A hybrid energy storage based shelter optimal discharge control method and system

Abstract

The application provides a hybrid energy storage based shelter optimal discharge control method and system, which utilizes sensor networks to monitor the state parameters of energy storage units in real time, generates discharge strategies through deep reinforcement learning algorithms. Genetic algorithms perform multi-objective optimization on the strategies, and fuzzy logic decision support systems evaluate and select the optimal solution. Intelligent scheduling algorithms output discharge instructions based on the optimal solution, dynamically adjust charging and discharging behavior, and ensure the best operating state of the shelter. Finally, machine learning algorithms are used to learn the deviation between actual results and preset targets, automatically adjust parameters, optimize discharge strategies, and form a self-adaptive closed-loop control mechanism. The application significantly improves the operating efficiency, economy and reliability of the shelter power supply system, ensuring stable power output in various complex application scenarios.

Description

Technical Field

[0001] This invention relates to the field of hybrid energy storage modular container technology, and in particular to an optimal discharge control method and system for modular containers based on hybrid energy storage. Background Technology

[0002] With the widespread application of renewable energy and the development of the energy internet, the demand for mobile energy storage systems (such as modular storage units) is increasing. These systems require high flexibility, reliability, and intelligent management capabilities to cope with various complex application scenarios, such as power supply in remote areas, emergency power in emergencies, and power supply for temporary large-scale events. In these applications, the energy storage system within the modular storage unit must not only ensure stable power output but also consider economic costs, environmental impact, and system health to ensure long-term efficient operation. However, most existing modular energy storage systems use a single type of energy storage unit, such as lithium-ion batteries or lead-acid batteries, and control the charging and discharging process through simple rules or preset programs, which limits the system's flexibility and performance.

[0003] Currently, some advanced systems are beginning to combine multiple energy storage technologies and utilize simple algorithms for energy management and optimization. Furthermore, some research and practice have introduced preliminary intelligent control technologies, such as rule-based decision systems or traditional optimization algorithms, in an attempt to improve the overall system performance. However, these systems lack comprehensive state awareness, monitoring only a few key parameters and failing to fully grasp the real-time status of energy storage units, resulting in insufficient information when formulating discharge strategies. Secondly, the optimization level is limited; most existing optimization algorithms are based on static models and cannot adapt well to changes in external conditions (such as electricity market fluctuations and weather changes), leading to low system operating efficiency and high economic costs. Finally, they lack adaptive capabilities; existing control systems typically lack the ability to learn and adjust themselves. Once the operating environment changes or unforeseen circumstances arise, the system may fail to react in a timely manner, affecting the stability and reliability of power supply. Summary of the Invention

[0004] This invention provides a method and system for optimal discharge control of a container based on hybrid energy storage, which solves the problem in the prior art that the real-time status of the energy storage unit cannot be fully grasped, thus affecting the stability and reliability of power supply.

[0005] In a first aspect, embodiments of the present invention provide an optimal discharge control method for a container based on hybrid energy storage, comprising:

[0006] The sensor network is used to monitor and record the status parameters of different types of energy storage units in the cabin in real time, so as to obtain detailed status information of each energy storage unit. The detailed status information includes: state of charge, temperature, voltage, current and internal resistance. The different types of energy storage units include: lithium-ion batteries, lead-acid batteries and supercapacitors.

[0007] The deep reinforcement learning algorithm in the deep reinforcement model is used to comprehensively analyze the detailed state information, external electricity market price fluctuations, user load forecasts and weather forecasts to generate a discharge strategy.

[0008] The discharge strategy is optimized using a genetic algorithm to comprehensively calculate economic cost, health status, response speed, and environmental impact, resulting in multiple candidate discharge strategies.

[0009] The fuzzy logic decision support system is used to evaluate each of the multiple candidate discharge strategies. By setting different weight coefficients to reflect the priority under different scenarios, the optimal discharge scheme is selected.

[0010] Using an intelligent scheduling algorithm, discharge commands are output to the different types of energy storage units according to the optimal discharge scheme, and the charging and discharging behavior of each energy storage unit is dynamically adjusted to obtain the optimal operating state of the container power supply system.

[0011] Based on the optimal operating state of the cabin power supply system, a machine learning algorithm is used to learn the deviation between the actual effect of the system operation and the preset target during the discharge process, automatically adjust the model parameters, optimize the discharge strategy, and form an adaptive closed-loop control mechanism.

[0012] Optionally, a fuzzy logic decision support system is used to evaluate each of the multiple candidate discharge strategies, and different weighting coefficients are set to reflect the priority under different scenarios to select the optimal discharge scheme, including:

[0013] By using scenario demand analysis to analyze the specific needs of the current application scenario, multiple evaluation indicators are obtained, including: economic cost, health status, response speed, environmental impact, and user satisfaction.

[0014] The multiple evaluation indicators are defined based on the specific requirements of the current application scenario, resulting in a set of evaluation indicators;

[0015] By using preset standards, real-time environmental changes, and user preferences, an adjustable weight coefficient is set for each evaluation indicator in the evaluation indicator set, resulting in a dynamically adjusted set of weight coefficients.

[0016] By using a multi-level fuzzy logic model combined with the weight coefficient set, the performance of each candidate discharge strategy on each evaluation index is quantified to obtain the fuzzy membership function value set of each candidate discharge strategy.

[0017] The comprehensive performance of each candidate discharge strategy is calculated by applying the fuzzy comprehensive evaluation method in combination with the fuzzy membership function value set and the weight coefficient set, and the comprehensive evaluation value of each candidate discharge strategy is obtained.

[0018] A threshold judgment mechanism is used to filter the comprehensive evaluation value to obtain a candidate discharge strategy set. By comparing the comprehensive evaluation value of each strategy in the candidate discharge strategy set, the discharge strategy with the best comprehensive performance is selected to obtain the optimal discharge scheme.

[0019] Optionally, the fuzzy comprehensive evaluation method is applied in combination with the fuzzy membership function value set and the weight coefficient set to calculate the comprehensive performance of each candidate discharge strategy, thereby obtaining the comprehensive evaluation value of each candidate discharge strategy, including:

[0020] The weighted membership value of each evaluation index is obtained by multiplying each value in the set of fuzzy membership function values ​​with the weight coefficients in the set of corresponding weight coefficients using the fuzzy comprehensive evaluation method.

[0021] The weighted membership values ​​of all evaluation indicators for each candidate discharge strategy are summed to obtain a preliminary comprehensive evaluation value;

[0022] The preliminary comprehensive evaluation value is normalized using a normalization method to normalize it to the [0,1] interval, thus obtaining the normalized comprehensive evaluation value;

[0023] A multi-objective optimization algorithm is introduced to optimize the normalized comprehensive evaluation value in multiple dimensions, calculating the balance between economic cost, health status, response speed, environmental impact and user satisfaction, and obtaining the target comprehensive evaluation value.

[0024] Optionally, a threshold judgment mechanism is used to filter the comprehensive evaluation values ​​to obtain a candidate discharge strategy set. By comparing the comprehensive evaluation values ​​of each strategy in the candidate discharge strategy set, the discharge strategy with the best comprehensive performance is selected to obtain the optimal discharge scheme, including:

[0025] The comprehensive evaluation value is filtered using a threshold judgment mechanism. Multiple preset thresholds are set, and the comprehensive evaluation value is classified according to the preset thresholds to obtain multiple candidate discharge strategy subsets.

[0026] The multiple candidate discharge strategy subsets are screened layer by layer using a multi-level screening algorithm. Based on the comprehensive evaluation value of the candidate discharge strategies in each subset, candidate discharge strategies below a preset threshold are eliminated, and candidate discharge strategies above a preset threshold are retained to obtain an initial set of candidate discharge strategies.

