Control method and system of charging square cabin

Abstract

The invention provides a control method and system for a charging shelter, and the method comprises the steps: receiving a power demand signal of equipment, comprehensively analyzing the power demand, the working state of a charging station, the remaining power resources and the emergency degree of the equipment through a multi-objective optimization algorithm, and generating an optimal charging station scheduling scheme and a charging path. And performing power demand trend prediction by adopting a machine learning algorithm in combination with historical data and current environmental factors to form a charging preparation plan. The power consumption, the connection state and the environmental parameters of the equipment are monitored in real time through the Internet of Things technology, abnormal conditions are detected, and an emergency processing instruction is generated. And comprehensively evaluating the emergency processing instruction and the charging completion condition by using a big data analysis technology, and generating a charging effect evaluation report. According to the invention, the charging efficiency, the resource utilization rate and the system reliability are improved, and the quick response capability to abnormal conditions and the overall operation effect are enhanced.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of charging cabins, and in particular to a control method and system for a charging cabin. Background Art

[0002] With the increasing number of medical devices and the continuous improvement of their intelligence level, the power supply of medical equipment has become an important issue in hospital management, especially in emergency medical rescue and temporary medical facilities, such as mobile medical vehicles and field hospitals. It is particularly important to provide power support for medical equipment quickly and efficiently.

[0003] However, the existing charging cabin control methods have obvious deficiencies in response speed, scheduling efficiency, monitoring and processing capabilities, and data analysis. The current control methods mainly include manual scheduling, simple algorithm scheduling, basic monitoring systems, and preliminary data analysis. Manual scheduling and simple algorithm scheduling cannot respond quickly in emergency situations, resulting in charging delays. In addition, the scheduling has a single goal and cannot comprehensively consider multiple dimensions such as charging efficiency, cost-effectiveness, equipment urgency, and path safety, resulting in low overall scheduling efficiency. There is a lack of intelligent anomaly detection and processing mechanisms, and it is impossible to promptly discover and handle abnormal situations during the charging process, which increases safety risks. At the level of data statistics and simple trend analysis, there is a lack of in-depth mining and optimization suggestions for charging data, and it is impossible to effectively improve the overall operational efficiency and service quality of the charging cabin. Summary of the invention

[0004] The embodiments of the present invention provide a control method and system for a charging cabin, which are used to solve the obvious deficiencies in the prior art in terms of response speed, scheduling efficiency, monitoring and processing capabilities, and data analysis, such as the inability to respond quickly in emergency situations, charging delays, low overall scheduling efficiency, lack of intelligent anomaly detection and processing mechanisms, inability to promptly discover and process abnormal situations during the charging process, increased safety risks, lack of in-depth mining of charging data and optimization suggestions, and inability to effectively improve the overall operational efficiency and service quality of the charging cabin.

[0005] In a first aspect, an embodiment of the present invention provides a control method for a charging cabin, including:

[0006] Receiving a power demand signal from a device, the power demand signal including: the type of device, the current power level, the required charging time, and the urgency level of the device;

[0007] A multi-objective optimization algorithm is used to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain an optimal charging station scheduling plan and charging path;

[0008] Using a machine learning algorithm, the optimal charging station scheduling scheme and charging path are combined with historical data and current environmental factors to predict the power demand trend and form a charging preparation plan;

[0009] The power consumption of the equipment, the connection status with the charging cabin and the surrounding environment parameters during the implementation of the charging preparation plan are monitored in real time through the Internet of Things technology, and when an abnormal situation is detected, an emergency processing instruction is generated;

[0010] Big data analysis technology is used to conduct a comprehensive evaluation of the emergency handling instructions and the situation after charging is completed to obtain a charging effect evaluation report.

[0011] Optionally, a multi-objective optimization algorithm is used to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain an optimal charging station scheduling plan and charging path, including:

[0012] A genetic algorithm and a particle swarm optimization algorithm are used to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain a set of charging station candidates that meet the preliminary conditions;

[0013] The charging station candidate set is classified by using a fuzzy C-means clustering algorithm to obtain multiple charging station clusters, and a virtual agent model is established within the charging station cluster using a reinforcement learning algorithm to simulate the charging efficiency and resource utilization under different charging strategies, and generate an actual performance evaluation of each charging station;

[0014] By using Bayesian network analysis, combined with historical charging data and current environmental variables, the working state of the actual performance evaluation is predicted to obtain the working state change trend of each charging station cluster;

[0015] By adopting the fusion technology of ant colony algorithm and Dijkstra algorithm, based on the working state change trend, the optimal path from the charging cabin entrance to the target charging station is found, and the optimal path solution is determined;

[0016] Comprehensively evaluate the optimal path plan through a multi-criteria decision analysis method, calculate charging efficiency, cost-effectiveness, equipment urgency and path safety, and select the optimal charging station scheduling plan and charging path;

[0017] Optionally, the optimal path plan is comprehensively evaluated by a multi-criteria decision analysis method, the charging efficiency, cost-effectiveness, equipment urgency and path safety are calculated, and the optimal charging station scheduling plan and charging path are selected, including:

[0018] The importance of the four dimensions of charging efficiency, cost-effectiveness, equipment urgency, and route safety was quantified using the analytic hierarchy process, and an evaluation index system was constructed to obtain the weight value of each dimension.

[0019] The data envelopment analysis method is used to evaluate the efficiency of the performance of the optimal path solution under the evaluation index system, and the comprehensive efficiency score of each path solution is calculated by combining the weight value of each dimension;

[0020] Applying grey correlation analysis and combining the comprehensive efficiency score, the correlation between each path plan and the ideal plan is calculated to obtain the correlation ranking of the path plans;

[0021] Using fuzzy comprehensive evaluation method to refine the correlation ranking, calculate the influence of uncertainty and subjective judgment, and generate the final path evaluation result;

[0022] The final path evaluation result and historical charging data are analyzed through an integrated learning algorithm to predict charging demand, dynamically adjust the charging station scheduling plan, form an adjusted charging station scheduling plan, and select the optimal charging station scheduling plan and charging path based on the adjusted charging station scheduling plan.

[0023] Optionally, a data envelopment analysis method is used to evaluate the efficiency of the performance of the optimal path solution under the evaluation index system, and the comprehensive efficiency score of each path solution is calculated in combination with the weight value of each dimension, including:

[0024] The data envelopment analysis method is used to evaluate the efficiency of the optimal path solution in four dimensions: charging efficiency, cost-effectiveness, equipment urgency, and path safety, and the efficiency value of each path solution in each dimension is obtained;

[0025] Multiplying the efficiency value of each dimension by the weight value of each dimension obtained by the hierarchical analysis method to obtain the weighted efficiency value of each path plan in each dimension;

[0026] Normalizing the weighted efficiency value to eliminate the dimension effect and obtain a normalized weighted efficiency value;

[0027] Using principal component analysis to perform dimensionality reduction processing on the normalized weighted efficiency value, extracting the main components, and obtaining main component data;

[0028] A multi-layer perceptron neural network is used to perform nonlinear mapping on the main component data, comprehensively calculate the interactions between the path plans, and generate a preliminary comprehensive efficiency score for each path plan;

[0029] The preliminary comprehensive efficiency scores are randomly sampled through Monte Carlo simulation to evaluate the stability of the scores and determine the comprehensive efficiency score of each path solution.

[0030] Optionally, a multi-layer perceptron neural network is used to perform nonlinear mapping on the main component data, comprehensively calculate the interactions between the path solutions, and generate a preliminary comprehensive efficiency score for each path solution, including:

[0031] A multi-layer perceptron neural network model is used to perform nonlinear mapping on the main component data, wherein the input layer receives the main component data, and the output layer is used to output a preliminary comprehensive efficiency score of each path plan;

[0032] Introducing an attention mechanism into the multi-layer perceptron neural network model to perform weighted processing on the main component data, highlighting the influence of key components, and obtaining weighted main component data;

[0033] Using a long short-term memory network to model the time series characteristics of the weighted principal component data, capturing the time-varying interactions between path plans and generating time series features;

[0034] Extract local features of the time series features through a convolutional neural network, identify the spatial correlation between path plans, and obtain spatial and temporal feature vectors;

[0035] The spatial and temporal feature vectors are input into the fully connected layer, and after nonlinear transformation, the interactions between the path plans are comprehensively calculated to generate a preliminary comprehensive efficiency score for each path plan.

