 input layer nodes, 16 hidden layer nodes, and 4 output layer nodes. The ReLU activation function is used. The use of lightweight neural networks can reduce hardware consumption and improve computing efficiency at the expense of some accuracy. This step establishes a highly adaptable lighting brightness adjustment model to achieve adaptive dimming control.

[0135] The specific implementation of step S80 is as follows:

[0136] Based on the brightness adaptive model established in step S70, the system calculates the target brightness value L of the lighting unit in each minimal partition. f, and sends brightness adjustment instructions to each lighting unit to adjust the actual brightness to the target value. This step applies the results of the optimization calculation to the actual lighting system to achieve intelligent adaptive brightness control.

[0137] In addition, the first preset threshold value θ1 and the second preset threshold value θ2 are obtained as follows:

[0138] The first preset threshold θ1 is calculated based on the historical operation data to calculate the average control similarity of the lighting units in the smallest partition, and the smallest partition that meets the lighting uniformity requirements and has stable control response is selected, and the mean of the average control similarity is counted as the first preset threshold, and the default value is 0.85.

[0139] The second preset threshold θ2 is calculated based on the historical operation data to calculate the control similarity of the lighting units between adjacent minimal partitions, and a partition scheme with good lighting effects and low energy consumption is selected, and the mean value of the control similarity between adjacent partitions is counted as the second preset threshold, and the default value is 0.3.

[0140] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are run in a computer, they are used to execute the above-mentioned method for adaptive brightness adjustment of a smart city lighting system.

[0141] A third aspect of the present invention provides a smart city lighting system adaptive brightness adjustment system, which includes the above-mentioned computer-readable storage medium.

[0142] Specifically, the principle of the present invention is:

[0143] First, the system obtains the partition information of the smart city lighting system and the parameter matrix of each lighting unit, providing a data basis for subsequent optimization calculations and decision-making.

[0144] Secondly, based on the real-time collected data such as ambient light, pedestrian density, and vehicle density, the system conducts similarity analysis on each lighting unit, establishes a multi-objective optimization model that includes lighting demand satisfaction, energy consumption minimization, and lighting uniformity, and optimizes the zoning scheme. The purpose of this step is to evaluate the operating status of each lighting unit and provide a basis for optimization decisions.

[0145] Subsequently, the system further optimizes the zoning scheme to ensure high coordination of lighting units within the partitions, while minimizing the number of partitions to improve the overall energy efficiency performance of the system.

[0146] Finally, for each optimized partition, the system establishes an adaptive brightness adjustment model including a fuzzy controller, a brightness adaptive equation group, and a lightweight neural network. The fuzzy controller outputs the reference brightness according to environmental factors, the brightness adaptive equation group realizes dynamic compensation and response to the reference brightness, and the lightweight neural network is used to optimize the model parameters. The purpose of this step is to establish a highly adaptable lighting brightness adjustment model for different partitions and realize intelligent adaptive dimming control.

[0147] In summary, the adaptive brightness adjustment method of the smart city lighting system of the present invention makes full use of the lighting equipment in various areas of the city, combines real-time perceived environmental factor data, and adopts a multi-objective optimization algorithm to perform partition optimization and adaptive brightness control. It can quickly respond to dynamic changes in the urban environment, effectively balance lighting effects and energy consumption, and realize adaptive control of the smart city lighting system, which greatly reduces energy consumption while improving lighting quality.

[0148] An embodiment of a specific application scenario of the present invention is provided below:

[0149] The smart city lighting system in this embodiment consists of a central control center and multiple lighting units distributed in each block, and each lighting unit is equipped with a light sensor, a pedestrian flow detection sensor, and a vehicle flow detection sensor.

[0150] The central control center established the smart city lighting system matrix M as shown in Table 1 based on the system partition information and the parameters of each lighting unit:

[0151] Table 1 Smart city lighting system matrix M data table;

[0152]

[0153] On the basis of establishing the matrix M, the system collects the ambient light illumination E and the crowd density D of the lighting units in each partition in real time. p , traffic density D v And weather conditions W data, the specific data are shown in Table 2:

[0154] Table 2 Real-time operation data

[0155]

[0156] Based on the above data, the system first performs similarity analysis on each lighting unit using the formula:

[0157]

[0158] Among them, d ij is the geographical distance between the two lighting units, θ ij is the azimuth difference between the two lighting units, Li ,L j is the brightness value, P i ,P j are energy consumption values, α1=0.3, α2=0.2, α3=0.3, α4=0.2.