[0027] Based on the set of weight coefficients, each candidate discharge strategy in the initial screening of candidate discharge strategies is re-evaluated. Combining real-time environmental changes and user preferences, the comprehensive evaluation value of each candidate discharge strategy is dynamically adjusted to obtain a dynamic comprehensive evaluation value set.

[0028] By using the fuzzy comprehensive evaluation method in conjunction with the dynamic comprehensive evaluation value set, the comprehensive performance of each candidate discharge strategy is recalculated to obtain the target comprehensive evaluation value set;

[0029] The target comprehensive evaluation value set is sorted using a sorting algorithm. Based on the sorting results, the discharge strategy with the best comprehensive performance is selected as the optimal discharge scheme.

[0030] Optionally, a multi-objective optimization algorithm is introduced to optimize the normalized comprehensive evaluation value from multiple dimensions, calculating the balance between economic cost, health status, response speed, environmental impact, and user satisfaction, to obtain the target comprehensive evaluation value, including:

[0031] The normalized comprehensive evaluation value is optimized in multiple dimensions using a multi-objective optimization algorithm. The balance between economic cost, health status, response speed, environmental impact and user satisfaction is comprehensively calculated to obtain the multi-dimensional optimization result.

[0032] Based on the multidimensional optimization results, a multi-objective optimization objective function is defined, and a dynamic adjustment factor is introduced to dynamically adjust the weights of each indicator according to real-time environmental changes and user needs, thereby obtaining the optimization objective function. The objective function includes indicators such as economic cost, health status, response speed, environmental impact, and user satisfaction.

[0033] A multi-objective optimization algorithm is run, and the objective function is optimized by combining historical data and real-time data to obtain an initial comprehensive evaluation value;

[0034] Scenario simulation technology is introduced to conduct simulation tests on the discharge strategy corresponding to the initial comprehensive evaluation value under various scenarios, evaluate its performance under different environments and user needs, and obtain simulation test results. The various scenarios include: normal operation scenario, extreme weather scenario and sudden failure scenario.

[0035] The simulation test results are analyzed using machine learning algorithms to identify key influencing factors and potential risk points, and a risk assessment report is generated.

[0036] Based on the risk assessment report and simulation test results, the initial comprehensive evaluation value is corrected to obtain the target comprehensive evaluation value.

[0037] Optionally, an intelligent scheduling algorithm is used to output discharge commands to the different types of energy storage units according to the optimal discharge scheme, dynamically adjusting the charging and discharging behavior of each energy storage unit to obtain the optimal operating state of the modular power supply system, including:

[0038] Based on the optimal discharge scheme, an intelligent scheduling algorithm is used to generate discharge instructions, which include the charging and discharging power, time, and sequence of each energy storage unit.

[0039] The discharge command is sent to the controller of the different types of energy storage units using a real-time communication module to obtain the execution result of the discharge command;

[0040] The charging and discharging behavior of each energy storage unit is dynamically adjusted based on the execution results, and the adjusted charging and discharging behavior is obtained by combining the real-time monitored energy storage unit status parameters.

[0041] By monitoring the operating status of the energy storage unit in real time and using a feedback control mechanism, the adjusted charging and discharging behavior is optimized to obtain the optimized system operating status.

[0042] Adaptive control algorithm and multi-objective optimization algorithm are introduced. Based on the deviation between the optimized system operating state and the preset target, the discharge strategy is automatically adjusted. The balance between economic cost, health status, response speed, environmental impact and user satisfaction is calculated. By combining historical data and real-time data, long-term trend prediction of the discharge strategy is made, and the discharge strategy is adjusted in advance to obtain the optimal operating state of the container power supply system.

[0043] Optionally, based on the optimal operating state of the cabin power supply system, a machine learning algorithm is used to learn the deviation between the actual effect of the system operation and the preset target during the discharge process, automatically adjust the model parameters, optimize the discharge strategy, and form an adaptive closed-loop control mechanism, including:

[0044] The optimal operating state of the modular power supply system is monitored in real time using machine learning algorithms in a deep reinforcement learning model. The actual performance data of the system operation is collected, including: economic cost, health status, response speed, environmental impact and user satisfaction.

[0045] The actual performance data is compared with the preset target, the deviation value of each indicator is calculated, and the deviation analysis results are obtained.

[0046] Based on the results of deviation analysis, the parameters in the deep reinforcement learning model and intelligent scheduling algorithm are automatically adjusted to optimize the prediction and control capabilities of the deep reinforcement learning model and obtain the optimized parameters of the deep reinforcement learning model.

[0047] An adaptive control algorithm is introduced to dynamically adjust the discharge strategy based on the optimized deep reinforcement learning model parameters, thereby obtaining an adaptively adjusted discharge strategy.

[0048] By combining historical and real-time data, machine learning algorithms are used to predict the long-term trend of the adaptively adjusted discharge strategy, identify potential problems and risks in advance, and further optimize the discharge strategy to form an adaptive closed-loop control mechanism.

[0049] Secondly, embodiments of this application provide an optimal discharge control system for a container based on hybrid energy storage, comprising:

[0050] The monitoring module uses a sensor network to monitor and record the status parameters of different types of energy storage units in the cabin in real time, and obtains detailed status information of each energy storage unit. The detailed status information includes: state of charge, temperature, voltage, current and internal resistance. The different types of energy storage units include: lithium-ion batteries, lead-acid batteries and supercapacitors.

[0051] The analysis module uses the deep reinforcement learning algorithm in the deep reinforcement model to comprehensively analyze the detailed state information, external electricity market price fluctuations, user load forecasts and weather forecasts, and generate a discharge strategy.

[0052] The optimization module uses a genetic algorithm to perform multi-objective optimization of the discharge strategy, comprehensively calculating economic cost, health status, response speed, and environmental impact to obtain multiple sets of candidate discharge strategies.

[0053] The evaluation module uses a fuzzy logic decision support system to evaluate each of the multiple candidate discharge strategies. By setting different weight coefficients to reflect the priority under different scenarios, the optimal discharge scheme is selected.

[0054] The output module uses an intelligent scheduling algorithm to output discharge commands to the different types of energy storage units according to the optimal discharge scheme, dynamically adjusts the charging and discharging behavior of each energy storage unit, and obtains the optimal operating state of the container power supply system.

[0055] The adjustment module, based on the optimal operating state of the cabin power supply system, utilizes machine learning algorithms to learn the deviation between the actual system operation and the preset target during the discharge process, automatically adjusts model parameters, optimizes the discharge strategy, and forms an adaptive closed-loop control mechanism. Thirdly, embodiments of the present invention provide 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 the optimal discharge control method for a container based on hybrid energy storage as described in any of the first aspects.

[0056] Fourthly, embodiments of the present invention provide a computer storage medium storing computer program instructions, wherein the computer program instructions, when executed by a processor, implement the optimal discharge control method for a container based on hybrid energy storage as described in any one of the first aspects.

[0057] In this embodiment of the invention, a sensor network is used to monitor and record the state parameters of different types of energy storage units within the container cabin in real time, obtaining detailed state information for each energy storage unit. A deep reinforcement learning algorithm in a deep reinforcement model is used to comprehensively analyze the detailed state information, external electricity market price fluctuations, user load forecasts, and weather forecasts to generate a discharge strategy. A genetic algorithm is used to perform multi-objective optimization of the discharge strategy, calculating economic costs, health status, response speed, and environmental impact, resulting in multiple candidate discharge strategies. A fuzzy logic decision support system is used to evaluate each candidate discharge strategy, setting different weight coefficients to reflect priorities under different scenarios, and selecting the optimal discharge scheme. An intelligent scheduling algorithm is used to output discharge commands to different types of energy storage units according to the optimal discharge scheme, dynamically adjusting the charging and discharging behavior of each energy storage unit to obtain the optimal operating state of the container cabin power supply system. Based on the optimal operating state of the container cabin power supply system, a machine learning algorithm is used to learn the deviation between the actual system operation and the preset target during the discharge process, automatically adjusting model parameters, optimizing the discharge strategy, and forming an adaptive closed-loop control mechanism.