[0036] Optionally, a machine learning algorithm is used to predict the power demand trend of the optimal charging station scheduling scheme and charging path in combination with historical data and current environmental factors to form a charging preparation plan, including:

[0037] Analyzing the optimal charging station scheduling scheme and historical data of charging paths using a time series analysis algorithm to identify periodic and seasonal patterns of power demand and obtain historical demand patterns;

[0038] Using an integrated learning algorithm, combining the historical demand pattern and current environmental factors, to predict power demand and generate preliminary power demand forecast results;

[0039] Using a deep learning algorithm to refine the preliminary power demand forecast result, introducing equipment type and urgency level as additional inputs, optimizing the forecast accuracy, and obtaining an optimized power demand forecast result;

[0040] Through simulation technology, based on the optimized power demand forecast results, the operation of the charging cabin under different charging strategies is simulated, the effects of various strategies are evaluated, and a variety of alternative charging preparation plans are formed;

[0041] A multi-objective optimization algorithm is applied to comprehensively evaluate the various alternative charging preparation plans, calculate charging efficiency, resource utilization, equipment urgency and user satisfaction, and select the best charging preparation plan.

[0042] Optionally, the power consumption of the equipment, the connection status with the charging cabin and the surrounding environmental parameters during the implementation of the charging preparation plan are monitored in real time by the Internet of Things technology, and when an abnormality is detected, an emergency processing instruction is generated, including:

[0043] The Internet of Things technology is used to collect real-time data on the power consumption of the equipment, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan to obtain real-time monitoring data;

[0044] An anomaly detection algorithm is used to analyze the real-time monitoring data, identify abnormal power consumption, abnormal connection status, and abnormal environmental parameters, and generate an anomaly detection result;

[0045] Applying situational awareness technology in combination with the abnormal detection results and the urgency level of the device to assess the severity and urgency of the abnormal situation and form a situational awareness assessment report;

[0046] Using an expert system, based on the situational awareness assessment report, calling a predefined emergency handling rule base, matching an emergency handling strategy, and generating preliminary emergency handling instructions;

[0047] The preliminary emergency handling instructions are optimized through a multi-agent collaborative decision-making algorithm, the charging requirements and resource allocation of other equipment in the charging cabin are calculated, and the final emergency handling instructions are generated.

[0048] In a second aspect, an embodiment of the present application provides a control system for a charging cabin, including:

[0049] A receiving module receives a power demand signal from a device, wherein the power demand signal includes: the type of device, the current power level, the required charging time, and the urgency level of the device;

[0050] The analysis module uses a multi-objective optimization algorithm to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain the optimal charging station scheduling plan and charging path;

[0051] A prediction module, using a machine learning algorithm, predicts the power demand trend based on the optimal charging station scheduling plan and charging path, combined with historical data and current environmental factors, to form a charging preparation plan;

[0052] A monitoring module, which uses the Internet of Things technology to monitor the power consumption of the equipment, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan in real time, and generates emergency processing instructions when an abnormal situation is detected;

[0053] The evaluation module uses big data analysis technology to comprehensively evaluate the emergency handling instructions and the situation after charging is completed to obtain a charging effect evaluation report.

[0054] In a third aspect, an embodiment of the present invention provides a computing device, including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a control method for a charging cabin as described in any one of the first aspects.

[0055] In a fourth aspect, an embodiment of the present invention provides a computer storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement a method for controlling a charging cabin as described in any one of the first aspects.

[0056] In an embodiment of the present invention, a power demand signal is received from a device, wherein the power demand signal includes: the type of device, the current power level, the required charging time, and the urgency level of the device; a multi-objective optimization algorithm is used to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the device to obtain an optimal charging station scheduling plan and charging path; a machine learning algorithm is used to predict the power demand trend of the optimal charging station scheduling plan and charging path in combination with historical data and current environmental factors to form a charging preparation plan; the power consumption of the device, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan are monitored in real time through the Internet of Things technology, and when an abnormal situation is detected, an emergency processing instruction is generated; and big data analysis technology is used to comprehensively evaluate the emergency processing instructions and the situation after charging is completed to obtain a charging effect evaluation report.

[0057] The technical solution of the present invention has the following beneficial effects:

[0058] Rapidly respond to the power needs of different devices through multi-objective optimization algorithms, comprehensively calculate multiple factors to generate optimal charging station scheduling plans and paths, significantly improve charging efficiency and reduce waiting time. Use machine learning algorithms combined with historical data and environmental factors to predict power demand trends and form an orderly and efficient charging preparation plan. Use IoT technology to monitor power consumption, connection status and environmental parameters in real time, promptly detect and handle abnormal situations, and ensure charging safety and stability. Use big data analysis technology to evaluate emergency handling instructions and charging effects, and provide data support for continuous optimization;

[0059] Furthermore, the genetic algorithm and particle swarm optimization algorithm are used to comprehensively analyze the power demand signal, the working status of the charging station, the remaining power resources and the urgency of the equipment to generate a reasonable and diverse candidate set of charging stations. The fuzzy C-means clustering algorithm is used to classify the candidate set, and the reinforcement learning algorithm is used to establish a virtual agent model to evaluate the performance of the charging station. The Bayesian network analysis is used to predict the trend of working status changes, making the scheduling more flexible and adaptable to future changes. The ant colony algorithm and the Dijkstra algorithm are integrated to find the optimal path to ensure that the path selection is scientific and reasonable. Finally, the charging efficiency, cost-effectiveness, urgency and safety are comprehensively evaluated through a multi-criteria decision analysis method, and the optimal charging station scheduling plan and path are selected to improve the overall operational efficiency and service quality. These aspects or other aspects of the present invention will be more concise and easy to understand in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0061] Figure 1 A flow chart of a control method for a charging cubicle provided in an embodiment of the present invention;

[0062] Figure 2 A schematic diagram of the structure of a control system of a charging cabin provided by an embodiment of the present invention;

[0063] Figure 3 A schematic diagram of the structure of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0064] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions 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.

[0065] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0067] Since the existing charging cabin control methods have obvious deficiencies in response speed, scheduling efficiency, monitoring and processing capabilities, and data analysis, the current control methods mainly include manual scheduling, simple algorithm scheduling, basic monitoring system, and preliminary data analysis. Manual scheduling and simple algorithm scheduling cannot respond quickly in emergency situations, resulting in charging delays. Based on this, the present invention provides a control method for a charging cabin, such as Figure 1 ,include:

[0068] Step 101: receiving a power demand signal from a device, wherein the power demand signal includes: the type of device, the current power level, the required charging time, and the urgency level of the device;

[0069] In this step, the power demand signal includes the type of device, the current power level, the required charging time and the urgency level of the device, which are used to determine the charging priority and resource allocation of the device;

[0070] Device type refers to the type of medical equipment, such as ECG monitors, ventilators, etc. Different types of equipment may have different charging requirements and priorities; Current power refers to the current battery power of the device, which is used to determine the urgency of charging the device; Required charging time refers to the estimated time required for the device to complete charging, which is used for scheduling and route planning; The urgency level is divided into different urgency levels, such as high, medium, and low, according to the importance and usage of the equipment.

[0071] Suppose an ECG monitor on a mobile medical vehicle sends out a power demand signal. The signal contains the following information: device type: ECG monitor, current power level: 20%, required charging time: 30 minutes, urgency level: high. After the system receives the signal, it will use it as input for subsequent processing.

[0072] Step 102: using a multi-objective optimization algorithm to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the emergency level of the equipment to obtain an optimal charging station scheduling plan and charging path;

[0073] In this step, the multi-objective optimization algorithm is used to generate the optimal charging station scheduling plan and charging path by integrating multiple objectives; the working status of the charging station refers to the current usage of each charging station, such as whether it is idle, charging progress, etc.; the remaining power resources refer to the total amount of power resources available in the charging cabin; the equipment urgency level refers to the charging priority of the equipment;

[0074] The genetic algorithm and particle swarm optimization algorithm are used to comprehensively analyze the above information and generate a set of candidate charging stations that meet the preliminary conditions. For example, two idle charging stations A and B are identified, of which A is closer but has a lower charging rate, and B is slightly farther but has a higher charging rate. By comprehensively calculating the urgency level of the ECG monitor and other factors, charging station B is finally selected as the optimal charging station, and the shortest path from the mobile medical vehicle to charging station B is planned.

[0075] Step 103: using a machine learning algorithm to predict the power demand trend of the optimal charging station scheduling plan and charging path in combination with historical data and current environmental factors to form a charging preparation plan;

[0076] In this step, machine learning algorithms are used to combine historical data and current environmental factors to predict future electricity demand trends and form a charging preparation plan. Historical data refers to previous charging records, equipment usage frequency, etc.; current environmental factors include weather conditions, holidays and other factors that affect charging demand.

[0077] For example, by using time series analysis algorithms and ensemble learning methods to analyze historical charging data, it was found that the charging demand for ECG monitors is higher in the afternoon. Combined with current environmental factors, it is predicted that the charging demand for ECG monitors will increase in the future, and a charging preparation plan is made in advance to ensure sufficient charging resources.

