A city lighting control fusion gateway system

Abstract

The present invention relates to the technical field of urban lighting control, and provides an urban lighting control fusion gateway system, including a multimodal perception network module, a causal reasoning diagnosis module, a destruction-resistant topology reconstruction module, a physiological adaptation dimming module, and a disaster chain blocking module. The multimodal perception network module collects structural displacements, environmental data, and weather forecasts of lighting facilities through drones, ground nodes, and meteorological radars. The causal reasoning diagnosis module analyzes the causes of facility aging through causal graph modeling. The destruction-resistant topology reconstruction module optimizes the power supply network topology. The physiological adaptation dimming module generates color temperature-brightness parameters according to the flow of people and the optical characteristics of the atmosphere. The disaster chain blocking module identifies disaster propagation nodes and triggers reinforcement instructions. The present invention can improve the reliability, energy saving, and safety of the lighting system, and optimize the comfort level and emergency response capabilities of the urban environment.

Description

Technical Field

[0001] The present invention relates to the technical field of urban lighting control. More specifically, the present invention relates to an urban lighting control fusion gateway system. Background Art

[0002] As an important part of urban infrastructure, the intelligent management and control of urban lighting systems have always been a research hotspot in the field of urban management. Traditional urban lighting control mainly relies on timed switches and simple light sensing control, making it difficult to cope with complex urban environments and diverse lighting needs. With the development of Internet of Things, big data, and artificial intelligence technologies, existing urban lighting control systems have gradually introduced more sensors and data analysis methods, but there are still many limitations. For example, existing systems often lack deep integration of multi-source data and cannot accurately identify the root causes of lighting facility aging; in terms of power supply network topology optimization, existing technologies mostly use static planning and are difficult to dynamically respond to sudden failures; in addition, traditional lighting control fails to fully consider the human physiological rhythm and real-time changes in the urban environment, resulting in a mismatch between lighting effects and human comfort. In terms of disaster prevention, existing systems usually rely on the simulation of a single physical field and are difficult to accurately predict and block the propagation of disaster chains.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: Existing urban lighting control systems have obvious deficiencies in multi-source data fusion, dynamic topology optimization, physiological adaptation dimming, and disaster chain blocking, and it is difficult to achieve precise, efficient, and safe urban lighting management. Summary of the Invention

[0004] The present invention provides an urban lighting control fusion gateway system, including:

[0005] A multi-modal perception network module, configured to synchronously collect millimeter-level structural displacement data, ambient temperature and humidity gradient distribution data, and short-term extreme weather prediction data of lighting facilities through high-precision deformation sensors carried by low-altitude drone swarms and micro-environment monitoring units of distributed ground nodes, and fuse the underground pipe network topology data in the geographic information system;

[0006] A causal reasoning and diagnosis module, configured to perform spatio-temporal causal graph modeling on the multi-dimensional data output by the multi-modal perception network module, separate the root causes of facility aging from apparent associated factors through intervention-based causal analysis, and generate a causal feature map;

[0007] A damage-resistant topology reconstruction module, configured to generate a dynamic topology adjustment strategy based on the causal feature map through an adaptive resilience optimization algorithm for the power supply network, and the algorithm includes a quantitative evaluation of the failure probability of key nodes and an intelligent planning of multi-path redundant power supply links;

[0008] The physiological adaptation dimming module is used to generate color temperature-brightness combination parameters that conform to the human body clock through the visual-rhythm coupling model according to the real-time pedestrian and vehicle flow and atmospheric optical characteristics in the urban area, and send them to the intelligent dimming terminal;

[0009] The disaster chain blocking module is used to identify key hub nodes for disaster propagation based on a multi-physics field coupling simulation engine and a historical disaster evolution database, and trigger targeted infrastructure reinforcement instructions.

[0010] Furthermore, the causal reasoning diagnosis module includes:

[0011] The causal graph construction unit is used to establish a conditional independence relationship network between facility state variables and multi-source environmental variables. The specific execution steps include:

[0012] An initial causal graph is constructed based on the counterfactual reasoning framework, with nodes including metal fatigue, salt spray corrosion rate, and instantaneous wind load.

[0013] The marginal causal contribution of each variable to the failure rate is calculated by the gradient intervention method. If the contribution exceeds the threshold, a directional causal edge is established.

[0014] The root cause mining unit is used to extract the key cause chain from the cause-effect diagram, including:

[0015] Identify the shortest causal path leading to the failure event and quantify the time-varying impact weight of each node in the path;

[0016] A causal strength attenuation model is constructed to calculate the residual effects of historical intervention measures on the current failure rate, and the residual effects are input into the invulnerability topology reconstruction module to optimize the redundant design of the power supply link.

[0017] Furthermore, the execution steps of the adaptive toughness optimization algorithm include:

[0018] Construct a dynamic vulnerability heat map of the power supply network, where the node vulnerability is calculated jointly by service life, leakage history, and the intensity of surrounding geological activities;

[0019] When it is detected that the vulnerability of key hub nodes exceeds the critical value, the three-level response mechanism is activated:

[0020] The first-level response generates a backup power supply plan based on the minimum cut set, and preferentially calls the low residual effect nodes identified in the causal reasoning diagnosis module as the backup power supply access points;

[0021] The secondary response optimizes the balanced line load distribution through impedance matching. Specifically, an asymmetric power flow constraint model is used to ensure that the load fluctuation rate of each branch is less than 5%, and the optimized load data is synchronized to the disaster chain blocking module to update the disaster propagation model.

[0022] The third-level response triggers the coordinated discharge of energy storage devices in the adjacent areas, and the discharge strategy is dynamically adjusted according to the energy consumption demand for the next hour predicted by the physiological adaptation dimming module.

[0023] Furthermore, the construction of the visual-rhythm coupling model includes:

[0024] Establish the response curve of human retinal ganglion cells and quantify the visual comfort index under different color temperatures;

[0025] Integrate wireless positioning data to build a heat map of human flow activities and dynamically divide high-density activity areas and silent areas;

[0026] The energy-saving efficiency and physiological adaptation requirements are balanced through a dual-objective optimization algorithm, where the objective function is: in, It is a dynamic weight factor, which is automatically adjusted according to the real-time electricity price and crowd density. For energy efficiency, For biorhythm adaptation, The optimized color temperature-brightness parameters are sent to the intelligent dimming terminal and synchronized to the disaster chain blocking module as the reference parameters of the emergency lighting plan. Furthermore, the multi-physics field coupling simulation engine includes:

[0027] Map meteorological data, geological deformation data and power supply network status to a unified spatiotemporal grid;

[0028] The interaction process between wind force, structure and electromagnetic field is simulated by the bidirectional coupling solver to predict the stress concentration area of ​​key facilities. When it exceeds 1.2, the prestress compensation instruction is triggered. is the real-time stress, is the yield strength of the material; the compensation instruction synchronously drives the anti-destruction topology reconstruction module to start the secondary response, and adjusts the load distribution of the power supply network to reduce the pressure on the damaged line.