[0058] The technical solution provided by this invention achieves comprehensive monitoring and intelligent optimization of energy storage units. This not only significantly improves the operational efficiency and stability of modular energy storage systems, but also generates optimal discharge strategies based on a comprehensive consideration of economic costs, health status, response speed, and environmental impact, minimizing operating costs and maximizing economic benefits. Furthermore, this invention dynamically adjusts the charging and discharging behavior of energy storage units through a fuzzy logic decision support system and intelligent scheduling algorithms. It also utilizes machine learning algorithms to automatically adjust model parameters during discharge, forming an adaptive closed-loop control mechanism. This enhances the system's adaptability and reliability, ensuring stable power output under various complex application scenarios.

[0059] These or other aspects of the invention will become more apparent from the following description of the embodiments. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0061] Figure 1 A flowchart of an optimal discharge control method for a container based on hybrid energy storage provided in an embodiment of the present invention;

[0062] Figure 2 A schematic diagram of a modular container optimal discharge control system based on hybrid energy storage provided in an embodiment of the present invention;

[0063] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of the present invention. Detailed Implementation

[0064] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0065] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

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

[0067] Existing technologies lack comprehensive state awareness, monitoring only a few key parameters and failing to fully grasp the real-time status of energy storage units. This results in insufficient information when formulating discharge strategies. Therefore, this invention provides an optimal discharge control method for modular energy storage units based on hybrid energy storage, such as... Figure 1 The specific steps are as follows:

[0068] Step 101: Use a sensor network to monitor and record the status parameters of different types of energy storage units in the cabin in real time to obtain detailed status information of each energy storage unit. The detailed status information includes: state of charge, temperature, voltage, current and internal resistance. The different types of energy storage units include: lithium-ion batteries, lead-acid batteries and supercapacitors.

[0069] In this step, each sensor node in the sensor network is installed on a different energy storage unit to collect the aforementioned status parameters in real time and transmit the data to the central control system via wireless or wired means. The central control system stores and performs preliminary processing on the received data to ensure its accuracy and integrity.

[0070] Step 102: Utilize the deep reinforcement learning algorithm in the deep reinforcement model to comprehensively analyze the detailed state information, external electricity market price fluctuations, user load forecasts, and weather forecasts to generate a discharge strategy;

[0071] In this step, the deep reinforcement learning algorithm learns the optimal charging and discharging mode for the energy storage unit using historical and real-time data. The algorithm inputs include the energy storage unit's state parameters, external electricity market price fluctuations, user load forecasts, and weather forecasts. Through continuous trial and error and optimization, the algorithm generates the optimal discharging strategy, ensuring that user needs are met while maximizing economic benefits and system performance.

[0072] Step 103: Use a genetic algorithm to perform multi-objective optimization on the discharge strategy, comprehensively calculate economic cost, health status, response speed and environmental impact, and obtain multiple sets of candidate discharge strategies;

[0073] In this step, the genetic algorithm optimizes the discharge strategy by simulating natural selection and genetic mechanisms. The algorithm defines multiple objective functions, including economic cost, energy storage unit health status, response speed, and environmental impact. Through operations such as crossover and mutation, multiple candidate discharge strategies are generated, and the comprehensive performance of each strategy is evaluated to ensure that the globally optimal solution is found.

[0074] Step 104: Use a fuzzy logic decision support system to evaluate each of the multiple candidate discharge strategies, and select the optimal discharge scheme by setting different weight coefficients to reflect the priority under different scenarios.

[0075] In this step, the fuzzy logic decision support system sets the weight coefficients of each objective function according to different application scenarios. For example, in an economically priority scenario, economic cost has a higher weight; in an environmentally friendly scenario, environmental impact has a higher weight. The system evaluates the merits of each candidate discharge strategy through fuzzy logic reasoning and ultimately selects the optimal solution.

[0076] Step 105: Using an intelligent scheduling algorithm, output discharge commands to the different types of energy storage units according to the optimal discharge scheme, dynamically adjust the charging and discharging behavior of each energy storage unit, and obtain the optimal operating state of the container power supply system.

[0077] In this step, the intelligent scheduling algorithm generates specific charging and discharging commands based on the selected optimal discharge scheme and sends them to each energy storage unit through the control system. The algorithm dynamically adjusts the charging and discharging behavior of each energy storage unit based on real-time status parameters and user needs, ensuring that the modular power supply system is always in optimal operating condition.

[0078] Step 106: Based on the optimal operating state of the cabin power supply system, a machine learning algorithm is used to learn the deviation between the actual effect of the system operation and the preset target during the discharge process, automatically adjust the model parameters, optimize the discharge strategy, and form an adaptive closed-loop control mechanism.

[0079] In this step, the machine learning algorithm collects actual data on system operation and learns the deviation between the actual performance and the preset target. Based on these deviations, the algorithm automatically adjusts the parameters of the deep reinforcement learning model and other related models to optimize the discharge strategy. This adaptive closed-loop control mechanism ensures that the system can continuously improve, adapt to various changing operating environments, and maintain optimal performance.

[0080] This invention, through the aforementioned steps of comprehensive monitoring and intelligent optimization of the energy storage unit, significantly improves the operational efficiency and stability of the modular energy storage system. The generated optimal discharge strategy not only considers economic costs but also takes into account the health status of the energy storage unit, response speed, and environmental impact, minimizing operating costs and maximizing economic benefits. Furthermore, the adaptive closed-loop control mechanism enables the system to react rapidly to changes in the operating environment, ensuring the stability and reliability of power supply and extending the system's lifespan.

[0081] Based on this, the present invention provides a specific embodiment in which, based on step 104, a fuzzy logic decision support system is used to evaluate each of the multiple candidate discharge strategies, and different weight coefficients are set to reflect the priority under different scenarios to select the optimal discharge scheme. The specific steps include:

[0082] Step 201: Analyze the specific needs of the current application scenario using scenario demand analysis to obtain multiple evaluation indicators, including: economic cost, health status, response speed, environmental impact, and user satisfaction.

[0083] This step begins with a detailed analysis of the specific requirements of the current application scenario to determine the key indicators and requirements that the system must meet. For example, in an emergency power supply application scenario in a remote area, key considerations might include economic cost, health status, response speed, environmental impact, and user satisfaction. By analyzing these requirements, multiple evaluation indicators are obtained, including economic cost, health status, response speed, environmental impact, and user satisfaction.

[0084] Step 202: Define the multiple evaluation indicators based on the specific requirements of the current application scenario to obtain a set of evaluation indicators;

[0085] In this step, multiple evaluation indicators are defined and processed according to the specific needs of the current application scenario, forming a set of evaluation indicators. For example, economic cost can be defined as the system's operating and maintenance costs; health status can be defined as the lifespan and performance degradation of the energy storage unit; response speed can be defined as the time from receiving an instruction to its actual execution; environmental impact can be defined as the system's carbon emissions and noise level; and user satisfaction can be defined as the user's satisfaction with the system's performance and services.

[0086] Step 203: Using preset standards, real-time environmental changes, and user preferences, set adjustable weight coefficients for each evaluation indicator in the evaluation indicator set to obtain a dynamically adjusted set of weight coefficients.