[0078] Step 104: Using the Internet of Things technology, the power consumption of the equipment, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan are monitored in real time, and when an abnormal situation is detected, an emergency processing instruction is generated;

[0079] In this step, sensors and network connections in the Internet of Things technology are used to monitor the power consumption during the charging process, the connection status between the device and the charging cabin, and the surrounding environmental parameters in real time; the power consumption refers to the real-time power changes of the device during the charging process; the connection status refers to the connection stability between the device and the charging station; the environmental parameters include environmental factors such as temperature and humidity;

[0080] The charging process of the ECG monitor is monitored in real time through sensors installed on the charging station and equipment. When an abnormal charging current or unstable connection is detected, emergency processing instructions are generated. For example, if a sudden drop in charging current is detected, which may be due to poor contact or cable failure, it will automatically switch to the backup charging station and notify maintenance personnel to check.

[0081] Step 105: Use big data analysis technology to comprehensively evaluate the emergency handling instructions and the situation after charging is completed to obtain a charging effect evaluation report;

[0082] In this step, the emergency processing instructions and the situation after charging are comprehensively evaluated through big data analysis technology to generate a charging effect evaluation report, where the emergency processing instructions refer to the processing instructions generated when an abnormal situation is detected; the situation after charging includes the state of the device after charging, charging time, charging efficiency, etc.

[0083] Collect and analyze emergency handling instructions and ECG monitor data after charging is completed, and generate a charging effect evaluation report, which includes the following: Charging efficiency: comparison between actual charging time and expected charging time; Cost-effectiveness: comparison between the cost of this charging and the expected cost; Safety: whether any abnormal situation occurs during the charging process and the results of its handling; User satisfaction: evaluation based on user feedback and device usage;

[0084] Through big data analysis, charging strategies can be continuously optimized to improve overall operational efficiency and service quality. For example, if the same type of abnormal situation occurs multiple times, relevant parameters can be automatically adjusted or hardware facilities can be improved to reduce the occurrence of similar problems.

[0085] The embodiments of the present invention improve the charging response speed and scheduling efficiency through the above steps, enhance the monitoring and exception handling capabilities, provide in-depth data analysis and optimization suggestions, thereby improving the overall operational efficiency and service quality of the charging cabin, and ensuring timely and reliable power supply to medical equipment.

[0086] Based on this, the present invention provides a specific embodiment, wherein step 102 receives collection rule parameters input by a user, wherein the collection rule parameters include collection frequency, collection type, and collection scope, and specifically includes the following steps:

[0087] Step 201: using a genetic algorithm and a particle swarm optimization algorithm to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment, and obtain a set of charging station candidates that meet preliminary conditions;

[0088] In this step, genetic algorithm refers to a search optimization algorithm that simulates natural selection and genetic mechanisms, and finds the optimal solution through selection, crossover and mutation operations; particle swarm optimization algorithm refers to an optimization algorithm based on swarm intelligence, which finds the optimal solution by simulating the flight behavior of bird flocks; power demand signals include device type, current power, required charging time and urgency level; charging station working status refers to the current usage of each charging station, such as whether it is idle, charging progress, etc.; remaining power resources refer to the total amount of power resources available in the charging cabin; equipment urgency level refers to the charging priority of the equipment;

[0089] Assuming that there are multiple devices on a mobile medical vehicle that need to be charged, when the power demand signals of these devices are received, the genetic algorithm and particle swarm optimization algorithm are used to comprehensively analyze these signals as well as the working status of the charging station in the charging cabin, the remaining power resources and the equipment urgency level, and generate a set of charging station candidate sets that meet the preliminary conditions. For example, three idle charging stations A, B and C are identified, among which A is close but has a low charging rate, B is slightly far away but has a high charging rate, and C is moderately close and has a medium charging rate. The urgency of the equipment and other factors are comprehensively calculated to generate a candidate set containing these three charging stations.

[0090] Step 202: using a fuzzy C-means clustering algorithm to classify the charging station candidate set to obtain multiple charging station clusters, using a reinforcement learning algorithm to establish a virtual agent model within the charging station cluster, simulating the charging efficiency and resource utilization under different charging strategies, and generating an actual performance evaluation of each charging station;

[0091] In this step, the fuzzy C-means clustering algorithm refers to a clustering method based on fuzzy set theory, which can divide data points into multiple clusters and allow data points to belong to multiple clusters; the charging station cluster refers to a group of charging stations formed according to factors such as the location, charging capacity, and current load of the charging station; the reinforcement learning algorithm refers to a machine learning method that learns the optimal strategy in the environment through an agent model; the virtual agent model is used to simulate the charging efficiency and resource utilization under different charging strategies;

[0092] The fuzzy C-means clustering algorithm is used to classify the charging station candidate set generated in step 1 to form multiple charging station clusters. For example, charging stations A, B and C are divided into two clusters: cluster 1 contains A and C, and cluster 2 contains B. Then, a virtual agent model is established using the reinforcement learning algorithm to simulate the charging efficiency and resource utilization under different charging strategies. For example, three different charging strategies are simulated: fast charging, balanced charging and energy-saving charging, and the actual performance evaluation results of each charging station are generated.

[0093] Step 203: By using Bayesian network analysis, combining historical charging data and current environmental variables, the working state of the actual performance evaluation is predicted to obtain the working state change trend of each charging station cluster;

[0094] In this step, the Bayesian network is a probabilistic graphical model used to represent the dependencies between variables and perform probabilistic reasoning;

[0095] Among them, historical charging data includes previous charging records, including charging time, charging amount, equipment type, etc.; current environmental variables refer to environmental factors that affect charging demand, such as weather conditions, holidays, etc.; working status change trends refer to the usage forecast of the charging station in the future.

[0096] Through Bayesian network analysis, combined with historical charging data and current environmental variables, the actual performance evaluation generated in step 2 is used to predict the working status. For example, it is found that the charging demand is high in the afternoon in the past week. Combined with the current weather forecast showing rainfall in the next few hours, it is predicted that the utilization rate of charging stations A and C will increase in the future, while the utilization rate of charging station B will remain stable. A working status change trend report for each charging station cluster is generated.

[0097] Step 204: using the fusion technology of the ant colony algorithm and the Dijkstra algorithm, based on the working state change trend, finding the optimal path from the charging cabin entrance to the target charging station, and determining the optimal path solution;

[0098] In this step, the ant colony algorithm is an optimization algorithm that simulates the foraging behavior of ants and finds the shortest path through the pheromone update mechanism; the Dijkstra algorithm is a classic shortest path algorithm used to find the shortest path from the starting point to the end point in the graph; the working status change trend refers to the working status change trend of the charging station cluster generated in step 3; the optimal path solution refers to the shortest and optimal path from the entrance of the charging cabin to the target charging station.

[0099] The ant colony algorithm and Dijkstra algorithm fusion technology are used to find the optimal path from the entrance of the charging cabin to the target charging station based on the working state change trend generated in step 3. For example, the system simulates multiple paths from the mobile medical vehicle to charging stations A, B, and C. Through the pheromone update mechanism of the ant colony algorithm and the shortest path calculation of the Dijkstra algorithm, the optimal path from the mobile medical vehicle to charging station B is determined. This path is not only the shortest, but also takes into account the high charging rate of charging station B and the future working state change trend.

[0100] Step 205: Comprehensively evaluate the optimal path plan through a multi-criteria decision analysis method, calculate charging efficiency, cost-effectiveness, equipment urgency and path safety, and select the optimal charging station scheduling plan and charging path;

[0101] In this step, the multi-criteria decision analysis method refers to a decision-making method that comprehensively considers multiple evaluation indicators, such as hierarchical analysis method, data envelopment analysis, etc. Charging efficiency refers to the ability to complete charging in unit time; cost-effectiveness refers to the cost-benefit ratio of the charging process; equipment urgency refers to the charging priority of the equipment; path safety refers to the safety and reliability of the path;

[0102] The optimal path plan is comprehensively evaluated through multi-criteria decision analysis methods. For example, the hierarchical analysis method is used to quantify the charging efficiency, cost-effectiveness, equipment urgency and path safety, and an evaluation index system is constructed. Then, the data envelopment analysis method is used to evaluate the efficiency of the path plan under each evaluation index, and the comprehensive efficiency score of each path plan is calculated by combining the weight value. Finally, the optimal charging station scheduling plan and charging path are selected through grey correlation analysis and fuzzy comprehensive evaluation method. For example, the path from the mobile medical vehicle to charging station B is finally selected because it performs well in terms of charging efficiency, cost-effectiveness, equipment urgency and path safety.