[0029] Furthermore, the multimodal perception network module also includes:

[0030] The drone swarm coordination unit is used to maintain the integrity of the sensing network in severe weather. The specific steps include:

[0031] Build a dynamic inspection priority map based on risk entropy value, risk entropy in is the sub-area failure probability; a distributed task allocation mechanism driven by game theory is adopted to input the priority map into the decision-making system of the drone, so that the drone cluster can autonomously optimize the coverage path according to the map data when the communication is interrupted;

[0032] Implement emergency obstacle avoidance and formation reorganization of the cluster through a bionic pulse coupling algorithm. The obstacle avoidance strategy calls the active heat map data in the physiological adaptation dimming module in real time to avoid crowded areas.

[0033] Furthermore, the implementation of the game theory-driven mechanism includes:

[0034] Define the multi-dimensional utility function of the UAV, including coverage benefit, energy consumption cost, and risk aversion factor. The specific formula is:

[0035] Among them, is the dynamic weight, which is dynamically adjusted according to the risk level predicted by the multi-physical field coupling simulation engine;

[0036] Construct a Bayesian Nash equilibrium model under incomplete information, optimize individual decisions through online policy iteration, and the decision results are fed back to the anti-destruction topology reconstruction module in real time to adjust the inspection priority of the power supply network;

[0037] Introduce a reputation mechanism to punish individuals with frequent conflicts, improve the cluster cooperation efficiency, and synchronize the reputation data to the disaster chain blocking module for evaluating the regional safety factor.

[0038] Furthermore, the disaster chain blocking module also includes:

[0039] The key hub identification unit is used to locate the super nodes in the disaster propagation network. The specific steps include:

[0040] Calculate the weighted comprehensive index of node betweenness centrality and eigenvector centrality, and the weights are dynamically allocated according to the proportion of the chain effect of node failures in historical disaster data;

[0041] Identify the key transmission paths that are frequently accessed, and input the path data into the multi-modal perception network module to guide the UAV cluster to focus on monitoring high-risk nodes;

[0042] When it is detected that the comprehensive influence index of a certain node exceeds 3 times the standard deviation of the global mean, it is marked as a super node that needs to be preferentially strengthened, and the three-level response mechanism of the anti-destruction topology reconstruction module is triggered to call the energy storage device to perform redundant protection on the power supply link of this node.

[0043] Furthermore, the physiological adaptation dimming module also includes:

[0044] The rhythm memory unit is used to learn the lighting preference patterns of specific areas. The specific steps include:

[0045] Construct a time series model of regional lighting habits through deep reinforcement learning. The model training data comes from the historical environment monitoring records of the multi-modal perception network module;

[0046] During major events, the mode enhancement mechanism is activated to integrate historical optimal lighting parameters with real-time environmental data to generate customized solutions. The solution data is synchronized to the disaster chain blocking module to optimize the lighting parameters of the emergency plan;

[0047] When a new crowd gathering pattern is detected, the incremental learning algorithm is activated to update the dimming strategy library. The updated strategy is sent to the smart dimming terminal in real time, and the drone cluster collaborative unit is triggered to conduct high-frequency status inspections of the changed area.

[0048] The above-mentioned embodiments of the present invention have at least the following beneficial effects: the present invention can improve the intelligent management level of urban lighting systems through the collaborative work of a multimodal perception network module, a causal reasoning diagnosis module, a resistant topology reconstruction module, a physiological adaptation dimming module, and a disaster chain blocking module. The multimodal perception network module can collect millimeter-level structural displacement data, environmental temperature and humidity gradient distribution data, and short-term extreme weather forecast data of lighting facilities in real time, and combine the underground pipe network topology data in the geographic information system to provide the system with comprehensive environmental perception capabilities. The causal reasoning diagnosis module can accurately identify the root causes of facility aging through spatiotemporal causal graph modeling and interventional causal analysis, generate causal feature maps, and provide a scientific basis for subsequent topological reconstruction and disaster prevention.

[0049] The anti-destruction topology reconstruction module can generate a dynamic topology adjustment strategy based on the causal feature map through an adaptive resilience optimization algorithm to ensure that the power supply network can maintain stable operation when key nodes fail. The physiological adaptation dimming module can generate color temperature-brightness combination parameters that conform to the human body's biological clock through a visual-rhythm coupling model based on the real-time pedestrian activity patterns and atmospheric optical properties in urban areas, thereby improving lighting comfort. The disaster chain blocking module can identify key hub nodes for disaster propagation based on a multi-physics field coupling simulation engine and a historical disaster evolution database, and trigger targeted infrastructure reinforcement instructions to effectively reduce disaster risks. On the whole, the present invention can improve the reliability, energy efficiency and safety of urban lighting systems, while optimizing the comfort and emergency response capabilities of the urban environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0051] Figure 1 A schematic diagram of the structure of an urban lighting control fusion gateway system provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0052] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0053] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0054] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0055] Reference below Figure 1 , Figure 1 This is a schematic diagram of the structure of the urban lighting control fusion gateway system provided by an embodiment of the present invention. Figure 1 As shown, a city lighting control fusion gateway system includes:

[0056] The multimodal perception network module 101 is used to synchronously collect millimeter-level structural displacement data of lighting facilities, environmental temperature and humidity gradient distribution data, and short-term extreme weather forecast data through high-precision deformation sensors carried by low-altitude drone clusters, microenvironment monitoring units of distributed ground nodes, and real-time storm tracking systems of meteorological radars, and integrate underground pipe network topology data in geographic information systems;

[0057] A causal reasoning diagnosis module 102 is used to perform spatiotemporal causal graph modeling on the multi-dimensional data output by the multimodal perception network module, separate the fundamental causes of facility aging from the apparent correlation factors through interventional causal analysis, and generate a causal feature map;

[0058] A survivability topology reconstruction module 103, configured to generate a dynamic topology adjustment strategy based on the causal feature graph through an adaptive resilience optimization algorithm of the power supply network, wherein the algorithm includes a quantitative evaluation of the failure probability of key nodes and an intelligent planning of multi-path redundant power supply links;

[0059] The physiological adaptation dimming module 104 is used to generate color temperature-brightness combination parameters that conform to the human body clock through a visual-rhythm coupling model according to the real-time flow of people and vehicles in the urban area and the optical characteristics of the atmosphere, and send them to the intelligent dimming terminal;

[0060] The disaster chain blocking module 105 is used to identify key hub nodes of disaster propagation based on a multi-physics field coupling simulation engine and a historical disaster evolution database, and trigger targeted infrastructure reinforcement instructions.