[0087] In this step, adjustable weighting coefficients are assigned to each evaluation indicator in the evaluation indicator set, utilizing preset standards, real-time environmental changes, and user preferences. For example, if the current application scenario prioritizes economic costs, the weighting coefficient for economic costs can be set to a higher value; if environmental impact is prioritized, the weighting coefficient for environmental impact can be set to a higher value. By dynamically adjusting the weighting coefficients, the evaluation results are ensured to meet the needs of the current application scenario.

[0088] Step 204: Using a multi-level fuzzy logic model combined with the weight coefficient set, quantify the performance of each candidate discharge strategy on each evaluation index to obtain the fuzzy membership function value set of each candidate discharge strategy.

[0089] In this step, adjustable weighting coefficients are assigned to each evaluation indicator in the evaluation indicator set, utilizing preset standards, real-time environmental changes, and user preferences. For example, if the current application scenario prioritizes economic costs, the weighting coefficient for economic costs can be set to a higher value; if environmental impact is prioritized, the weighting coefficient for environmental impact can be set to a higher value. By dynamically adjusting the weighting coefficients, the evaluation results are ensured to meet the needs of the current application scenario.

[0090] Step 205: Apply the fuzzy comprehensive evaluation method, combining the fuzzy membership function value set and the weight coefficient set, to calculate the comprehensive performance of each candidate discharge strategy and obtain the comprehensive evaluation value of each candidate discharge strategy;

[0091] In this step, the fuzzy comprehensive evaluation method is applied, combining the fuzzy membership function value set and the weight coefficient set to calculate the comprehensive performance of each candidate discharge strategy. The specific steps include: multiplying the fuzzy membership function value of each evaluation index by its corresponding weight coefficient, and then summing the results to obtain the comprehensive evaluation value for each candidate discharge strategy. A higher comprehensive evaluation value indicates better overall performance of the candidate discharge strategy.

[0092] Step 206: Use a threshold judgment mechanism to filter the comprehensive evaluation value to obtain a candidate discharge strategy set. By comparing the comprehensive evaluation value of each strategy in the candidate discharge strategy set, select the discharge strategy with the best comprehensive performance to obtain the optimal discharge scheme.

[0093] In this step, a threshold judgment mechanism is used to filter the comprehensive evaluation values ​​and eliminate candidate discharge strategies that do not meet the requirements. Then, by comparing the comprehensive evaluation values ​​of each strategy in the candidate discharge strategy set, the discharge strategy with the best comprehensive performance is selected as the optimal discharge scheme. For example, the candidate discharge strategy with the highest comprehensive evaluation value can be selected as the final optimal discharge scheme;

[0094] This invention, through the aforementioned steps, precisely analyzes the specific requirements of the current application scenario, ensuring a comprehensive evaluation of system performance. By employing a multi-level fuzzy logic model and fuzzy comprehensive evaluation method, combined with dynamically adjusted weighting coefficients, this invention can accurately quantify the performance of each candidate discharge strategy across various evaluation indicators, thereby selecting the discharge strategy with the best overall performance. This method not only improves the system's operational efficiency and stability but also ensures comprehensive optimization in terms of economic cost, health status, response speed, environmental impact, and user satisfaction. By dynamically adjusting the weighting coefficients, the system can flexibly respond to changes in the needs of different application scenarios, ensuring the provision of the optimal discharge solution in various complex environments.

[0095] Based on this, the present invention provides a specific embodiment in which, based on step 205, the fuzzy comprehensive evaluation method is applied in combination with the fuzzy membership function value set and the weight coefficient set to calculate the comprehensive performance of each candidate discharge strategy, thereby obtaining the comprehensive evaluation value of each candidate discharge strategy. The specific steps include the following:

[0096] Step 301: Use the fuzzy comprehensive evaluation method to multiply each value in the fuzzy membership function value set with the weight coefficient in the corresponding weight coefficient set to obtain the weighted membership value of each evaluation index;

[0097] In this step, the fuzzy comprehensive evaluation method is used to multiply each value in the fuzzy membership function set with the corresponding weight coefficient in the weight coefficient set to obtain the weighted membership value of each evaluation indicator. For example, assuming that the fuzzy membership function value of a candidate discharge strategy on economic cost is 0.8 and the weight coefficient is 0.3, then its weighted membership value on economic cost is 0.8 * 0.3 = 0.24. Similarly, the same calculation is performed on other evaluation indicators (health status, response speed, environmental impact, and user satisfaction) to obtain the weighted membership value of each evaluation indicator.

[0098] Step 302: Sum the weighted membership values ​​of all evaluation indicators for each candidate discharge strategy to obtain a preliminary comprehensive evaluation value;

[0099] In this step, the weighted membership values ​​of all evaluation indicators for each candidate discharge strategy are summed to obtain a preliminary comprehensive evaluation value. For example, assuming that the weighted membership values ​​of a candidate discharge strategy for economic cost, health status, response speed, environmental impact, and user satisfaction are 0.24, 0.18, 0.15, 0.20, and 0.23, respectively, then its preliminary comprehensive evaluation value is 0.24 + 0.18 + 0.15 + 0.20 + 0.23 = 1.00. Similarly, the same calculation is performed on other candidate discharge strategies to obtain the preliminary comprehensive evaluation value for each candidate discharge strategy.

[0100] Step 303: Normalize the preliminary comprehensive evaluation value using a normalization method, normalizing it to the [0,1] interval to obtain the normalized comprehensive evaluation value;

[0101] In this step, to eliminate the influence of differences in dimensions and magnitudes, a normalization method is used to normalize the preliminary comprehensive evaluation value, normalizing it to the [0,1] interval to obtain the normalized comprehensive evaluation value. The purpose of normalization is to ensure that the comprehensive evaluation values ​​of different candidate discharge strategies are compared on the same order of magnitude. For example, assuming that the preliminary comprehensive evaluation values ​​of all candidate discharge strategies are 1.00, 0.85, 0.92, and 0.78, respectively, through normalization, these values ​​are transformed to the [0,1] interval to obtain the normalized comprehensive evaluation value. The normalized value can be directly used to compare the comprehensive performance of different candidate discharge strategies;

[0102] Step 304: Introduce a multi-objective optimization algorithm to optimize the normalized comprehensive evaluation value in multiple dimensions, calculate the balance between economic cost, health status, response speed, environmental impact and user satisfaction, and obtain the target comprehensive evaluation value;

[0103] In this step, a multi-objective optimization algorithm is introduced to optimize the normalized comprehensive evaluation value from multiple dimensions. It calculates the balance between economic cost, health status, response speed, environmental impact, and user satisfaction to obtain the target comprehensive evaluation value. The multi-objective optimization algorithm ensures that the final selected discharge strategy achieves the best balance across all evaluation indicators by finding the optimal solution among multiple objectives. For example, the multi-objective optimization algorithm considers how to maintain high health status, fast response speed, minimal environmental impact, and high user satisfaction while keeping economic costs low, thus selecting the discharge strategy with the best overall performance.

[0104] This invention, through calculating weighted membership values ​​and preliminary comprehensive evaluation values, accurately quantifies the performance of each candidate discharge strategy across various evaluation metrics. The normalization method eliminates the influence of differences in dimensions and magnitudes, allowing for comparison of the comprehensive evaluation values ​​of different candidate discharge strategies on the same order of magnitude. The multi-objective optimization algorithm searches for the optimal solution among multiple evaluation metrics, ensuring that the final selected discharge strategy achieves the best balance in terms of economic cost, health status, response speed, environmental impact, and user satisfaction.