[0103] Among them, the calculation method of the optimal charging station scheduling plan and charging path is as follows:

[0104] S i =α·E i +β·C i +γ·U i +δ·S i +η·I i +θ·D i +φ·P i

[0105] Among them, S i is the comprehensive evaluation score of the i-th path plan; E i is the charging efficiency of the i-th path scheme; C i is the cost-effectiveness of the i-th path plan; Ui is the equipment urgency of the i-th path solution; S i is the path security of the i-th path solution; I i is the interaction factor of the i-th path plan; D i is the dynamic adaptability of the i-th path solution; P i is the predicted reliability of the i-th path plan; α, β, γ, δ, η, θ, φ are the weights of each dimension, which are determined by the analytic hierarchy process (AHP) or other multi-criteria decision-making methods, which usually involve expert evaluation and historical data analysis to ensure that the importance of different dimensions is reasonably reflected and satisfy α+β+γ+δ+η+θ+φ=1;

[0106] Wherein, the charging efficiency E i , cost-effectiveness C i 、Equipment urgency level U i , Path Security S i , prediction reliability P i , Interaction Factor I i , Dynamic Adaptability D i and the predicted reliability P i The calculation method is as follows:

[0107] Among them, the charging efficiency E i The calculation method is as follows:

[0108]

[0109] Among them, E i Indicates charging efficiency; Q i is the total charging capacity of the i-th path plan, obtained through the charging data recorded by the sensors or management system of the charging station; T i is the total charging time of the i-th path plan obtained through the time data recorded by the timer or management system of the charging station;

[0110] Among them, cost-effectiveness C i The calculation method is as follows:

[0111]

[0112] Among them, C i Represents cost efficiency; V i is the value of the i-th path solution, which is determined by the importance and frequency of use of the equipment and can be obtained through historical data and equipment classification; C i is the total cost of the i-th path plan, which is obtained through the electricity fee record and maintenance cost record of the charging station;

[0113] Among them, the equipment urgency level U iThe calculation method is:

[0114]

[0115] Among them, U i Indicates the urgency of the equipment; U i,base is the basic urgency of the device in the i-th path solution, which is determined by information such as device type, current power, and required charging time; U i,threshold is the value of the urgency, which is set according to historical data and experience; k is the sensitivity parameter, which is determined through experiments and historical data analysis to ensure the smoothness of the Sigmoid function;

[0116] Among them, the path security S i The calculation method is as follows:

[0117]

[0118] Among them, S i represents the path security, S i,base is the basic safety of the path in the i-th path plan, calculated based on historical accident data, environmental factors, etc.; S i,threshold is the safety threshold, which is set according to historical data and experience; k is the sensitivity parameter, which is determined through experiments and historical data analysis to ensure the smoothness of the Sigmoid function;

[0119] Among them, the interaction factor I i The calculation method is as follows:

[0120]

[0121] Among them, w ij is the interaction weight between the i-th path plan and the j-th path plan, and the mutual influence weight between different path plans is determined through expert evaluation or historical data analysis; E j and E i , is the charging efficiency of the j-th path scheme and the i-th path scheme; C j and C i is the cost-effectiveness of the j-th path plan and the i-th path plan; U j and U i is the equipment urgency of the j-th path plan and the i-th path plan; S j and S i is the path security of the j-th path solution and the i-th path solution;

[0122] Among them, dynamic adaptability D i The calculation method is as follows:

[0123]

[0124] Among them, D i,base is the adaptability of the i-th path plan in a dynamic environment. By comparing historical data with real-time data, the stability and flexibility of the path plan in different environments are evaluated; D i,threshold is the adaptability threshold, which is set according to actual needs and experience; k is the sensitivity parameter, which is determined through expert experience and historical data analysis;

[0125] Among them, the prediction reliability P i The calculation method is as follows:

[0126]

[0127] Among them, P i,base is the prediction reliability of the i-th path plan, and the prediction reliability of the path plan is evaluated by the error rate and confidence interval of the prediction model; P i,threshold is the reliability threshold, which is set according to actual needs and experience; k is the sensitivity parameter, which is obtained by determining the sensitivity parameter through expert experience and historical data analysis.

[0128] The embodiments of the present invention improve the charging response speed and scheduling efficiency through the above steps, enhance the monitoring and exception handling capabilities, provide in-depth data analysis and optimization suggestions, and improve the overall operating efficiency and service quality of the charging cabin.

[0129] Based on this, the present invention provides a specific embodiment, wherein the step 205 comprehensively evaluates the optimal path plan through a multi-criteria decision analysis method, calculates charging efficiency, cost-effectiveness, equipment urgency and path safety, and selects the optimal charging station scheduling plan and charging path, specifically including the following steps:

[0130] Step 301: quantify the importance of the four dimensions of charging efficiency, cost-effectiveness, equipment urgency, and route safety using the analytic hierarchy process, construct an evaluation index system, and obtain a weight value for each dimension;

[0131] In this step, the analytic hierarchy process is a multi-criteria decision-making method. By establishing a hierarchical model, complex problems are decomposed into multiple levels, and the importance of each factor is compared layer by layer, and finally the weight of each factor is obtained; the evaluation index system includes four dimensions: charging efficiency, cost-effectiveness, equipment urgency and path safety, which are used to comprehensively evaluate the performance of the charging path solution;

[0132] Assuming that a charging cabin needs to select the optimal charging path solution, firstly, through expert scoring or historical data analysis, a pairwise comparison matrix between the criteria is constructed. For example, the importance of charging efficiency relative to cost-effectiveness is 3, the importance of cost-effectiveness relative to the urgency of equipment is 2, and so on. Then, by solving the maximum eigenvalue root of the judgment matrix and its corresponding eigenvector, the weight of each criterion is obtained after normalization. After calculation, the charging efficiency α = 0.4, the cost-effectiveness β = 0.3, the equipment urgency γ = 0.2, and the path safety δ = 0.1 are obtained.

[0133] Step 302: using a data envelopment analysis method to evaluate the efficiency of the optimal path solution under the evaluation index system, and combining the weight value of each dimension to calculate the comprehensive efficiency score of each path solution;

[0134] In this step, the data envelopment analysis method, a linear programming-based approach, is used to evaluate the relative efficiency of decision-making units with multiple inputs and outputs;

[0135] The comprehensive efficiency score refers to the comprehensive efficiency score of each path plan in all dimensions calculated by the DEA method;

[0136] By collecting specific data on the four dimensions of charging efficiency, cost-effectiveness, equipment urgency, and path safety for each path plan, for example, the data for path plan 1 is: charging efficiency 0.85, cost-effectiveness 0.9, equipment urgency 0.7, and path safety 0.8. The relative efficiency score of each path plan is calculated using the DEA model, assuming that the DEA efficiency score of path plan 1 is 0.85. Then, the comprehensive efficiency score is calculated based on the weights. The comprehensive efficiency score of path plan 1 is S1 = 0.4.0.85 + 0.3 0.9 + 0.2 0.7 + 0.1 0.8 = 0.83;

[0137] Step 303: applying grey correlation analysis, combining the comprehensive efficiency score, calculating the correlation between each path solution and the ideal solution, and obtaining the correlation ranking of the path solutions;

[0138] In this step, grey relational analysis refers to a systematic analysis method that reflects the similarity between factors by calculating the correlation between them; the ideal solution refers to the path solution that achieves the best in all dimensions;

[0139] Assuming that the ideal solution scores 1 in all dimensions, the grey correlation analysis method is used to calculate the correlation between each path solution and the ideal solution. For example, the correlation of path solution 1 is 0.9, and the correlation of path solution 2 is 0.8. The path solutions are sorted from high to low according to the correlation. Path solution 1 has the highest correlation and ranks first.

[0140] Step 304: using a fuzzy comprehensive evaluation method to refine the correlation ranking, calculate the influence of uncertainty and subjective judgment, and generate a final path evaluation result;

[0141] In this step, fuzzy comprehensive evaluation method refers to a multi-criteria decision-making method for processing fuzzy information, which comprehensively evaluates multiple factors through fuzzy mathematical theory; uncertainty refers to uncertainty caused by incomplete data or subjective judgment;

[0142] Through expert scoring or historical data analysis, a fuzzy evaluation matrix is ​​constructed. For example, the fuzzy evaluation matrix of path plan 1 is: , the fuzzy comprehensive evaluation results of each path plan are calculated by the fuzzy comprehensive evaluation method. The fuzzy comprehensive evaluation results of path plan 1 are 0.85, and the fuzzy comprehensive evaluation results of path plan 2 are 0.75. According to the fuzzy comprehensive evaluation results, the final path evaluation results are generated. The final evaluation results of path plan 1 are 0.85, and the final evaluation results of path plan 2 are 0.75;

[0143] Step 305: Analyze the final path evaluation result and historical charging data through an integrated learning algorithm, predict charging demand, dynamically adjust the charging station scheduling plan, form an adjusted charging station scheduling plan, and select the optimal charging station scheduling plan and charging path based on the adjusted charging station scheduling plan;

[0144] In this step, ensemble learning algorithm refers to a machine learning method that improves prediction accuracy by combining the prediction results of multiple models; historical charging data refers to past charging records, including charging time, charging amount, device type, etc.; charging demand prediction refers to predicting future charging demand based on historical data and current environmental variables; dynamic adjustment of charging station scheduling plan refers to dynamically adjusting the charging station scheduling plan based on the prediction results to adapt to actual demand changes.