[0061] It should be noted that the present invention relates to an urban lighting control fusion gateway system, which includes multiple modules for realizing intelligent management of urban lighting facilities. The multimodal perception network module is the core part of the system. Through a variety of equipment such as low-altitude drone clusters, distributed ground nodes and meteorological radars, the millimeter-level structural displacement data, environmental temperature and humidity gradient distribution data and short-term extreme weather forecast data of lighting facilities are synchronously collected. These data are integrated with the underground pipe network topology data in the geographic information system to provide comprehensive environmental perception capabilities for subsequent decision-making. The "low-altitude drone cluster" in the multimodal perception network module refers to a collaborative working system composed of multiple drones flying at low altitudes, which can be equipped with high-precision deformation sensors for data collection.

[0062] Specifically, the high-precision deformation sensors in the multimodal perception network module can detect tiny displacement changes in lighting facilities with millimeter-level accuracy, ensuring real-time monitoring of the health of the facility structure. The microenvironment monitoring unit of the distributed ground node can collect environmental temperature and humidity gradient distribution data to help the system understand the microclimate conditions in different regions. The real-time storm tracking system of the meteorological radar can predict short-term extreme weather, such as heavy rain and strong winds, and provide early warning of possible disaster risks. The underground pipe network topology data in the geographic information system provides layout information of the city's underground pipe network, helping the system to better understand the relationship between lighting facilities and underground infrastructure. The fusion of these data can provide comprehensive input for the subsequent causal reasoning diagnosis module.

[0063] Preferably, low-altitude drone swarms can still work in bad weather. Through dynamic inspection priority maps and game theory-driven task allocation mechanisms, drones can autonomously optimize coverage paths when communications are interrupted. The bionic pulse coupling algorithm can ensure that drone swarms can achieve obstacle avoidance and formation reorganization in emergency situations to avoid conflicts with crowded areas. In addition, the inspection data of the drone swarm can be fed back to the multimodal perception network module in real time to ensure the integrity of the perception network and the real-time nature of the data. This design not only improves the reliability of the system, but also enhances its adaptability in complex environments.

[0064] In some embodiments, the causal reasoning diagnosis module includes:

[0065] The causal graph construction unit is used to establish a conditional independence relationship network between facility state variables and multi-source environmental variables. The specific execution steps include:

[0066] An initial causal graph is constructed based on the counterfactual reasoning framework, with nodes including metal fatigue, salt spray corrosion rate, and instantaneous wind load.

[0067] The marginal causal contribution of each variable to the failure rate is calculated by the gradient intervention method. If the contribution exceeds the threshold, a directional causal edge is established.

[0068] The root cause mining unit is used to extract the key cause chain from the cause-effect diagram, including:

[0069] Identify the shortest causal path leading to the failure event and quantify the time-varying impact weight of each node in the path;

[0070] A causal strength attenuation model is constructed to calculate the residual effects of historical intervention measures on the current failure rate, and the residual effects are input into the invulnerability topology reconstruction module to optimize the redundant design of the power supply link.

[0071] It should be noted that the causal reasoning diagnosis module is a key part of the present invention, which is used to model the spatiotemporal causal graph of the multi-dimensional data output by the multimodal perception network module, and to separate the fundamental causes and apparent correlation factors of facility aging through interventional causal analysis to generate a causal feature map. The causal graph construction unit is the core component of this module, which is used to establish a conditional independent relationship network between facility state variables and multi-source environmental variables. Specifically, the unit constructs an initial causal graph based on the counterfactual reasoning framework, and the nodes include key variables such as metal fatigue, salt spray corrosion rate, and instantaneous wind load. The counterfactual reasoning framework is a model that analyzes causal relationships through hypothetical scenarios, which can help identify causal relationships between variables.

[0072] Specifically, the causal graph construction unit calculates the marginal causal contribution of each variable to the failure rate through the gradient intervention method. If the contribution exceeds the threshold, a directed causal edge is established. The gradient intervention method is a mathematical method that evaluates the impact of variables on results by gradually adjusting their values. It can quantify the contribution of different variables to the failure rate. The root cause mining unit extracts the key cause chain from the causal graph, identifies the shortest causal path leading to the failure event, and quantifies the time-varying impact weight of each node in the path. In addition, the unit also constructs a causal strength attenuation model, calculates the residual effects of historical intervention measures on the current failure rate, and inputs these effects into the anti-destruction topology reconstruction module to optimize the redundant design of the power supply link.

[0073] Preferably, the causal graph construction unit can combine historical fault data and real-time monitoring data when constructing the initial causal graph to ensure the accuracy of the causal relationship. When identifying the shortest causal path, the root cause mining unit can adopt a dynamic weight adjustment strategy to update the time-varying influence weight of each node according to real-time data. The calculation of the causal strength attenuation model can be based on time series analysis to ensure that the residual effects of historical intervention measures can be accurately quantified. These optimization measures can improve the accuracy and reliability of the causal reasoning diagnosis module and provide a more scientific basis for subsequent topology reconstruction and disaster prevention.

[0074] In some embodiments, the steps of executing the adaptive toughness optimization algorithm include:

[0075] Construct a dynamic vulnerability heat map of the power supply network, where the node vulnerability is calculated jointly by service life, leakage history, and the intensity of surrounding geological activities;

[0076] When it is detected that the vulnerability of key hub nodes exceeds the critical value, the three-level response mechanism is activated:

[0077] The first-level response generates a backup power supply plan based on the minimum cut set, and preferentially calls the low residual effect nodes identified in the causal reasoning diagnosis module as the backup power supply access points;

[0078] The secondary response optimizes the balanced line load distribution through impedance matching. Specifically, an asymmetric power flow constraint model is used to ensure that the load fluctuation rate of each branch is less than 5%, and the optimized load data is synchronized to the disaster chain blocking module to update the disaster propagation model.