[0105] Based on this, the present invention provides a specific embodiment. Step 206 involves using a threshold judgment mechanism to filter the comprehensive evaluation value to obtain a candidate discharge strategy set. By comparing the comprehensive evaluation value of each strategy in the candidate discharge strategy set, the discharge strategy with the best comprehensive performance is selected to obtain the optimal discharge scheme. Specifically, this includes the following steps:

[0106] Step 401: Use a threshold judgment mechanism to filter the comprehensive evaluation value, set multiple preset thresholds, classify the comprehensive evaluation value according to the preset thresholds, and obtain multiple candidate discharge strategy subsets;

[0107] In this step, a threshold judgment mechanism is used to filter the comprehensive evaluation value. Multiple preset thresholds are set, and the comprehensive evaluation value is classified according to these preset thresholds to obtain multiple subsets of candidate discharge strategies. For example, three preset thresholds can be set: 0.7, 0.8, and 0.9. Candidate discharge strategies with a comprehensive evaluation value less than 0.7 are classified into the first category, those between 0.7 and 0.8 into the second category, those between 0.8 and 0.9 into the third category, and those greater than 0.9 into the fourth category. This divides the candidate discharge strategies into four subsets, facilitating subsequent layer-by-layer screening.

[0108] Step 402: Use a multi-level screening algorithm to screen the multiple candidate discharge strategy subsets layer by layer. Based on the comprehensive evaluation value of the candidate discharge strategies in each subset, remove candidate discharge strategies that are lower than the preset threshold and retain candidate discharge strategies that are higher than the preset threshold to obtain an initial set of candidate discharge strategies.

[0109] In this step, a multi-level screening algorithm is used to screen the multiple candidate discharge strategy subsets layer by layer. Based on the comprehensive evaluation value of the candidate discharge strategies in each subset, candidate discharge strategies below a preset threshold are eliminated, and candidate discharge strategies above the preset threshold are retained, resulting in an initial set of screened candidate discharge strategies. For example, firstly, all candidate discharge strategies are eliminated from the first category because their comprehensive evaluation values ​​are all below 0.7. Then, candidate discharge strategies with comprehensive evaluation values ​​above 0.7 are retained from the second category, those with comprehensive evaluation values ​​above 0.8 are retained from the third category, and those with comprehensive evaluation values ​​above 0.9 are retained from the fourth category. Finally, the retained candidate discharge strategies are merged into an initial set of screened candidate discharge strategies.

[0110] Step 403: Based on the weight coefficient set, each candidate discharge strategy in the initial screening candidate discharge strategy is re-evaluated. Combining real-time environmental changes and user preferences, the comprehensive evaluation value of each candidate discharge strategy is dynamically adjusted to obtain a dynamic comprehensive evaluation value set.

[0111] In this step, based on the set of weight coefficients, each candidate discharge strategy in the initial screening is re-evaluated. Taking into account real-time environmental changes and user preferences, the comprehensive evaluation value of each candidate discharge strategy is dynamically adjusted to obtain a dynamic comprehensive evaluation value set. For example, if current environmental changes cause users to pay more attention to environmental impact, the weight coefficients can be dynamically adjusted to increase the weight of environmental impact, and the comprehensive evaluation value of each candidate discharge strategy can be recalculated. This ensures that the comprehensive evaluation value better reflects the current situation and user needs.

[0112] The evaluation method combines the aforementioned dynamic comprehensive evaluation value set to recalculate the comprehensive performance of each candidate discharge strategy, thereby obtaining the target comprehensive evaluation value set.

[0113] In this step, the comprehensive performance of each candidate discharge strategy is recalculated using the fuzzy comprehensive evaluation method combined with the dynamic comprehensive evaluation value set, resulting in a target comprehensive evaluation value set. The fuzzy comprehensive evaluation method recalculates the comprehensive performance score of each candidate discharge strategy by combining the dynamic comprehensive evaluation value and weighting coefficients. For example, assuming a candidate discharge strategy has dynamic comprehensive evaluation values ​​of 0.8, 0.7, 0.9, 0.85, and 0.75 for economic cost, health status, response speed, environmental impact, and user satisfaction, respectively, and weighting coefficients of 0.2, 0.2, 0.2, 0.2, and 0.2, its target comprehensive evaluation value is calculated using the fuzzy comprehensive evaluation method. Similarly, the same calculation is performed on other candidate discharge strategies to obtain the target comprehensive evaluation value set.

[0114] Step 405: Sort the target comprehensive evaluation value set using a sorting algorithm, and select the discharge strategy with the best comprehensive performance as the optimal discharge scheme based on the sorting results.

[0115] In this step, a ranking algorithm is used to sort the target comprehensive evaluation value set. Based on the ranking result, the discharge strategy with the best overall performance is selected as the optimal discharge scheme. For example, all candidate discharge strategies in the target comprehensive evaluation value set are ranked from high to low according to their comprehensive evaluation value, and the candidate discharge strategy with the highest comprehensive evaluation value is selected as the optimal discharge scheme. The ranking algorithm ensures that the selected discharge strategy achieves an optimal balance in terms of economic cost, health status, response speed, environmental impact, and user satisfaction.

[0116] This invention employs multi-level screening and dynamic optimization of candidate discharge strategies. Specifically, through a threshold judgment mechanism and a multi-level screening algorithm, this invention can effectively screen candidate discharge strategies with high overall performance, reducing unnecessary calculations and evaluations. Dynamically adjusting the comprehensive evaluation value and recalculating the comprehensive performance ensures that the discharge strategy can adapt to real-time environmental changes and user needs, improving the system's flexibility and adaptability. Finally, a ranking algorithm selects the discharge strategy with the best overall performance, ensuring that the system achieves an optimal balance in terms of economic cost, health status, response speed, environmental impact, and user satisfaction. This method not only improves the system's operating efficiency and security but also ensures that it can provide the optimal discharge solution in various complex application scenarios.

[0117] Based on this, the present invention provides a specific embodiment. Step 304 involves introducing a multi-objective optimization algorithm to optimize the normalized comprehensive evaluation value from multiple dimensions, calculating the balance between economic cost, health status, response speed, environmental impact, and user satisfaction to obtain the target comprehensive evaluation value. Specifically, this includes the following steps:

[0118] Step 501: Utilize a multi-objective optimization algorithm to perform multi-dimensional optimization on the normalized comprehensive evaluation value, comprehensively calculate the balance between economic cost, health status, response speed, environmental impact, and user satisfaction, and obtain the multi-dimensional optimization result;

[0119] In this step, a multi-objective optimization algorithm is used to optimize the normalized comprehensive evaluation value from multiple dimensions. It comprehensively calculates the balance between economic cost, health status, response speed, environmental impact, and user satisfaction to obtain a multi-dimensional optimization result. The multi-objective optimization algorithm ensures that the final selected discharge strategy achieves the best balance across all evaluation indicators by finding the optimal solution among multiple objectives. For example, the algorithm considers how to maintain high health status, fast response speed, minimal environmental impact, and high user satisfaction while keeping economic costs low.

[0120] Step 502: Define the objective function of multi-objective optimization based on the multi-dimensional optimization results, and introduce a dynamic adjustment factor to dynamically adjust the weight of each indicator according to real-time environmental changes and user needs to obtain the optimization objective function. The objective function includes indicators of economic cost, health status, response speed, environmental impact and user satisfaction.

[0121] In this step, combining the results of multidimensional optimization, a multi-objective optimization objective function is defined, and a dynamic adjustment factor is introduced. The weights of each indicator are dynamically adjusted based on real-time environmental changes and user needs to obtain the optimized objective function. The objective function includes indicators such as economic cost, health status, response speed, environmental impact, and user satisfaction. For example, if current environmental changes cause users to pay more attention to environmental impact, the weight of environmental impact can be increased through the dynamic adjustment factor, thereby adjusting the optimized objective function.