[0159] The similarity analysis results show that the control similarity of lighting unit 1 and unit 2 in partition 1 is sim 12 =0.82, the control similarity of lighting unit 1 and unit 2 in partition 2 is sim 12 =0.78, the control similarity sim between lighting unit 1 and unit 2 in partition K 12 =0.88.

[0160] like Figure 2 As shown, the similarity analysis results of the lighting units are displayed, including the changing trends of the four components of distance similarity, azimuth similarity, brightness similarity and energy consumption similarity, as well as the total similarity curve obtained by their weighted combination.

[0161] Based on the similarity analysis results, the system established the first multi-objective optimization model including lighting demand satisfaction, energy consumption minimization and lighting uniformity:

[0162] minF1=(f1,f2,f3);

[0163]

[0164]

[0165] Among them, L i is the brightness value of the i-th lighting unit, L req = 90% is the lighting demand value, P i is the energy consumption value of the i-th lighting unit, N i is the neighborhood set of the i-th lighting unit.

[0166] like Figure 3 As shown in the figure, the performance analysis of the multi-objective optimization model is demonstrated, and the relationship between the lighting brightness ratio, energy consumption and the optimization target value is displayed through a three-dimensional surface. The system uses this optimization model to calculate each partition and obtains the optimized partition scheme. On this basis, the system further establishes a second multi-objective optimization model that includes minimizing the energy consumption of the lighting unit, maximizing the lighting uniformity and minimizing the number of partitions:

[0167] minF2=(g1,g2,g3);

[0168]

[0169] g3=K;

[0170] Among them, C k is the set of lighting units of the kth partition.

[0171] By solving the optimization model, the system finally divides the original K partitions into K′=15 minimal partitions, satisfying the following conditions:

[0172] 1) The average control similarity of the lighting units in the smallest partition is greater than the first preset threshold value θ1=0.85;

[0173] 2) The control similarity of the lighting units between adjacent minimal partitions is less than the second preset threshold θ2=0.3.

[0174] For each minimal partition, the system establishes and fits an adaptive brightness adjustment model including a fuzzy controller, a brightness adaptive equation group, and a lightweight neural network. The input of the fuzzy controller is the ambient light intensity E, the crowd density D p , traffic density D v and weather conditions W, the output is the reference brightness value L of the lighting unit b The brightness adaptation equations include:

[0175] Light compensation equation:

[0176]

[0177] Crowd response equation:

[0178]

[0179] Traffic flow response equation:

[0180]

[0181] Energy balance equation:

[0182]

[0183] Among them, ω is the angular frequency of diurnal variation, φ is the phase difference, ∈1,∈2,∈3,∈4 are error terms. The lightweight neural network is used to optimize the parameters of the above equations. Figure 4 As shown in Figure 1, the response surface of the adaptive brightness adjustment model is shown, reflecting the influence of ambient light illumination and crowd density on the final adjusted brightness. Figure 5 As shown in the figure, the optimization results of the minimum partition of the lighting system are displayed, and the spatial distribution and partition of each lighting unit are shown through a scatter plot.

[0184] Taking lighting units 1 and 2 in zone 1 as an example, their real-time operating data is as follows:

[0185] Unit 1:

[0186] E=8000lux,D p =15 people / 100m 2 ,D v =4 vehicles / 100m, W = sunny;

[0187] L = 80%, P = 120W;

[0188] Unit 2:

[0189] E=7500lux,D p =12 people / 100m 2 ,D v =3 vehicles / 100m, W = sunny;

[0190] L = 85%, P = 130W;

[0191] According to the above data, the fuzzy controller first calculates the reference brightness value L b =90%.

[0192] Then, the brightness adaptation equations dynamically compensate and respond to the reference brightness:

[0193] Target brightness after illumination compensation:

[0194]

[0195] Target brightness after crowd response:

[0196]

[0197] Target brightness after traffic response:

[0198]

[0199] Final target brightness after energy consumption balance:

[0200]

[0201] Through the above calculations, the system finally determined that the target brightness values ​​of lighting unit 1 and unit 2 in partition 1 were 92% and 92% respectively, and sent brightness adjustment instructions to the two lighting units.

[0202] like Figure 6 As shown, the dynamic response curve of lighting control is displayed, including brightness changes in four stages: baseline brightness, lighting compensation, pedestrian flow response and vehicle flow response.