[0079] The third-level response triggers the coordinated discharge of energy storage devices in the adjacent areas, and the discharge strategy is dynamically adjusted according to the energy consumption demand for the next hour predicted by the physiological adaptation dimming module.

[0080] It should be noted that the anti-destruction topology reconstruction module is the core component of the present invention, which is used to generate dynamic topology adjustment strategies based on the causal feature map through the adaptive resilience optimization algorithm of the power supply network. The algorithm includes a quantitative evaluation of the failure probability of key nodes and intelligent planning of multi-path redundant power supply links to ensure that the power supply network can maintain stable operation in the face of sudden failures. The adaptive resilience optimization algorithm is an intelligent algorithm that dynamically adjusts the topology of the power supply network. It can evaluate the vulnerability of key nodes based on real-time data and generate corresponding adjustment strategies.

[0081] Specifically, this module first constructs a dynamic vulnerability heat map of the power supply network, and the node vulnerability is jointly calculated by the service life, leakage history and the intensity of surrounding geological activities. Service life refers to the usage time of the power supply equipment, the leakage history records the past leakage of the equipment, and the intensity of surrounding geological activities reflects the geological stability of the area where the equipment is located. When it is detected that the vulnerability of the key hub node exceeds the critical value, the system initiates the three-level response mechanism. The first-level response generates a backup power supply plan based on the minimum cut set, and gives priority to calling the low residual effect nodes identified in the causal reasoning diagnosis module as the backup power supply access point. The minimum cut set refers to the set of nodes that disconnect the least connection in the network to split the network, ensuring the efficiency of the backup power supply plan.

[0082] Preferably, the secondary response optimizes the balanced line load distribution through impedance matching, adopts an asymmetric power flow constraint model to ensure that the load fluctuation rate of each branch is lower than a certain threshold, and synchronizes the optimized load data to the disaster chain blocking module to update the disaster propagation model. The tertiary response triggers the coordinated discharge of energy storage devices in adjacent areas, and the discharge strategy is dynamically adjusted according to the future energy consumption demand predicted by the physiological adaptation dimming module. The coordinated discharge of energy storage devices can adjust the discharge rate according to real-time needs to ensure the stability of the power supply network. These optimization measures can improve the anti-destruction and reliability of the power supply network and ensure the stable operation of the urban lighting system in complex environments.

[0083] In some embodiments, the construction of the visual-rhythm coupling model includes:

[0084] Establish the response curve of human retinal ganglion cells and quantify the visual comfort index under different color temperatures;

[0085] Integrate wireless positioning data to build a heat map of human flow activities and dynamically divide high-density activity areas and silent areas;

[0086] The energy-saving efficiency and physiological adaptation requirements are balanced through a dual-objective optimization algorithm, where the objective function is: in, It is a dynamic weight factor, which is automatically adjusted according to the real-time electricity price and crowd density. For energy efficiency, For biorhythm adaptation, As the optimization target; the optimized color temperature-brightness parameters are sent to the intelligent dimming terminal, and synchronized to the disaster chain blocking module as the benchmark parameters of the emergency lighting plan. It should be noted that the physiological adaptation dimming module is a key part of the present invention, which is used to generate color temperature-brightness combination parameters that conform to the human body's biological clock through the visual-rhythm coupling model according to the real-time pedestrian activity pattern and atmospheric optical characteristics in the urban area, and send them to the intelligent dimming terminal. The visual-rhythm coupling model is a mathematical model that combines the visual characteristics of the human eye and the human biological rhythm, and can dynamically adjust the lighting parameters according to environmental changes and pedestrian activities. This module first establishes the response curve of the retinal ganglion cells of the human eye, quantifies the visual comfort index under different color temperatures, and ensures that the lighting effect meets the visual needs of the human body.

[0087] Specifically, the physiological adaptation dimming module constructs a heat map of human flow activities by integrating wireless positioning data, and dynamically divides high-density activity areas and silent areas. Wireless positioning data can come from smartphones, wearable devices, etc., to help the system grasp the distribution of human flow in urban areas in real time. By balancing energy-saving efficiency and physiological adaptation needs through a dual-objective optimization algorithm, the system can automatically adjust the dynamic weight factor according to the real-time electricity price and human flow density to ensure that the lighting system meets human comfort needs while saving energy. The optimized color temperature-brightness parameters will be sent to the intelligent dimming terminal and synchronized to the disaster chain blocking module as the benchmark parameters of the emergency lighting plan.

[0088] Preferably, the visual-rhythm coupling model can be constructed using deep learning technology, and the model can be trained through historical data to improve its prediction accuracy. The heat map of human flow activities can be dynamically adjusted according to different time periods and weather conditions to ensure that the system can adapt to complex environmental changes. The dual-objective optimization algorithm can use a multi-objective genetic algorithm to ensure that the optimal balance is found between energy-saving efficiency and physiological adaptation needs. These optimization measures can improve the intelligence level of the lighting system and ensure that urban lighting provides a comfortable visual experience while meeting energy-saving requirements.

[0089] In some embodiments, the calculation of the biorhythm fitness includes:

[0090] Through the heart rate variability data of the crowd collected by wearable devices, the collective stress level can be inferred;

[0091] Establish a nonlinear mapping relationship between blue light spectrum intensity and melatonin suppression rate;

[0092] When it is detected that the regional pressure level exceeds the threshold, the spectrum proportion of the 450-480nm band is automatically reduced, while the proportion of warm tones in the long-wave band is increased, and the adjusted spectral data is fed back to the multimodal perception network module to update the environmental monitoring database.

[0093] It should be noted that the calculation of biorhythm adaptation in the physiological adaptation dimming module is a key link of the present invention, which is used to ensure that the lighting system can dynamically adjust the lighting parameters according to the physiological rhythm of the human body. The calculation uses the heart rate variability data of the crowd collected by wearable devices to reverse the collective stress level, and combines the nonlinear mapping relationship between the blue light spectrum intensity and the melatonin inhibition rate to dynamically adjust the lighting spectrum. Heart rate variability data refers to the changes in the heartbeat interval, which can reflect the stress state of the human body; melatonin is a hormone related to sleep regulation, and its secretion is affected by the blue light spectrum.