[0122] Step 503: Run a multi-objective optimization algorithm, combine historical data and real-time data to optimize the objective function, and obtain an initial comprehensive evaluation value;

[0123] In this step, a multi-objective optimization algorithm is run, combining historical and real-time data to optimize the objective function and obtain an initial comprehensive evaluation value. The optimization algorithm finds the optimal discharge strategy under current conditions by analyzing historical and real-time data. For example, the algorithm considers the best discharge strategy under similar past conditions and combines it with current real-time data to generate a new initial comprehensive evaluation value.

[0124] Step 504: Introduce scenario simulation technology to conduct simulation tests on the discharge strategy corresponding to the initial comprehensive evaluation value under various scenarios, evaluate its performance under different environments and user needs, and obtain simulation test results. The various scenarios include: normal operation scenario, extreme weather scenario and sudden failure scenario.

[0125] In this step, scenario simulation technology is introduced to conduct simulation tests on the discharge strategy corresponding to the initial comprehensive evaluation value under various scenarios, evaluating its performance under different environments and user requirements, and obtaining simulation test results. These scenarios include normal operation scenarios, extreme weather scenarios, and sudden failure scenarios. For example, by simulating system operation under extreme weather conditions, the performance of the discharge strategy under high load and harsh environments is evaluated, ensuring stable system operation under various conditions.

[0126] Step 505: Analyze the simulation test results using machine learning algorithms to identify key influencing factors and potential risk points, and generate a risk assessment report;

[0127] In this step, machine learning algorithms are used to analyze simulation test results, identify key influencing factors and potential risk points, and generate a risk assessment report. The machine learning algorithms analyze simulation test results to identify key factors affecting system performance, such as temperature, humidity, and user load, and identify potential risk points, such as overheating and overload. The risk assessment report provides detailed analysis results to help decision-makers understand the system's performance and potential problems under different scenarios.

[0128] Step 506: Based on the risk assessment report and simulation test results, correct the initial comprehensive evaluation value to obtain the target comprehensive evaluation value;

[0129] In this step, the initial comprehensive evaluation value is corrected by combining the risk assessment report and simulation test results to obtain the target comprehensive evaluation value. By analyzing the risk assessment report and simulation test results, areas for improvement are identified, and the initial comprehensive evaluation value is adjusted to ensure that the finally selected discharge strategy exhibits optimal performance under various scenarios. For example, if simulation test results show that the system is prone to overheating under extreme weather conditions, the discharge strategy can be adjusted to reduce the discharge intensity under high-temperature conditions, thereby improving the system's stability and safety.

[0130] Furthermore, since traditional optimization methods often optimize under a single scenario, they are difficult to adapt to complex and ever-changing real-world environments. In addition, traditional optimization methods typically ignore the impact of simulation test results on the final optimization result. Therefore, an expression for correcting the target comprehensive evaluation value by incorporating the simulation test results is given:

[0131]

[0132] Among them, V final The final comprehensive evaluation value reflects the overall performance of the discharge strategy under various scenarios, providing a basis for selecting the optimal discharge strategy; V i S represents the comprehensive evaluation value of the objective under the i-th scenario. This value is obtained by optimizing the algorithm using a multi-objective optimization method that combines historical and real-time data. i This represents the simulation test result for scenario i. This value reflects the performance of the discharge strategy in scenario i. The simulation test result can be a score, indicating the performance of the discharge strategy in that scenario; β i W represents the weighting index of the simulation test results under the i-th scenario. This index can be dynamically adjusted according to the importance and complexity of the scenario to reflect the influence of the simulation test results under different scenarios. i γ represents the weight for the i-th scenario. This weight can be set according to the frequency and importance of the scenario to ensure that different scenarios have a reasonable impact on the final result; i This is the dynamic adjustment factor for the i-th scenario. This factor can be dynamically adjusted according to real-time environmental changes and user needs, reflecting adaptability under different scenarios; D i Let be the deviation value under the i-th scenario. This value reflects the deviation between the simulation test results and the preset target, and can be used to adjust the weight of the simulation test results; exp(-γ i ·D i The exponential decay function is used to adjust the weight of simulation test results. Through this function, the impact of simulation test results under different scenarios on the final comprehensive evaluation value can be dynamically adjusted.

[0133] This formula, through multi-level weights, weight exponents, dynamic adjustment factors, and deviation values, more precisely reflects the impact of simulation test results under different scenarios on the initial comprehensive evaluation value, thereby improving the robustness and adaptability of the discharge strategy. It comprehensively considers simulation test results and their weights under multiple scenarios to ensure the final comprehensive evaluation value V... final This reflects the final overall performance of the discharge strategy under various scenarios;

[0134] The following is a brief introduction to the design rationale behind each term of the formula:

[0135] Regarding the first part of the items

[0136] This part reflects the simulation test results S i The influence under different scenarios. This is determined by introducing the exponent β. i The importance of simulation results can be dynamically adjusted. If a scenario is particularly important or complex, the importance can be increased by adding β. iThis increases the weight of simulation results in this scenario, thereby ensuring that critical scenarios receive sufficient attention.

[0137] W i Each scenario has its inherent importance, which may be based on factors such as its frequency of occurrence and its impact on the system. This can be addressed by setting different weights W. i This allows the model to focus more on the more critical scenarios while appropriately reducing the impact of less important scenarios.

[0138] exp(-γ i ·D i This section adjusts the weights of the simulation results using an exponential decay function. When the deviation D between the simulation results and the preset target... i A large value indicates poor performance in that scenario; in this case, exp(-γ) can be used to... i ·D i To minimize the impact of abnormal situations, a higher weight is maintained if the deviation is small; conversely, if the deviation is small, a higher weight is maintained. This design helps reduce the impact of abnormal situations on the final decision while retaining sensitivity within the normal range.

[0139] Regarding the second part of the items

[0140] This section describes the comprehensive evaluation value V of the target under the i-th scenario. i The result is the product of the above correction factors. Its design aims to combine the ideal evaluation V given by a multi-objective optimization algorithm. i And actual simulation test results S i Performance, by considering the importance and adaptability of the situation (via β) i W i ,γ i D i To adjust V i This makes the final evaluation more closely reflect the actual situation and better reflect the comprehensive performance of the discharge strategy under various scenarios.

[0141] The following is a brief introduction to how the parameters of this formula are obtained:

[0142] By combining historical and real-time data with multi-objective optimization algorithms (such as genetic algorithms and particle swarm optimization), the discharge strategy is optimized to obtain the comprehensive evaluation value V for each scenario. i Using simulation software or platforms, a simulation model of the mobile cabin power supply system is constructed. The discharge strategy corresponding to the initial comprehensive evaluation value is input, parameters and conditions under different scenarios are set, simulation tests are run, and simulation test results S are obtained. iThe weighting is dynamically adjusted based on the importance and complexity of the scenario. For example, some scenarios may have a greater impact on system performance, and therefore can be assigned a higher weighting index β. i W i The weighting should be adjusted based on the frequency and importance of the scenario; for example, common scenarios can be assigned a higher weight, while rare but important scenarios can also have their weighting appropriately increased. i It dynamically adjusts based on real-time environmental changes and user needs. For example, when user needs change or environmental conditions change, γ can be dynamically adjusted. i The value of ; by comparing the difference between the simulation test results and the preset target, the deviation value D is calculated. i .