[0145] Collect charging records over the past period of time, including charging time, charging amount, device type, etc. Use a random forest model or other integrated learning algorithm to train the model and predict future charging demand. For example, historical data shows that in the past week, the charging demand was higher in the afternoon every day. According to the prediction results, dynamically adjust the charging station scheduling plan, increase the number of charging stations in this time period or adjust the charging priority. Finally, based on the adjusted charging station scheduling plan, select the optimal charging station scheduling plan and charging path. For example, select path plan 1 as the optimal plan because it performs well in terms of charging efficiency, cost-effectiveness, equipment urgency, and path safety.

[0146] The embodiment of the present invention can analyze historical charging data, predict future demand and dynamically adjust the scheduling plan through the above steps, thereby improving charging efficiency, cost-effectiveness and safety.

[0147] Based on this, the present invention provides a specific embodiment, in which step 302 uses a data envelopment analysis method to evaluate the efficiency of the performance of the optimal path solution under the evaluation index system, and calculates the comprehensive efficiency score of each path solution in combination with the weight value of each dimension, which specifically includes the following steps:

[0148] Step 401: using a data envelopment analysis method to evaluate the performance of the optimal path solution in four dimensions: charging efficiency, cost-effectiveness, equipment urgency, and path safety, and obtain the efficiency value of each path solution in each dimension;

[0149] In this step, the data envelopment analysis method refers to a method based on linear programming, which is used to evaluate the relative efficiency of decision-making units with multiple inputs and outputs. By comparing the performance of various path plans in different dimensions, the efficiency value of each path plan in each dimension is calculated;

[0150] Among them, charging efficiency refers to the ability to complete charging in a unit of time, cost-effectiveness refers to the cost-benefit ratio of the charging process, equipment urgency refers to the charging priority of the equipment, and path safety refers to the safety and reliability of the path.

[0151] Assume that a charging cabin has three path plans A, B and C, and collect the specific data of these path plans in four dimensions: charging efficiency, cost-effectiveness, equipment urgency and path safety. For example, the data of path plan A is: charging efficiency 0.85, cost-effectiveness 0.9, equipment urgency 0.7, path safety 0.8. The efficiency value of each path plan in each dimension is calculated by the DEA model. Assume that the efficiency values ​​of path plan A are: charging efficiency 0.85, cost-effectiveness 0.9, equipment urgency 0.7, path safety 0.8;

[0152] Step 402: multiplying the efficiency value of each dimension by the weight value of each dimension obtained by the hierarchical analysis method to obtain the weighted efficiency value of each path solution in each dimension;

[0153] In this step, the analytic hierarchy process refers to a multi-criteria decision-making method, which establishes a hierarchical model, compares the importance of each factor layer by layer, and finally obtains the weight of each factor;

[0154] The weighted efficiency value means that the weighted efficiency value of each path scheme in each dimension is obtained by multiplying the efficiency value of each dimension with the corresponding weight value;

[0155] Assume that the weight values ​​obtained by the hierarchical analysis method are: charging efficiency α = 0.4α = 0.4, cost-effectiveness β = 0.3β = 0.3, equipment urgency γ = 0.2γ = 0.2, and path safety δ = 0.1δ = 0.1. Multiply the efficiency value of path plan A by the weight value to obtain the weighted efficiency value:

[0156] Charging efficiency: 0.85×0.4=0.340.85×0.4=0.34, cost-effectiveness: 0.9×0.3=0.270.9×0.3=0.27, equipment urgency: 0.7×0.2=0.140.7×0.2=0.14, path safety: 0.8×0.1=0.080.8×0.1=0.08;

[0157] Step 403: normalizing the weighted efficiency value to eliminate the dimension effect and obtain a normalized weighted efficiency value;

[0158] In this step, normalization is used to standardize the data so that they are within the same dimensional range to facilitate subsequent analysis and comparison;

[0159] Assume that the weighted efficiency values ​​of path plan A are: charging efficiency 0.34, cost-effectiveness 0.27, equipment urgency 0.14, and path safety 0.08. To normalize these weighted efficiency values, the minimum-maximum normalization method can be used: Normalized value = original value - minimum value Maximum value - minimum value Normalized value = maximum value - minimum value Original value - minimum value Assume that the weighted efficiency values ​​of all path plans range from 0 to 1, and the normalized weighted efficiency values ​​remain unchanged: Charging efficiency: 0.34, cost-effectiveness: 0.27, equipment urgency: 0.14, path safety: 0.08;

[0160] Step 404: using principal component analysis to perform dimensionality reduction processing on the normalized weighted efficiency value, extracting the main components, and obtaining main component data;

[0161] In this step, principal component analysis refers to a statistical method that transforms the original data into a new set of uncorrelated variables, namely principal components, through linear transformation, thereby reducing the dimensionality of the data while retaining as much information as possible;

[0162] Assume that the normalized weighted efficiency values ​​of path schemes A, B, and C are: Scheme A: [0.34, 0.27, 0.14, 0.08], Scheme B: [0.30, 0.25, 0.16, 0.09], Scheme C: [0.32, 0.28, 0.15, 0.07];

[0163] PCA is used to reduce the dimension of these data and extract the main components. Assuming that two main components are extracted, the main component data obtained are: Scheme A: [0.5, 0.3], Scheme B: [0.4, 0.4], Scheme C: [0.45, 0.35];

[0164] Step 405: Use a multi-layer perceptron neural network to perform nonlinear mapping on the main component data, comprehensively calculate the interactions between the path solutions, and generate a preliminary comprehensive efficiency score for each path solution;

[0165] In this step, the multilayer perceptron neural network refers to a feedforward artificial neural network that can capture the complex relationship between data through nonlinear mapping of multiple layers of neurons;

[0166] Assume that the main component data obtained by PCA are: Scheme A: [0.5, 0.3], Scheme B: [0.4, 0.4], Scheme C: [0.45, 0.35];

[0167] Construct an MLP neural network with two nodes in the input layer, one or more nodes in the hidden layer, and one node in the output layer. Train the neural network to generate a preliminary comprehensive efficiency score for each path plan. For example, after training, the preliminary comprehensive efficiency scores obtained are: Plan A: 0.8, Plan B: 0.75, Plan C: 0.78;

[0168] Step 406: Randomly sample the preliminary comprehensive efficiency scores through Monte Carlo simulation to evaluate the stability of the scores and determine the comprehensive efficiency score of each path solution;

[0169] In this step, Monte Carlo simulation refers to a method of estimating numerical results through random sampling, which is often used to assess uncertainty or risk;

[0170] Assume that the preliminary comprehensive efficiency scores obtained by the MLP neural network are: Scheme A: 0.8, Scheme B: 0.75, Scheme C: 0.78;

[0171] Through Monte Carlo simulation, these preliminary comprehensive efficiency scores are randomly sampled multiple times, and a certain random disturbance is introduced in each sampling to evaluate the stability of the scores. It is assumed that after 1,000 simulations, the average comprehensive efficiency scores obtained are: Plan A: 0.79, Plan B: 0.76, Plan C: 0.77;

[0172] Finally, based on the results of the Monte Carlo simulation, the comprehensive efficiency score of each path plan was determined. For example, the comprehensive efficiency score of plan A was 0.79, the comprehensive efficiency score of plan B was 0.76, and the comprehensive efficiency score of plan C was 0.77.

[0173] The embodiments of the present invention can improve not only the charging efficiency, cost-effectiveness and safety through the above steps, but also the response speed and scheduling flexibility of the system.

[0174] Based on this, the present invention provides a specific embodiment, wherein the step 405 uses a multi-layer perceptron neural network to perform nonlinear mapping on the main component data, comprehensively calculate the interactions between the path solutions, and generate a preliminary comprehensive efficiency score for each path solution, which specifically includes the following steps:

[0175] Step 501: nonlinearly mapping the main component data using a multi-layer perceptron neural network model, wherein the input layer receives the main component data and the output layer is used to output a preliminary comprehensive efficiency score of each path solution;

[0176] In this step, the multi-layer perceptron neural network refers to a feed-forward artificial neural network, which consists of multiple fully connected layers. Through nonlinear activation functions, nonlinear mapping is performed. MLP can capture the complex relationship between input data and generate output;

[0177] Principal component data refers to the main component data extracted by principal component analysis, which represents the main information in the original data;

[0178] Example: Assume that the main component data obtained by PCA are: Scheme A: [0.5, 0.3], Scheme B: [0.4, 0.4], Scheme C: [0.45, 0.35];

[0179] Construct an MLP model with two nodes in the input layer, one or more nodes in the hidden layer, and one node in the output layer. Train the MLP model, input the main component data, and output the preliminary comprehensive efficiency score of each path plan. For example, after training, the preliminary comprehensive efficiency scores obtained are: Plan A: 0.8, Plan B: 0.75, Plan C: 0.78;