[0094] Specifically, the calculation of biorhythm fitness first collects the heart rate variability data of the crowd through wearable devices to analyze the collective stress level. When the regional stress level is detected to exceed the preset threshold, the system will automatically reduce the proportion of blue light spectrum in the 450-480nm band, and increase the proportion of warm tones in the long-wave band to reduce the inhibition of human melatonin secretion. The adjusted spectral data will be fed back to the multimodal perception network module to update the environmental monitoring database to ensure that the system can respond to environmental changes in real time. The adjustment range of the blue light spectrum intensity can be dynamically set according to different time periods and crowd activity patterns to ensure that the lighting effect meets physiological needs and does not have a negative impact on visual comfort. Preferably, the system can use a machine learning algorithm to analyze the heart rate variability data in real time to improve the accuracy of stress level detection. The nonlinear mapping relationship between the blue light spectrum intensity and the melatonin suppression rate can be calibrated through experimental data to ensure that the adjusted spectral parameters can effectively relieve human stress. In addition, the system can dynamically adjust the adjustment strategy of the lighting spectrum according to the collective stress level of different regions to ensure a softer lighting environment in high-pressure areas. These optimization measures can improve the physiological adaptability of the lighting system and ensure that urban lighting provides a healthier and more comfortable visual experience while meeting energy-saving requirements.

[0095] In some embodiments, the multi-physics coupling simulation engine includes:

[0096] Map meteorological data, geological deformation data and power supply network status to a unified spatiotemporal grid;

[0097] The interaction process of wind force, structure and electromagnetic field is simulated by bidirectional coupling solver to predict the stress concentration areas of key facilities;

[0098] When the cumulative damage value of a grid is detected When it exceeds 1.2, the prestress compensation instruction is triggered, is the real-time stress, is the yield strength of the material; the compensation instruction synchronously drives the anti-destruction topology reconstruction module to start the secondary response, and adjusts the load distribution of the power supply network to reduce the pressure on the damaged line.

[0099] It should be noted that the multi-physics coupling simulation engine is one of the core components of the present invention, which is used to map meteorological data, geological deformation data and power supply network status to a unified space-time grid, simulate the interaction process of wind-structure-electromagnetic field through a bidirectional coupling solver, and predict the stress concentration area of ​​key facilities. The multi-physics coupling simulation engine is a computational model that can handle the interaction of multiple physical fields at the same time, and can simulate the stress conditions of facilities in complex environments. When it is detected that the cumulative damage value of a grid exceeds the preset threshold, the system will trigger the prestress compensation instruction to ensure the safety of the facility under extreme conditions.

[0100] Specifically, the multi-physics coupling simulation engine first maps meteorological data, geological deformation data and power supply network status to a unified space-time grid to ensure that data from different physical fields can be analyzed in the same space-time framework. The bidirectional coupling solver is used to simulate the interaction between wind, structure and electromagnetic fields, and predict stress concentration areas of key facilities. When it is detected that the cumulative damage value of a grid exceeds the preset threshold, the system triggers the prestress compensation instruction and adjusts the load distribution of the power supply network to reduce the pressure on the damaged line. The cumulative damage value is calculated by the ratio of real-time stress to material yield strength, which can reflect the fatigue degree of the facility.

[0101] Preferably, the multi-physics coupling simulation engine can adopt high-performance computing technology to improve the accuracy and efficiency of the simulation. The bidirectional coupling solver can dynamically adjust the simulation parameters according to real-time data to ensure the accuracy of the prediction results. The trigger threshold of the prestress compensation instruction can be dynamically set according to the material properties and use environment of different facilities to ensure that the system can respond to potential risks in a timely manner. In addition, the system can optimize the parameter settings of the simulation engine based on historical data and real-time monitoring data to improve the reliability of the prediction. These optimization measures can improve the disaster prevention capabilities of the system and ensure the safe operation of urban lighting facilities in complex environments.

[0102] In some embodiments, the multimodal perception network module further includes:

[0103] The drone swarm coordination unit is used to maintain the integrity of the perception network in bad weather. The specific steps include: building a dynamic inspection priority map based on risk entropy value, risk entropy value ,in is the sub-area failure probability; a distributed task allocation mechanism driven by game theory is adopted to input the priority map into the decision-making system of the drone, so that the drone cluster can autonomously optimize the coverage path according to the map data when the communication is interrupted;

[0104] The bionic pulse coupling algorithm is used to realize emergency obstacle avoidance and formation reorganization of the cluster. The obstacle avoidance strategy calls the activity heat map data in the physiological adaptation dimming module in real time to avoid areas with dense crowds.

[0105] It should be noted that the drone cluster collaboration unit in the multimodal perception network module is a key part of the present invention, which is used to maintain the integrity of the perception network in bad weather. This unit constructs a dynamic inspection priority map based on risk entropy value and adopts a distributed task allocation mechanism driven by game theory to ensure that the drone cluster can autonomously optimize the coverage path when communication is interrupted. The risk entropy value is an indicator used to quantify the risk of regional failure, which can help the system identify high-priority inspection areas. The distributed task allocation mechanism driven by game theory is a task allocation method based on multi-agent collaboration, which can ensure that drone clusters can work efficiently and collaboratively in complex environments.

[0106] Specifically, the drone cluster coordination unit first constructs a dynamic inspection priority map based on the risk entropy value. The risk entropy value is calculated by the sub-area failure probability to ensure that high-failure risk areas can be inspected first. A distributed task allocation mechanism driven by game theory is used to input the priority map into the drone's decision-making system, so that the drone cluster can autonomously optimize the coverage path according to the map data when the communication is interrupted. The bionic pulse coupling algorithm is used to achieve emergency obstacle avoidance and formation reorganization of the cluster. The obstacle avoidance strategy calls the activity heat map data in the physiological adaptation dimming module in real time to ensure that the drone avoids crowded areas. The bionic pulse coupling algorithm is an algorithm that simulates the behavior of biological groups and can achieve autonomous obstacle avoidance and formation adjustment of drone clusters.