[0143] Suppose there are three scenarios, and the parameters for each scenario are as follows:

[0144] Scenario 1: V1:0.8, S1:0.9, β1:1.2, W1:0.7, γ1:0.5, D1:0.2; Scenario 2: V2:0.7, S2:0.8, β2:1.1, W2:0.6, γ2:1.0, D2:0.5; Scenario 3: V3:0.6, S3:0.7, β3:1.0, W3:0.5, γ3:0.8, D3:1.0;

[0145] First, calculate the weight term for each scenario:

[0146] The calculation method for scenario 1 is as follows:

[0147]

[0148] The calculation method for scenario 2 is as follows:

[0149]

[0150] The calculation method for scenario 3 is as follows:

[0151]

[0152] Next, the numerator is calculated as follows:

[0153]

[0154] Then, calculate the denominator as follows:

[0155]

[0156] The final comprehensive evaluation value is calculated as follows:

[0157]

[0158] Through the above calculations, the final comprehensive evaluation value V of the target was obtained. final The value is approximately 0.739. This value takes into account the simulation test results and their weights under various scenarios, and reflects the final comprehensive performance of the discharge strategy under various scenarios, which helps to select the optimal discharge strategy.

[0159] This invention provides a comprehensive evaluation and optimization of candidate discharge strategies. Specifically, through a multi-objective optimization algorithm, it finds the optimal solution among multiple evaluation indicators, ensuring that the discharge strategy achieves the best balance in terms of economic cost, health status, response speed, environmental impact, and user satisfaction. The introduction of a dynamic adjustment factor allows for dynamic adjustment of the weights of each indicator based on real-time environmental changes and user needs, improving flexibility and adaptability. Scenario simulation technology and simulation testing ensure stability and reliability under various complex environments. Machine learning algorithms analyze simulation test results to identify key influencing factors and potential risk points, generating risk assessment reports that provide a scientific basis for decision-making. Finally, by correcting the initial comprehensive evaluation value, a target comprehensive evaluation value is obtained, ensuring optimal performance under various scenarios.

[0160] Based on this, the present invention provides a specific embodiment. Based on step 106, an intelligent scheduling algorithm is used to output discharge commands to the different types of energy storage units according to the optimal discharge scheme, dynamically adjusting the charging and discharging behavior of each energy storage unit to obtain the optimal operating state of the container power supply system. The specific steps include the following:

[0161] Step 601: Using an intelligent scheduling algorithm, generate a discharge command based on the optimal discharge scheme. The discharge command includes the charging and discharging power, time, and sequence of each energy storage unit.

[0162] In this step, an intelligent scheduling algorithm generates discharge instructions based on the optimal discharge scheme. The discharge instructions include the charging and discharging power, time, and sequence for each energy storage unit. For example, assuming the optimal discharge scheme requires the lithium-ion battery to discharge at 10kW from 10 AM to 12 PM, the lead-acid battery to charge at 8kW from 2 PM to 4 PM, and the supercapacitor to discharge at 5kW from 8 PM to 10 PM, the intelligent scheduling algorithm will generate corresponding discharge instructions to ensure that each energy storage unit is charged and discharged according to the predetermined power, time, and sequence.

[0163] Step 602: Use the real-time communication module to send the discharge command to the controller of the different types of energy storage units to obtain the execution result of the discharge command;

[0164] In this step, a real-time communication module is used to send the generated discharge command to the controllers of different types of energy storage units, obtaining the execution results of the discharge command. The real-time communication module ensures that the discharge command can be quickly and accurately transmitted to the controller of each energy storage unit, and the controller returns the execution result after executing the command, so that the central control system can make subsequent adjustments and optimizations. For example, after receiving the discharge command, the controller will start the corresponding charging and discharging process and feed back the execution status (such as actual charging and discharging power, time, etc.) to the central control system.

[0165] Step 603: Dynamically adjust the charging and discharging behavior of each energy storage unit based on the execution results, and combine the real-time monitored energy storage unit status parameters to obtain the adjusted charging and discharging behavior;

[0166] In this step, based on the execution results and the real-time monitored state parameters of the energy storage units, the charging and discharging behavior of each energy storage unit is dynamically adjusted to obtain the adjusted charging and discharging behavior. For example, if the actual charging and discharging power of a certain energy storage unit does not match the command, or if its temperature, voltage, or other state parameters exceed the safe range, the central control system will immediately adjust the charging and discharging behavior of that energy storage unit to ensure stable system operation. The adjusted charging and discharging behavior will be more in line with the current actual operating conditions.

[0167] Step 604: By monitoring the operating status of the energy storage unit in real time and using a feedback control mechanism, the adjusted charging and discharging behavior is optimized to obtain the optimized system operating status;

[0168] In this step, the operating status of the energy storage units is monitored in real time, and the adjusted charging and discharging behavior is optimized using a feedback control mechanism to obtain an optimized system operating state. For example, the central control system continuously monitors parameters such as the state of charge, temperature, voltage, current, and internal resistance of each energy storage unit. Based on changes in these parameters, the charging and discharging behavior is dynamically adjusted to ensure that the system is always in the optimal operating state, and the feedback control mechanism can ensure a rapid response to any abnormal situations.

[0169] Step 605: Incorporate adaptive control algorithm and multi-objective optimization algorithm. Based on the deviation between the optimized system operating state and the preset target, automatically adjust the discharge strategy, calculate the balance between economic cost, health status, response speed, environmental impact and user satisfaction, combine historical data and real-time data to make long-term trend prediction of the discharge strategy, adjust the discharge strategy in advance, and obtain the optimal operating state of the cabin power supply system.

[0170] In this step, adaptive control and multi-objective optimization algorithms are introduced to automatically adjust the discharge strategy based on the deviation between the optimized system operating state and the preset target. The balance between economic cost, health status, response speed, environmental impact, and user satisfaction is calculated. Combining historical and real-time data, long-term trend predictions of the discharge strategy are made, allowing for advance adjustments to the discharge strategy and achieving the optimal operating state of the shelter power supply system. For example, the adaptive control algorithm automatically adjusts the discharge strategy based on the current operating state and changes in the external environment, ensuring stable system operation under various conditions. The multi-objective optimization algorithm finds the optimal balance among economic cost, health status, response speed, environmental impact, and user satisfaction, generating a new discharge strategy. By combining historical and real-time data, future operating trends can be predicted, allowing for advance adjustments to the discharge strategy and ensuring the system is always in optimal operating condition.

[0171] This invention achieves comprehensive optimization and intelligent management of the cabin power supply system. Specifically, an intelligent scheduling algorithm generates discharge commands, ensuring that each energy storage unit charges and discharges according to the optimal plan. A real-time communication module and feedback control mechanism ensure accurate execution of commands and real-time system adjustments, improving system stability and reliability. Adaptive control and multi-objective optimization algorithms, combined with historical and real-time data, adjust the discharge strategy in advance, ensuring the system exhibits optimal performance under various conditions. This approach not only improves system operating efficiency and safety but also ensures an optimal balance in terms of economic cost, health status, response speed, environmental impact, and user satisfaction.

[0172] Based on this, the present invention provides a specific embodiment. Based on step 107, based on the optimal operating state of the cabin power supply system, a machine learning algorithm is used to learn the deviation between the actual effect of the system operation and the preset target during the discharge process, automatically adjust the model parameters, optimize the discharge strategy, and form an adaptive closed-loop control mechanism, specifically including the following steps:

[0173] Step 701: Use the machine learning algorithm in the deep reinforcement learning model to monitor the optimal operating state of the modular power supply system in real time, and collect the actual effect data of the system operation. The actual effect data includes: economic cost, health status, response speed, environmental impact and user satisfaction.