[0180] Step 502: Introduce an attention mechanism into the multi-layer perceptron neural network model to perform weighted processing on the main component data, highlight the influence of key components, and obtain weighted main component data; In this step, the attention mechanism refers to a method for enhancing a neural network model, which enables the model to focus on key information and ignore unimportant information by learning the importance weights of different parts of the input data; the weighted main component data refers to data after the main component data is weighted by the attention mechanism, highlighting the influence of key components;

[0181] Assume that the main component data obtained by the MLP model are: Scheme A: [0.5, 0.3], Scheme B: [0.4, 0.4], Scheme C: [0.45, 0.35];

[0182] Introduce the attention mechanism into the MLP model to perform weighted processing on these main component data. Assume that the weights calculated by the attention mechanism are: Scheme A: [0.7, 0.3], Scheme B: [0.6, 0.4], Scheme C: [0.65, 0.35];

[0183] The weighted data are: Plan A: [0.5*0.7, 0.3*0.3] = [0.35, 0.09], Plan B: [0.4*0.6, 0.4*0.4] = [0.24, 0.16], Plan C: [0.45*0.65, 0.35*0.35] = [0.2925, 0.1225];

[0184] Step 503: Modeling the time series characteristics of the weighted principal component data using a long short-term memory network to capture the time-varying interactions between path solutions and generate time series features;

[0185] In this step, the long short-term memory network (LSTM) refers to a special recurrent neural network (RNN) that can effectively capture the long-term dependencies in time series data. LSTM controls the flow of information through a gating mechanism to avoid the gradient vanishing problem;

[0186] Time series features refer to the characteristics of path solutions that change over time captured by the LSTM model, reflecting the dynamic interactions between path solutions;

[0187] Assume that the weighted principal component data are arranged in chronological order as follows: time t1: [0.35, 0.09],

[0188] Time t2: [0.24, 0.16], time t3: [0.2925, 0.1225];

[0189] Construct an LSTM model, input the above time series data, and output the time series features. Assume that the time series features output by the LSTM model are: time t1: [0.3, 0.1], time t2: [0.2, 0.2], time t3: [0.25, 0.15]; Convolutional neural network (CNN) refers to a neural network commonly used in image processing and time series analysis. It extracts local features through the convolution layer and reduces the feature dimension through the pooling layer, and finally generates a high-level feature representation;

[0190] The spatial and temporal feature vectors refer to the local features and spatial correlations in the time series features extracted by the CNN model, forming a feature vector containing spatial and temporal information;

[0191] Assume that the time series features output by the LSTM model are: time t1: [0.3, 0.1], time t2: [0.2, 0.2], time t3: [0.25, 0.15];

[0192] Construct a CNN model, input the above time series features, extract local features through convolutional layers and pooling layers, and assume that the spatial and temporal feature vectors output by the CNN model are:

[0193] Eigenvector 1: [0.28, 0.12, 0.05], Eigenvector 2: [0.22, 0.18, 0.02], Eigenvector 3: [0.24, 0.13, 0.07];

[0194] Step 504: extract local features from the time series features through a convolutional neural network, identify the spatial correlation between path solutions, and obtain spatial and temporal feature vectors;

[0195] In this step, convolutional neural network (CNN) refers to a neural network commonly used in image processing and time series analysis. It extracts local features through convolutional layers, reduces feature dimensions through pooling layers, and finally generates high-level feature representations;

[0196] The spatial and temporal feature vectors refer to the local features and spatial correlations in the time series features extracted by the CNN model, forming a feature vector containing spatial and temporal information;

[0197] Assume that the time series features output by the LSTM model are: time t1: [0.3, 0.1], time t2: [0.2, 0.2], time t3: [0.25, 0.15];

[0198] Construct a CNN model, input the above time series features, extract local features through convolutional layers and pooling layers, and assume that the spatial and temporal feature vectors output by the CNN model are: feature vector 1: [0.28, 0.12, 0.05], feature vector 2: [0.22, 0.18, 0.02], feature vector 3: [0.24, 0.13, 0.07];

[0199] Step 505: input the spatial and temporal feature vectors into the fully connected layer, and after nonlinear transformation, comprehensively calculate the interactions between the path solutions to generate a preliminary comprehensive efficiency score for each path solution;

[0200] In this step, the fully connected layer refers to a layer in the neural network, in which each node is connected to all nodes in the previous layer, mapping the input data to the output space through nonlinear transformation;

[0201] The preliminary comprehensive efficiency score refers to the comprehensive efficiency score of each path solution calculated by the fully connected layer, which reflects the overall performance of the path solution;

[0202] Assume that the spatial and temporal feature vectors output by the CNN model are: Feature vector 1: [0.28, 0.12, 0.05], Feature vector 2: [0.22, 0.18, 0.02], Feature vector 3: [0.24, 0.13, 0.07];

[0203] Construct a fully connected layer, input the above feature vector, and generate the preliminary comprehensive efficiency score of each path plan after nonlinear transformation. Assume that the preliminary comprehensive efficiency scores output by the fully connected layer are: Plan A: 0.82, Plan B: 0.77, Plan C: 0.79;

[0204] The embodiment of the present invention improves the model's attention to important information through the above steps, better reflects the dynamic interaction between path solutions, and provides a more comprehensive feature representation.

[0205] Based on this, the present invention provides a specific embodiment, wherein the step 103 uses a machine learning algorithm to predict the power demand trend of the optimal charging station scheduling plan and charging path in combination with historical data and current environmental factors to form a charging preparation plan, which specifically includes the following steps:

[0206] Step 601: Analyze the optimal charging station scheduling plan and historical data of charging paths using a time series analysis algorithm to identify periodic and seasonal patterns of power demand and obtain historical demand patterns;

[0207] In this step, the time series analysis algorithm refers to a statistical method used to analyze data that changes over time and identify trends, periodicity, and seasonal patterns in the data; historical data includes past charging records, such as charging time, charging amount, device type, etc.; historical demand patterns refer to the periodicity and seasonality patterns of electricity demand identified through time series analysis;

[0208] Assume that a charging cubicle has charging record data for the past year. Use time series analysis algorithms to analyze this data and identify the periodic and seasonal patterns of electricity demand; for example, it is found that 4 to 6 p.m. every day is the peak charging period, while the charging demand on weekends is lower than on weekdays. In addition, the charging demand in summer is higher than in winter.

[0209] Step 602: using an integrated learning algorithm, combining the historical demand pattern and current environmental factors, to predict the power demand and generate a preliminary power demand prediction result;

[0210] In this step, ensemble learning algorithm refers to a machine learning method that improves prediction accuracy by combining the prediction results of multiple models. Common ensemble learning algorithms include random forest, gradient boosting tree, etc.; current environmental factors include environmental variables that affect power demand, such as weather conditions, holidays, special events, etc.; preliminary power demand forecast results refer to power demand forecast results generated by ensemble learning algorithms;

[0211] Assuming that the historical demand pattern has been identified, combined with the current environmental factors, the random forest algorithm is used to predict the power demand. The preliminary power demand forecast results generated are: in the next week, the charging demand from 4 to 6 pm every day will increase by 15%, and the charging demand during the National Day holiday will increase by 20% compared to usual;

[0212] Step 603: using a deep learning algorithm to refine the preliminary power demand forecast result, introducing equipment type and urgency level as additional inputs, optimizing the forecast accuracy, and obtaining an optimized power demand forecast result;

[0213] In this step, deep learning algorithm refers to a machine learning method based on neural networks, which can process complex data relationships and improve prediction accuracy; device type refers to different types of devices having different charging requirements; urgency level refers to the charging priority of the device; optimized power demand forecast result refers to the power demand forecast result after refined processing by deep learning algorithm;

[0214] Assuming that preliminary power demand forecast results have been generated, these forecast results are further refined using deep learning algorithms, and equipment type and urgency level are introduced as additional inputs. After training and optimization, optimized power demand forecast results are generated. For example, the forecast results show that in the next week, the charging demand for medical equipment will increase by 20%, while the charging demand for communication equipment will increase by 10%.