[0107] Preferably, the dynamic inspection priority map can be dynamically updated according to the real-time monitoring data to ensure that the drone cluster can respond to environmental changes in a timely manner. The game theory-driven distributed task allocation mechanism can adopt a multi-agent reinforcement learning algorithm to improve the efficiency and accuracy of task allocation. The bionic pulse coupling algorithm can dynamically adjust the obstacle avoidance strategy according to the real-time position and speed of the drone to ensure the safe flight of the cluster in complex environments. In addition, the system can optimize the inspection path and task allocation strategy of the drone cluster based on historical inspection data and real-time monitoring data to improve the integrity and reliability of the perception network. These optimization measures can improve the working efficiency of the drone cluster in bad weather and ensure the continuous operation of the perception network.

[0108] In some embodiments, the implementation of the game theory driven mechanism includes:

[0109] Define the multi-dimensional utility function of drones, including coverage benefits, energy consumption costs and risk aversion factors. The specific formula is:

[0110] in, It is a dynamic weight, which is adjusted dynamically according to the risk level predicted by the multi-physics coupling simulation engine;

[0111] A Bayesian Nash equilibrium model under incomplete information is constructed, and individual decisions are optimized through online policy iteration. The decision results are fed back to the invulnerability topology reconstruction module in real time to adjust the inspection priority of the power supply network.

[0112] A reputation mechanism is introduced to punish individuals with frequent conflicts and improve cluster collaboration efficiency. The reputation data is synchronized to the disaster chain blocking module to evaluate the regional safety factor.

[0113] It should be noted that the game theory driven mechanism in the drone cluster collaborative unit is the core part of the present invention, which is used to ensure the efficient collaboration of drones in complex environments. This mechanism defines a multi-dimensional utility function of drones, including coverage benefits, energy consumption costs and risk aversion factors, to ensure that drones can comprehensively consider multiple factors when assigning tasks. Coverage benefits refer to the effectiveness of drones in completing inspection tasks, energy costs refer to the energy consumption of drones when performing tasks, and risk aversion factors are used to assess the collision risks faced by drones when performing tasks. By constructing a Bayesian Nash equilibrium model under incomplete information, the system can optimize individual decisions online to ensure that the overall efficiency of the drone cluster is maximized.

[0114] Specifically, the game theory-driven mechanism first defines the multi-dimensional utility function of the drone, and the dynamic weight is adjusted according to the risk level predicted by the multi-physics field coupling simulation engine to ensure that the drone can give priority to high-risk areas when assigning tasks. The Bayesian Nash equilibrium model under incomplete information is used to optimize individual decisions. Through online strategy iteration, the drone can adjust the task allocation strategy according to real-time data. The decision results are fed back to the anti-destruction topology reconstruction module in real time to adjust the inspection priority of the power supply network. In addition, the system introduces a reputation mechanism to punish individuals with frequent conflicts and improve the efficiency of cluster collaboration. The reputation data is synchronized to the disaster chain blocking module to evaluate the regional safety factor.

[0115] Preferably, the dynamic weights of the multi-dimensional utility function can be adjusted according to the real-time status and environmental data of the drone to ensure the flexibility and adaptability of task allocation. The Bayesian Nash equilibrium model uses distributed computing technology to improve the efficiency and real-time performance of decision optimization. The reputation mechanism can be dynamically adjusted according to the historical performance of the drone to ensure the efficiency of cluster collaboration. In addition, the system can optimize the parameter settings of the utility function based on historical mission data and real-time monitoring data to improve the task execution efficiency of the drone cluster. These optimization measures can improve the collaborative effectiveness of drone clusters in complex environments and ensure the continuous operation and efficient inspection of the perception network.

[0116] In some embodiments, the disaster chain blocking module further includes:

[0117] The key hub identification unit is used to locate super nodes in the disaster propagation network. The specific steps include:

[0118] Calculate the weighted comprehensive index of betweenness centrality and eigenvector centrality of computing nodes, and the weights are dynamically allocated according to the proportion of the cascading effect of node failures in historical disaster data;

[0119] Identify the key transmission paths that are frequently accessed, and input the path data into the multi-modal perception network module to guide the key monitoring of high-risk nodes by the UAV cluster;

[0120] When it is detected that the comprehensive influence index of a certain node exceeds 3 times the standard deviation of the global mean, it is marked as a super node that needs to be preferentially strengthened, and a three-level response mechanism of the anti-destruction topology reconstruction module is triggered, and energy storage devices are called to perform redundant protection on the power supply link of this node.

[0121] It should be noted that the key hub identification unit in the disaster chain blocking module is the core part of the present invention, which is used to locate the super nodes in the disaster propagation network to ensure that the system can timely block the propagation of the disaster chain. This unit calculates the weighted comprehensive index of betweenness centrality and eigenvector centrality of nodes to identify the key nodes in the disaster propagation network. Betweenness centrality refers to the frequency of a node acting as a bridge in the network, and eigenvector centrality reflects the degree of connection between a node and other important nodes. By constructing a random walk model of disaster propagation, the system can identify the key transmission paths that are frequently accessed to ensure the key monitoring of high-risk nodes.

[0122] Specifically, the key hub identification unit first calculates the weighted comprehensive index of betweenness centrality and eigenvector centrality of nodes, and the weights are dynamically allocated according to the proportion of the cascading effect of node failures in historical disaster data to ensure the accuracy of key node identification. The random walk model is used to simulate the propagation path of disasters in the network to identify the key transmission paths that are frequently accessed. The path data is input into the multi-modal perception network module to guide the key monitoring of high-risk nodes by the UAV cluster. When it is detected that the comprehensive influence index of a certain node exceeds three times the standard deviation of the global mean, the system will mark this node as a super node that needs to be preferentially strengthened, and trigger a three-level response mechanism of the anti-destruction topology reconstruction module, and call energy storage devices to perform redundant protection on the power supply link of this node.

[0123] Preferably, the weighted comprehensive index of betweenness centrality and eigenvector centrality can be dynamically adjusted according to real-time monitoring data to ensure the accuracy of key node identification. The random walk model can use the Monte Carlo simulation method to improve the accuracy and efficiency of path identification. The threshold of the comprehensive influence index can be dynamically set according to the historical disaster data of different regions to ensure that the system can respond to potential risks in a timely manner. In addition, the system can optimize the parameter settings of the key hub identification unit based on historical disaster data and real-time monitoring data to improve the accuracy and timeliness of disaster chain blocking. These optimization measures can improve the disaster prevention capabilities of the system and ensure the safe operation of urban lighting facilities in complex environments.