[0203] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for adaptive brightness adjustment of a smart city lighting system, characterized in that: The following steps are involved: S10, obtaining partition information of the smart city lighting system, wherein each partition includes a plurality of lighting units; S20. Establish a smart city lighting system matrix; S30, collecting real-time operating data of all lighting units in each of the partitions; S40, performing similarity analysis on each lighting unit in each of the partitions based on the real-time operation data, and establishing a first multi-objective optimization model including lighting demand satisfaction, energy consumption minimization, and lighting uniformity; S50, using the first multi-objective optimization model to perform optimization calculation on each of the partitions, calculate the control similarity matrix between the lighting units in the partition, construct lighting unit group division constraint conditions based on the control similarity matrix, and obtain an optimized partitioning scheme; S60. According to the optimized partitioning scheme, a second multi-objective optimization model including minimizing the energy consumption of the lighting unit, maximizing the lighting uniformity, and minimizing the number of partitions is established, and a plurality of extremely small partitions are obtained by solving the model, wherein the control similarity of the lighting units in the extremely small partitions is greater than a first preset threshold, and the control similarity with the adjacent lighting units is less than a second preset threshold; S70, establishing and fitting a brightness adaptive model including a fuzzy controller for each of the extremely small partitions, setting the input variables of the fuzzy controller to the ambient light illumination, crowd density, vehicle density, and weather conditions, and setting the output variable to the target brightness of the lighting unit; the brightness adaptive model includes a fuzzy controller, a brightness adaptive equation group, and a lightweight neural network; S80. Calculate the target brightness value of the lighting unit in each of the extremely small partitions based on the brightness adaptive model, and send a brightness adjustment instruction to all the lighting units in each of the extremely small partitions to adjust the actual brightness of the lighting units to the target brightness value.

2. The method for adaptive brightness adjustment of a smart city lighting system according to claim 1, characterized in that: The lighting unit refers to an intelligent street lamp monomer composed of a controller, an LED light source, and a sensor. The sensor includes a light sensor, a pedestrian flow detection sensor, and a vehicle flow detection sensor.

3. The method for adaptive brightness adjustment of a smart city lighting system according to claim 2, characterized in that: The elements of the smart city lighting system matrix include lighting unit identification, lighting unit geographic location information, lighting unit brightness parameters, lighting unit energy consumption parameters, and lighting unit operating status parameters.

4. The method for adaptive brightness adjustment of a smart city lighting system according to claim 3, characterized in that: The real-time operation data includes ambient light illumination data, pedestrian flow density data, vehicle flow density data, and weather condition data.

5. The method for adaptive brightness adjustment of a smart city lighting system according to claim 4, characterized in that: The fuzzy controller is used to realize the mapping relationship from environmental factors to lighting control strategies. The input is ambient light illumination data, pedestrian density data, vehicle density data, and weather condition data. The output is the reference brightness value of the lighting unit. The structure adopts a double-layer fuzzy reasoning system. The specific structure is that the first-layer fuzzy system processes the ambient light illumination and weather condition data to obtain the environmental evaluation value, and the second-layer fuzzy system processes the pedestrian density, vehicle density data and the environmental evaluation value to obtain the reference brightness value.

6. The method for adaptive brightness adjustment of a smart city lighting system according to claim 5, characterized in that: The brightness adaptive equation group includes an illumination compensation equation, a pedestrian flow response equation, a vehicle flow response equation and an energy consumption balance equation.

7. The method for adaptive brightness adjustment of a smart city lighting system according to claim 6, characterized in that: The illumination compensation equation is used to dynamically compensate the reference brightness value, the input is the ambient light illumination data and the reference brightness value, and the output is the compensated target brightness value; The crowd response equation is used to adjust the lighting brightness according to the crowd density, the input is the crowd density data and the target brightness value after compensation, and the output is the target brightness value after the crowd response; The traffic flow response equation is used to adjust the lighting brightness according to the traffic flow density, the input is the traffic flow density data and the target brightness value after the pedestrian flow response, and the output is the target brightness value after the traffic flow response; The energy consumption balance equation is used to balance the lighting effect and energy consumption. The input is the target brightness value after the traffic flow response and the energy consumption parameter of the lighting unit, and the output is the final target brightness value.

8. The method for adaptive brightness adjustment of a smart city lighting system according to claim 7, characterized in that: The lightweight neural network is used to optimize the parameters of the brightness adaptation model. The input is historical operation data and current operation data, and the output is the optimization parameters of the brightness adaptation equation group. The structure adopts a three-layer feedforward neural network. The specific structure is that the number of input layer nodes is 8, the number of hidden layer nodes is 16, the number of output layer nodes is 4, and the ReLU activation function is adopted.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, and when the program instructions are executed in a computer, they are used to execute the method for adaptive brightness adjustment of a smart city lighting system according to any one of claims 1 to 8.

10. A smart city lighting system adaptive brightness adjustment system, characterized in that: A computer-readable storage medium comprising the computer-readable storage medium of claim 9.

Images