[0215] Step 604: Using simulation technology, based on the optimized power demand forecast result, simulate the operation of the charging cabin under different charging strategies, evaluate the effects of various strategies, and form a variety of alternative charging preparation plans;

[0216] In this step, true simulation technology refers to a technology that simulates the operation of the actual system through a computer to evaluate the effects of different strategies; charging strategies refer to different charging scheduling and path selection schemes; alternative charging preparation plans refer to charging preparation plans under different charging strategies generated through simulation;

[0217] Assuming that the optimized power demand forecast results have been obtained, simulation technology is used to simulate the operation of the charging cabin under different charging strategies, for example, simulating the fast charging strategy, balanced charging strategy and energy-saving charging strategy, and evaluating the effect of each strategy through simulation to generate multiple alternative charging preparation plans. For example, the fast charging strategy can meet the needs of most devices in a short time, but may result in low resource utilization; the balanced charging strategy can balance resource utilization and charging efficiency; the energy-saving charging strategy can reduce energy consumption, but may extend the charging time;

[0218] Step 605: Apply a multi-objective optimization algorithm to comprehensively evaluate the multiple alternative charging preparation plans, calculate charging efficiency, resource utilization, equipment urgency and user satisfaction, and select the best charging preparation plan;

[0219] In this step, the multi-objective optimization algorithm refers to an optimization method used to simultaneously optimize multiple objective functions, such as charging efficiency, resource utilization, equipment urgency, and user satisfaction; the optimal charging preparation plan refers to the optimal charging strategy selected after comprehensive evaluation by the multi-objective optimization algorithm;

[0220] Assuming that multiple alternative charging preparation plans have been generated, a multi-objective optimization algorithm is used to comprehensively evaluate these plans. The evaluation indicators include charging efficiency, resource utilization, equipment urgency, and user satisfaction. The best charging preparation plan is selected through the optimization algorithm. For example, the charging preparation plan finally selected can balance resource utilization and equipment urgency while ensuring high charging efficiency, and at the same time improve user satisfaction;

[0221] The embodiment of the present invention improves the accuracy of power demand forecasting through the above steps, further optimizes the forecasting precision, makes the forecasting results more in line with the actual situation, provides a variety of alternative charging preparation plans, increases the flexibility of decision-making, comprehensively calculates the charging efficiency, resource utilization, equipment urgency and user satisfaction, and ensures efficient operation and user satisfaction.

[0222] Based on this, the present invention provides a specific embodiment, in which step 104, the power consumption of the equipment, the connection status with the charging cabin and the surrounding environmental parameters during the implementation of the charging preparation plan are monitored in real time through the Internet of Things technology, and when an abnormal situation is detected, an emergency processing instruction is generated, which specifically includes the following steps:

[0223] Step 701: using the Internet of Things technology to collect real-time data on the power consumption of the equipment, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan, and obtain real-time monitoring data;

[0224] In this step, the Internet of Things (IoT) refers to a technology that connects various physical devices through a network, enabling the devices to collect and exchange data; real-time monitoring data includes data such as the power consumption of the device, the connection status with the charging cabin, and the surrounding environmental parameters, which are used for real-time monitoring and analysis;

[0225] Step 702: Analyze the real-time monitoring data using an anomaly detection algorithm to identify abnormal power consumption, abnormal connection status, and abnormal environmental parameters, and generate an anomaly detection result;

[0226] In this step, the anomaly detection algorithm refers to a machine learning method used to identify abnormal patterns in data. Common anomaly detection algorithms include isolation forest, local anomaly factor (LOF), etc.; the anomaly detection results include the results of power consumption anomaly, connection status anomaly, and environmental parameter anomaly generated by the anomaly detection algorithm;

[0227] Assume that real-time monitoring data has been collected and the Isolation Forest algorithm is used to detect anomalies in the data. For example, the following anomalies are detected:

[0228] Device A: Power consumption suddenly increased to 4.0A, which may be due to device failure or overload; Device B: The connection status is unstable and intermittent disconnection occurs; Environmental parameters: The temperature suddenly rose to 30°C, which may affect the normal operation of the device;

[0229] The generated abnormal detection results are: Device A: abnormal power consumption; Device B: abnormal connection status; Environmental parameters: abnormal temperature;

[0230] Step 703: Applying context awareness technology in combination with the abnormality detection result and the urgency level of the device to evaluate the severity and urgency of the abnormality and form a context awareness evaluation report;

[0231] In this step, situational awareness technology refers to a technology that understands and responds to a specific situation by analyzing the current environment and contextual information;

[0232] Situational awareness assessment report refers to a report that assesses the severity and urgency of an abnormal situation by combining the abnormal detection results and the urgency level of the equipment;

[0233] Assuming that anomaly detection results have been generated, context-aware technology is used to evaluate the urgency level of the device (such as high, medium, and low);

[0234] For example, device A: abnormal power consumption, the urgency level is high; device B: abnormal connection status, the urgency level is medium; environmental parameters: abnormal temperature, the urgency level is medium;

[0235] The situational awareness assessment report is as follows:

[0236] Device A: abnormal power consumption, high urgency level, needs to be dealt with immediately; Device B: abnormal connection status, medium urgency level, needs to be dealt with as soon as possible; Environmental parameters: abnormal temperature, medium urgency level, needs to be dealt with as soon as possible;

[0237] Step 704: using the expert system, according to the situational awareness assessment report, calling a predefined emergency handling rule library, matching the emergency handling strategy, and generating preliminary emergency handling instructions;

[0238] In this step, the expert system refers to an artificial intelligence system based on knowledge and reasoning rules, which is used to solve complex problems; the emergency handling rule base refers to a series of predefined emergency handling rules and strategies; the preliminary emergency handling instructions refer to the preliminary emergency handling instructions generated according to the situational awareness assessment report and the emergency handling rule base;

[0239] Assuming that a situational awareness assessment report has been generated, the expert system is used to call the pre-defined emergency processing rule base. For example, Rule 1: If the power consumption of the device is abnormal and the urgency level is high, immediately cut off the power supply and notify the maintenance personnel; Rule 2: If the device connection status is abnormal and the urgency level is medium, reconnect the device and monitor its status; Rule 3: If the ambient temperature is abnormal and the urgency level is medium, start the air conditioner to cool down and monitor the temperature change;

[0240] The initial emergency handling instructions generated are: Device A: immediately cut off the power supply and notify the maintenance personnel; Device B: reconnect the device and monitor its status; Environmental parameters: start the air conditioner to cool down and monitor the temperature changes;

[0241] Step 705: Optimize the preliminary emergency processing instructions through a multi-agent collaborative decision-making algorithm, calculate the charging requirements and resource allocation of other devices in the charging cabin, and generate a final emergency processing instruction;

[0242] In this step, the multi-agent collaborative decision-making algorithm refers to a method to optimize decision-making through collaboration and interaction between multiple agents;

[0243] The final emergency handling instructions refer to the emergency handling instructions optimized by the multi-agent collaborative decision-making algorithm;

[0244] Assuming that preliminary emergency handling instructions have been generated, these instructions are optimized through the multi-agent collaborative decision-making algorithm, for example: Device A: immediately cut off the power supply and notify the maintenance personnel; Device B: reconnect the device and monitor its status; Environmental parameters: start the air conditioner to cool down and monitor the temperature change;

[0245] Through the multi-agent collaborative decision-making algorithm, the charging needs and resource allocation of other devices in the charging cabin are taken into consideration to generate the final emergency processing instructions: Device A: immediately cut off the power supply and notify the maintenance personnel; at the same time, adjust the charging priority of other devices to ensure that the charging of key devices is not affected; Device B: reconnect the device and monitor its status; at the same time, adjust the charging path to avoid affecting the normal charging of other devices; for environmental parameters, start the air conditioner to cool down and monitor temperature changes; at the same time, adjust the ventilation system of the charging station to ensure the stability of the overall environment;

[0246] The embodiment of the present invention improves fault detection capability and safety through the above steps; applies situational awareness technology, combined with the urgency level of the equipment, to evaluate the severity and urgency of abnormal situations, making emergency handling more accurate and efficient; utilizes expert systems and predefined emergency handling rule bases to automatically generate preliminary emergency handling instructions, thereby improving the automation level of emergency handling; optimizes emergency handling instructions through a multi-agent collaborative decision-making algorithm, taking into account the charging needs and resource allocation of other equipment in the charging cabin, thereby improving overall efficiency and resource utilization.

[0247] Figure 2 A schematic diagram of the control system of a charging cabin is provided in the embodiment of the present application. Figure 2 As shown, the system includes:

[0248] Receiving module 01, receiving a power demand signal from a device, wherein the power demand signal includes: the type of device, the current power level, the required charging time, and the urgency level of the device;

[0249] Analysis module 02, using a multi-objective optimization algorithm to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain an optimal charging station scheduling plan and charging path;

[0250] Prediction module 03, using machine learning algorithms to predict the power demand trend of the optimal charging station scheduling plan and charging path in combination with historical data and current environmental factors to form a charging preparation plan;

[0251] Monitoring module 04, which uses the Internet of Things technology to monitor the power consumption of the equipment, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan in real time, and generates emergency processing instructions when an abnormal situation is detected;

[0252] Evaluation module 05 uses big data analysis technology to comprehensively evaluate the emergency handling instructions and the situation after charging is completed to obtain a charging effect evaluation report.

[0253] Figure 2The control system of the charging cabin can execute Figure 1 The implementation principle and technical effect of the xx method described in the embodiment shown will not be repeated. The specific way in which each module and unit performs operations in the control system of a charging cabin in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0254] Figure 2 A control system of a charging cabin in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0255] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0256] The processing component 32 is used for: right 1.

[0257] The processing component 32 includes one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be implemented by one or more application-specific integrated circuits (AICs), digital signal processors (DPs), digital signal processing devices (DPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0258] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (RAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0259] Computing devices also include other components, such as input / output interfaces, display components, and communication components.