[0124] In some embodiments, the physiological adaptation dimming module further includes:

[0125] Rhythm memory unit, used to learn the lighting preference pattern of a specific area, the specific steps include:

[0126] A time series model of regional lighting habits is constructed through deep reinforcement learning, and the model training data comes from the historical environmental monitoring records of the multimodal perception network module;

[0127] During major events, the mode enhancement mechanism is activated to integrate historical optimal lighting parameters with real-time environmental data to generate customized solutions. The solution data is synchronized to the disaster chain blocking module to optimize the lighting parameters of the emergency plan;

[0128] When a new crowd gathering pattern is detected, the incremental learning algorithm is activated to update the dimming strategy library. The updated strategy is sent to the smart dimming terminal in real time, and the drone cluster collaborative unit is triggered to conduct high-frequency status inspections of the changed area.

[0129] It should be noted that the rhythmic memory unit in the physiological adaptation dimming module is a key part of the present invention, which is used to learn the lighting preference pattern of a specific area and ensure that the lighting system can generate a customized lighting plan based on historical data and real-time environmental data. This unit constructs a time series model of regional lighting habits through deep reinforcement learning, and the model training data comes from the historical environmental monitoring records of the multimodal perception network module. Deep reinforcement learning is an algorithm that combines deep learning and reinforcement learning, and can learn lighting preference patterns through historical data. During major events, the system will activate the mode enhancement mechanism, which will fuse the historical optimal lighting parameters with real-time environmental data to generate a customized plan to ensure that the lighting effect meets specific needs.

[0130] Specifically, the rhythmic memory unit first constructs a time series model of regional lighting habits through deep reinforcement learning. The model training data includes historical environmental monitoring records and lighting parameter adjustment records to ensure that the model can accurately predict lighting preferences. During major events, the system will activate the mode enhancement mechanism to fuse the historical optimal lighting parameters with real-time environmental data to generate a customized solution. The solution data is synchronized to the disaster chain blocking module to optimize the lighting parameters of the emergency plan. When a new crowd gathering pattern is detected, the system will start the incremental learning algorithm to update the dimming strategy library to ensure that the lighting system can adapt to new lighting needs. The incremental learning algorithm is an algorithm that can update the model without losing historical data, ensuring that the system can continuously optimize the lighting strategy.

[0131] Preferably, the deep reinforcement learning model can adopt convolutional neural network or recurrent neural network to improve the learning ability and prediction accuracy of the model. The pattern enhancement mechanism can dynamically adjust the lighting parameters according to the needs of different activities to ensure that the lighting effect meets the requirements of specific scenarios. The incremental learning algorithm can dynamically update the model based on real-time monitoring data to ensure that the lighting system can respond to new lighting needs in a timely manner. In addition, the system can optimize the parameter settings of the rhythm memory unit based on historical data and real-time monitoring data to improve the intelligence level and adaptability of the lighting system. These optimization measures can improve the personalized service capabilities of the lighting system and ensure that urban lighting provides a more comfortable and humanized lighting experience while meeting energy-saving requirements.

[0132] The above-mentioned embodiments of the present invention have the following beneficial effects: the present invention can improve the intelligent management level of urban lighting systems through the collaborative work of a multimodal perception network module, a causal reasoning diagnosis module, a destructive topology reconstruction module, a physiological adaptation dimming module and a disaster chain blocking module. The multimodal perception network module can collect millimeter-level structural displacement data, environmental temperature and humidity gradient distribution data and short-term extreme weather forecast data of lighting facilities in real time, and combine the underground pipe network topology data in the geographic information system to provide the system with comprehensive environmental perception capabilities. The causal reasoning diagnosis module can accurately identify the root causes of facility aging through spatiotemporal causal graph modeling and interventional causal analysis, generate a causal feature map, and provide a scientific basis for subsequent topological reconstruction and disaster prevention. The destructive topology reconstruction module can generate a dynamic topology adjustment strategy based on the causal feature map through an adaptive resilience optimization algorithm to ensure that the power supply network can maintain stable operation when key nodes fail.

[0133] The physiological adaptation dimming module can generate color temperature-brightness combination parameters that are consistent with the human body clock through the visual-rhythm coupling model according to the real-time pedestrian activity patterns and atmospheric optical characteristics in the urban area, thereby improving lighting comfort. The disaster chain blocking module can identify the key hub nodes of disaster propagation based on the multi-physics field coupling simulation engine and the historical disaster evolution database, and trigger targeted infrastructure reinforcement instructions to effectively reduce disaster risks. In addition, the drone cluster collaboration unit can maintain the integrity of the perception network in severe weather, and realize emergency obstacle avoidance and formation reorganization of the cluster through a distributed task allocation mechanism driven by game theory and a bionic pulse coupling algorithm. On the whole, the present invention can improve the reliability, energy saving and safety of urban lighting systems, while optimizing the comfort and emergency response capabilities of the urban environment.

[0134] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0135] The above descriptions are only some preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A city lighting control fusion gateway system, characterized in that: Includes the following modules: The multimodal perception network module is used to synchronously collect millimeter-level structural displacement data of lighting facilities, environmental temperature and humidity gradient distribution data, and short-term extreme weather forecast data through high-precision deformation sensors carried by low-altitude drone clusters and micro-environment monitoring units of distributed ground nodes, and integrate underground pipe network topology data in the geographic information system; A causal reasoning diagnosis module is used to perform spatiotemporal causal graph modeling on the multi-dimensional data output by the multimodal perception network module, separate the fundamental causes of facility aging from the apparent correlation factors through interventional causal analysis, and generate a causal feature map; A survivability topology reconstruction module, which is used to generate a dynamic topology adjustment strategy based on the causal feature map through an adaptive resilience optimization algorithm of the power supply network, wherein the algorithm includes a quantitative evaluation of the failure probability of key nodes and an intelligent planning of multi-path redundant power supply links; The physiological adaptation dimming module is used to generate color temperature-brightness combination parameters that conform to the human body clock through the visual-rhythm coupling model according to the real-time pedestrian and vehicle flow and atmospheric optical characteristics in the urban area, and send them to the intelligent dimming terminal; Disaster chain blocking module, which is used to identify key hubs of disaster propagation based on a multi-physics field coupling simulation engine and a historical disaster evolution database, and trigger targeted infrastructure reinforcement instructions; The construction of the visual-rhythm coupling model includes: Establish the response curve of human retinal ganglion cells and quantify the visual comfort index under different color temperatures; Integrate wireless positioning data to build a heat map of human flow activities and dynamically divide high-density activity areas and silent areas; The energy-saving efficiency and physiological adaptation requirements are balanced through a dual-objective optimization algorithm, where the objective function is: in, It is a dynamic weight factor, which is automatically adjusted according to the real-time electricity price and crowd density. For energy efficiency, For biorhythm adaptation, To optimize the target; the optimized color temperature-brightness parameters are sent to the intelligent dimming terminal and synchronized to the disaster chain blocking module as the benchmark parameters of the emergency lighting plan.

2. The urban lighting control fusion gateway system according to claim 1 is characterized in that: The causal reasoning diagnosis module includes: The causal graph construction unit is used to establish a conditional independence relationship network between facility state variables and multi-source environmental variables. The specific execution steps include: An initial causal graph is constructed based on the counterfactual reasoning framework, with nodes including metal fatigue, salt spray corrosion rate, and instantaneous wind load. The marginal causal contribution of each variable to the failure rate is calculated by the gradient intervention method. If the contribution exceeds the threshold, a directed causal edge is established; The root cause mining unit is used to extract the key cause chain from the cause-effect diagram, including: Identify the shortest causal path leading to the failure event and quantify the time-varying impact weight of each node in the path; A causal strength attenuation model is constructed to calculate the residual effects of historical intervention measures on the current failure rate, and the residual effects are input into the invulnerability topology reconstruction module to optimize the redundant design of the power supply link.

3. The urban lighting control fusion gateway system according to claim 2 is characterized in that: The execution steps of the adaptive toughness optimization algorithm include: Construct a dynamic vulnerability heat map of the power supply network, where the node vulnerability is calculated jointly by service life, leakage history, and the intensity of surrounding geological activities; When it is detected that the vulnerability of key hub nodes exceeds the critical value, the three-level response mechanism is activated: The first-level response generates a backup power supply plan based on the minimum cut set, and preferentially calls the low residual effect nodes identified in the causal reasoning diagnosis module as the backup power supply access points; The secondary response optimizes the balanced line load distribution through impedance matching. Specifically, an asymmetric power flow constraint model is used to ensure that the load fluctuation rate of each branch is less than 5%, and the optimized load data is synchronized to the disaster chain blocking module to update the disaster propagation model. The third-level response triggers the coordinated discharge of energy storage devices in the neighboring areas, and the discharge strategy is dynamically adjusted according to the energy consumption demand for the next hour predicted by the physiological adaptation dimming module.

4. The urban lighting control fusion gateway system according to claim 3 is characterized in that: The multi-physics coupling simulation engine includes: Map meteorological data, geological deformation data and power supply network status to a unified spatiotemporal grid; The interaction process of wind force, structure and electromagnetic field is simulated by bidirectional coupling solver to predict the stress concentration areas of key facilities; When it is detected that the cumulative damage value of a certain grid exceeds the preset threshold, the prestress compensation instruction is triggered; the cumulative damage value is shown in the following formula; ,in, is the real-time stress, is the material yield strength, is the accumulated damage value; the compensation instruction synchronously drives the anti-destruction topology reconstruction module to start the secondary response, and adjusts the load distribution of the power supply network to reduce the pressure of the damaged line.

5. The urban lighting control fusion gateway system according to claim 1 is characterized in that: The multimodal perception network module also includes: The drone swarm coordination unit is used to maintain the integrity of the perception network in severe weather. The specific steps include: Construct a dynamic inspection priority map based on risk entropy value. The risk entropy is shown in the following formula; ,in is the sub-area failure probability, is the risk entropy value; a distributed task allocation mechanism driven by game theory is adopted to input the priority map into the decision-making system of the drone, so that the drone cluster can autonomously optimize the coverage path according to the map data when the communication is interrupted; The bionic pulse coupling algorithm is used to realize emergency obstacle avoidance and formation reorganization of the cluster. The obstacle avoidance strategy calls the activity heat map data in the physiological adaptation dimming module in real time to avoid areas with dense crowds.

6. The urban lighting control fusion gateway system according to claim 5 is characterized in that: The implementation of the game theory driven mechanism includes: Define the multi-dimensional utility function of drones, including coverage benefits, energy consumption costs and risk aversion factors. The specific formula is: in, is the dynamic weight, is the drone utility function, is the target area coverage, is the energy cost, is the collision risk index; a Bayesian Nash equilibrium model under incomplete information is constructed, and individual decisions are optimized through online strategy iteration. The decision results are fed back to the anti-destruction topology reconstruction module in real time to adjust the inspection priority of the power supply network; A reputation mechanism is introduced to punish individuals with frequent conflicts and improve cluster collaboration efficiency. The reputation data is synchronized to the disaster chain blocking module to evaluate the regional safety factor.

7. The urban lighting control fusion gateway system according to claim 3 is characterized in that: The disaster chain blocking module also includes: The key hub identification unit is used to locate super nodes in the disaster propagation network. The specific steps include: Calculate the weighted comprehensive index of node betweenness centrality and eigenvector centrality, and the weights are dynamically allocated according to the proportion of the chain effect of node failure in historical disaster data; Identify the key transmission paths that are frequently accessed, and input the path data into the multimodal perception network module to guide the drone cluster to focus on monitoring high-risk nodes; When it is detected that the comprehensive influence index of a node exceeds 3 standard deviations of the global mean, it is marked as a super node that needs priority reinforcement, and the three-level response mechanism of the anti-destruction topology reconstruction module is triggered, calling the energy storage device to perform redundant protection on the power supply link of the node.

8. The urban lighting control fusion gateway system according to claim 1 is characterized in that: The physiological adaptation dimming module also includes: Rhythm memory unit, used to learn the lighting preference pattern of a specific area, the specific steps include: A time series model of regional lighting habits is constructed through deep reinforcement learning, and the model training data comes from the historical environmental monitoring records of the multimodal perception network module; During major events, the mode enhancement mechanism is activated to integrate historical optimal lighting parameters with real-time environmental data to generate customized solutions. The solution data is synchronized to the disaster chain blocking module to optimize the lighting parameters of the emergency plan; When a new crowd gathering pattern is detected, the incremental learning algorithm is activated to update the dimming strategy library. The updated strategy is sent to the smart dimming terminal in real time, and the drone cluster collaborative unit is triggered to conduct high-frequency status inspections of the changed area.

Images