[0260] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above-mentioned peripheral interface module can be an output device or an input device.

[0261] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0262] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0263] The embodiment of the present invention further provides a computer storage medium storing a computer program, which can achieve the above-mentioned Figure 1 A control method and system for a charging cabin in the illustrated embodiment.

[0264] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0265] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0266] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0267] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A control method for a charging cabin, characterized in that: include: Receiving a power demand signal from a device, the power demand signal including: the type of device, the current power level, the required charging time, and the urgency level of the device; A multi-objective optimization algorithm is used to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain an optimal charging station scheduling plan and charging path; Using a machine learning algorithm, the optimal charging station scheduling scheme and charging path are combined with historical data and current environmental factors to predict the power demand trend and form a charging preparation plan; The power consumption of the equipment, the connection status with the charging cabin and the surrounding environment parameters during the implementation of the charging preparation plan are monitored in real time through the Internet of Things technology, and when an abnormal situation is detected, an emergency processing instruction is generated; Big data analysis technology is used to conduct a comprehensive evaluation of the emergency handling instructions and the situation after charging is completed to obtain a charging effect evaluation report.

2. The method according to claim 1, characterized in that A multi-objective optimization algorithm is used to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain the optimal charging station scheduling plan and charging path, including: A genetic algorithm and a particle swarm optimization algorithm are used to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain a set of charging station candidates that meet the preliminary conditions; The charging station candidate set is classified by using a fuzzy C-means clustering algorithm to obtain multiple charging station clusters, and a virtual agent model is established within the charging station cluster using a reinforcement learning algorithm to simulate the charging efficiency and resource utilization under different charging strategies, and generate an actual performance evaluation of each charging station; By using Bayesian network analysis, combined with historical charging data and current environmental variables, the working state of the actual performance evaluation is predicted to obtain the working state change trend of each charging station cluster; By adopting the fusion technology of ant colony algorithm and Dijkstra algorithm, based on the working state change trend, the optimal path from the charging cabin entrance to the target charging station is found, and the optimal path solution is determined; The optimal path plan is comprehensively evaluated through a multi-criteria decision analysis method to calculate the charging efficiency, cost-effectiveness, equipment urgency and path safety, and select the optimal charging station scheduling plan and charging path.

3. The method according to claim 2, characterized in that The optimal path scheme is comprehensively evaluated through a multi-criteria decision analysis method, the charging efficiency, cost-effectiveness, equipment urgency and path safety are calculated, and the optimal charging station scheduling scheme and charging path are selected, including: The importance of the four dimensions of charging efficiency, cost-effectiveness, equipment urgency, and route safety was quantified using the analytic hierarchy process, and an evaluation index system was constructed to obtain the weight value of each dimension. The data envelopment analysis method is used to evaluate the efficiency of the performance of the optimal path solution under the evaluation index system, and the comprehensive efficiency score of each path solution is calculated by combining the weight value of each dimension; Applying grey correlation analysis and combining the comprehensive efficiency score, the correlation between each path plan and the ideal plan is calculated to obtain the correlation ranking of the path plans; Using fuzzy comprehensive evaluation method to refine the correlation ranking, calculate the influence of uncertainty and subjective judgment, and generate the final path evaluation result; The final path evaluation result and historical charging data are analyzed through an integrated learning algorithm to predict charging demand, dynamically adjust the charging station scheduling plan, form an adjusted charging station scheduling plan, and select the optimal charging station scheduling plan and charging path based on the adjusted charging station scheduling plan.

4. The method according to claim 3, characterized in that The data envelopment analysis method is used to evaluate the efficiency of the optimal path solution under the evaluation index system, and the comprehensive efficiency score of each path solution is calculated by combining the weight value of each dimension, including: The data envelopment analysis method is used to evaluate the efficiency of the optimal path solution in four dimensions: charging efficiency, cost-effectiveness, equipment urgency, and path safety, and the efficiency value of each path solution in each dimension is obtained; Multiplying the efficiency value of each dimension by the weight value of each dimension obtained by the hierarchical analysis method to obtain the weighted efficiency value of each path plan in each dimension; Normalizing the weighted efficiency value to eliminate the dimension effect and obtain a normalized weighted efficiency value; Using principal component analysis to perform dimensionality reduction processing on the normalized weighted efficiency value, extracting the main components, and obtaining main component data; A multi-layer perceptron neural network is used to perform nonlinear mapping on the main component data, comprehensively calculate the interactions between the path plans, and generate a preliminary comprehensive efficiency score for each path plan; The preliminary comprehensive efficiency scores are randomly sampled through Monte Carlo simulation to evaluate the stability of the scores and determine the comprehensive efficiency score of each path solution.

5. The method according to claim 4, characterized in that A multi-layer perceptron neural network is used to perform nonlinear mapping on the main component data, comprehensively calculate the interactions between the path solutions, and generate a preliminary comprehensive efficiency score for each path solution, including: A multi-layer perceptron neural network model is used to perform nonlinear mapping on the main component data, wherein the input layer receives the main component data, and the output layer is used to output a preliminary comprehensive efficiency score of each path plan; Introducing an attention mechanism into the multi-layer perceptron neural network model to perform weighted processing on the main component data, highlighting the influence of key components, and obtaining weighted main component data; Using a long short-term memory network to model the time series characteristics of the weighted principal component data, capturing the time-varying interactions between path plans and generating time series features; Extract local features of the time series features through a convolutional neural network, identify the spatial correlation between path plans, and obtain spatial and temporal feature vectors; The spatial and temporal feature vectors are input into the fully connected layer, and after nonlinear transformation, the interactions between the path plans are comprehensively calculated to generate a preliminary comprehensive efficiency score for each path plan.

6. The method according to claim 1, characterized in that The optimal charging station scheduling scheme and charging path are predicted by using a machine learning algorithm, combined with historical data and current environmental factors, to predict the power demand trend and form a charging preparation plan, including: Analyzing the optimal charging station scheduling scheme and historical data of charging paths using a time series analysis algorithm to identify periodic and seasonal patterns of power demand and obtain historical demand patterns; Using an integrated learning algorithm, combining the historical demand pattern and current environmental factors, to predict power demand and generate preliminary power demand forecast results; Using a deep learning algorithm to refine the preliminary power demand forecast result, introducing equipment type and urgency level as additional inputs, optimizing the forecast accuracy, and obtaining an optimized power demand forecast result; Through simulation technology, based on the optimized power demand forecast results, the operation of the charging cabin under different charging strategies is simulated, the effects of various strategies are evaluated, and a variety of alternative charging preparation plans are formed; A multi-objective optimization algorithm is applied to comprehensively evaluate the various alternative charging preparation plans, calculate charging efficiency, resource utilization, equipment urgency and user satisfaction, and select the best charging preparation plan.

7. The method according to claim 1, characterized in that The power consumption of the equipment, the connection status with the charging cabin and the surrounding environmental parameters during the implementation of the charging preparation plan are monitored in real time through the Internet of Things technology. When an abnormal situation is detected, an emergency processing instruction is generated, including: The Internet of Things technology is used to collect real-time data on the power consumption of the equipment, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan to obtain real-time monitoring data; An anomaly detection algorithm is used to analyze the real-time monitoring data, identify abnormal power consumption, abnormal connection status, and abnormal environmental parameters, and generate an anomaly detection result; Applying situational awareness technology in combination with the abnormal detection results and the urgency level of the device to evaluate the severity and urgency of the abnormal situation and form a situational awareness assessment report; Using the expert system, according to the situational awareness assessment report, calling a predefined emergency handling rule base, matching the emergency handling strategy, and generating preliminary emergency handling instructions; The preliminary emergency handling instructions are optimized through a multi-agent collaborative decision-making algorithm, the charging requirements and resource allocation of other equipment in the charging cabin are calculated, and the final emergency handling instructions are generated.

8. A control system for a charging cabin, characterized in that: include: A receiving module receives a power demand signal from a device, wherein the power demand signal includes: the type of device, the current power level, the required charging time, and the urgency level of the device; The analysis module uses a multi-objective optimization algorithm to comprehensively analyze the power demand signal, the working status of each charging station in the charging cabin, the remaining power resources, and the urgency level of the equipment to obtain the optimal charging station scheduling plan and charging path; A prediction module, using a machine learning algorithm, predicts the power demand trend based on the optimal charging station scheduling plan and charging path, combined with historical data and current environmental factors, to form a charging preparation plan; A monitoring module, which uses the Internet of Things technology to monitor the power consumption of the equipment, the connection status with the charging cabin, and the surrounding environmental parameters during the implementation of the charging preparation plan in real time, and generates emergency processing instructions when an abnormal situation is detected; The evaluation module uses big data analysis technology to comprehensively evaluate the emergency handling instructions and the situation after charging is completed to obtain a charging effect evaluation report.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the method according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, the method according to any one of claims 1 to 7 is implemented.

Images