[0174] In this step, machine learning algorithms within a deep reinforcement learning model are used to monitor the optimal operating state of the mobile power supply system in real time, collecting actual performance data of the system. This performance data includes economic costs, health status, response speed, environmental impact, and user satisfaction. For example, real-time monitoring of parameters such as the state of charge, temperature, voltage, current, and internal resistance of the energy storage unit via a sensor network, along with information such as external electricity market price fluctuations, user load forecasts, and weather forecasts, generates the actual performance data.

[0175] Step 702: Compare the actual effect data with the preset target, calculate the deviation value of each indicator, and obtain the deviation analysis results;

[0176] In this step, the collected actual results data are compared with the preset targets, and the deviation values ​​of each indicator are calculated to obtain the deviation analysis results. For example, if the economic cost in the preset target is 100 yuan / hour, while the economic cost in the actual results data is 120 yuan / hour, then the deviation value of the economic cost is 20 yuan / hour. Similarly, deviation analysis is performed on indicators such as health status, response speed, environmental impact, and user satisfaction to obtain the deviation values ​​of each indicator.

[0177] Step 703: Based on the deviation analysis results, automatically adjust the parameters in the deep reinforcement learning model and the intelligent scheduling algorithm to optimize the prediction and control capabilities of the deep reinforcement learning model and obtain the optimized deep reinforcement learning model parameters;

[0178] In this step, based on the deviation analysis results, the parameters in the deep reinforcement learning model and the intelligent scheduling algorithm are automatically adjusted to optimize the prediction and control capabilities of the deep reinforcement learning model, resulting in optimized deep reinforcement learning model parameters. For example, if the deviation value of economic cost is large, the parameters related to economic cost in the deep reinforcement learning model can be adjusted to improve the model's prediction accuracy and control capability. Through continuous adjustment and optimization, it is ensured that the model can better adapt to actual operating conditions.

[0179] Step 704: Introduce an adaptive control algorithm to dynamically adjust the discharge strategy based on the optimized deep reinforcement learning model parameters, and obtain the adaptively adjusted discharge strategy;

[0180] In this step, an adaptive control algorithm is introduced to dynamically adjust the discharge strategy based on the optimized deep reinforcement learning model parameters, resulting in an adaptively adjusted discharge strategy. The adaptive control algorithm automatically adjusts the discharge strategy according to the current operating state and changes in the external environment, ensuring stable system operation under various conditions. For example, if the system detects that a weather forecast indicates extreme weather in the next few days, the adaptive control algorithm will adjust the discharge strategy in advance to ensure stability and safety under extreme weather conditions.

[0181] Step 705: Combining historical and real-time data, using machine learning algorithms to predict the long-term trend of the adaptively adjusted discharge strategy, identify potential problems and risks in advance, optimize the discharge strategy again, and form an adaptive closed-loop control mechanism.

[0182] In this step, historical and real-time data are combined, and machine learning algorithms are used to predict the long-term trend of the adaptively adjusted discharge strategy. This allows for the early detection of potential problems and risks, followed by further optimization of the discharge strategy, forming an adaptive closed-loop control mechanism. For example, by analyzing historical and real-time data, the system's operating trend over a future period can be predicted, identifying potential overloads, overheating, and other issues in advance, and adjusting the discharge strategy to avoid potential risks. Through continuous prediction and optimization, the system is ensured to always operate at its optimal state.

[0183] This invention provides a comprehensive monitoring and intelligent optimization system for the cabin power supply system. Specifically, it uses a deep reinforcement learning model and machine learning algorithms to monitor the actual performance data of the system in real time and compare it with preset targets, calculating the deviation values ​​of each indicator. Based on the deviation analysis results, the parameters in the deep reinforcement learning model and intelligent scheduling algorithm are automatically adjusted to optimize the model's predictive and control capabilities. The adaptive control algorithm dynamically adjusts the discharge strategy according to the optimized model parameters to ensure stable system operation under various conditions. Through long-term trend prediction, potential problems and risks are identified in advance, and the discharge strategy is further optimized to form an adaptive closed-loop control mechanism. This method not only improves the system's operating efficiency and safety but also ensures an optimal balance in terms of economic cost, health status, response speed, environmental impact, and user satisfaction.

[0184] Figure 2 This application provides a schematic diagram of the structure of an optimal discharge control system for a container based on hybrid energy storage, as shown in the embodiment of the present application. Figure 2 As shown, the system includes:

[0185] The monitoring module 21 uses a sensor network to monitor and record the status parameters of different types of energy storage units in the cabin in real time, and obtains detailed status information of each energy storage unit. The detailed status information includes: state of charge, temperature, voltage, current and internal resistance. The different types of energy storage units include: lithium-ion batteries, lead-acid batteries and supercapacitors.

[0186] Analysis module 22 uses the deep reinforcement learning algorithm in the deep reinforcement model to comprehensively analyze the detailed state information, external electricity market price fluctuations, user load forecasts and weather forecasts, and generate a discharge strategy;

[0187] Optimization module 23 uses a genetic algorithm to perform multi-objective optimization on the discharge strategy, comprehensively calculating economic cost, health status, response speed, and environmental impact to obtain multiple sets of candidate discharge strategies;

[0188] Evaluation module 24 uses a fuzzy logic decision support system to evaluate each of the multiple candidate discharge strategies, and selects the optimal discharge scheme by setting different weight coefficients to reflect the priority under different scenarios.

[0189] Output module 25 uses an intelligent scheduling algorithm to output discharge commands to different types of energy storage units according to the optimal discharge scheme, dynamically adjusts the charging and discharging behavior of each energy storage unit, and obtains the optimal operating state of the container power supply system.

[0190] The adjustment module 26, based on the optimal operating state of the cabin power supply system, uses machine learning algorithms to learn the deviation between the actual effect of the system operation and the preset target during the discharge process, automatically adjusts the model parameters, optimizes the discharge strategy, and forms an adaptive closed-loop control mechanism.

[0191] Figure 2 The aforementioned optimal discharge control system for modular container based on hybrid energy storage can execute... Figure 1 The implementation principle and technical effects of the optimal discharge control method for a container based on hybrid energy storage described in the illustrated embodiment will not be repeated here. The specific operation methods of each module and unit in the optimal discharge control system for a container based on hybrid energy storage in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0192] Figure 2 The optimal discharge control system for a container based on hybrid energy storage, as shown in the embodiment, 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;

[0193] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0194] The processing component 32 is used to: monitor and record the status parameters of different types of energy storage units in the cabin in real time using a sensor network to obtain detailed status information of each energy storage unit; use the deep reinforcement learning algorithm in the deep reinforcement model to comprehensively analyze the detailed status information, external electricity market price fluctuations, user load forecasts and weather forecasts to generate a discharge strategy; use a genetic algorithm to perform multi-objective optimization processing on the discharge strategy, comprehensively calculate economic costs, health status, response speed and environmental impact to obtain multiple sets of candidate discharge strategies; use a fuzzy logic decision support system to evaluate each of the multiple sets of candidate discharge strategies, set different weight coefficients to reflect the priority under different scenarios, select the optimal discharge scheme; use an intelligent scheduling algorithm to output discharge commands to the different types of energy storage units according to the optimal discharge scheme, dynamically adjust the charging and discharging behavior of each energy storage unit to obtain the optimal operating state of the cabin power supply system; based on the optimal operating state of the cabin power supply system, use a machine learning algorithm to learn the deviation between the actual effect of the system operation and the preset target during the discharge process, automatically adjust the model parameters, optimize the discharge strategy, and form an adaptive closed-loop control mechanism.

[0195] The processing component 32 includes one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component can be implemented as one or more application-specific integrated circuits (ASICs), 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-described method.

[0196] 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 storage, flash memory, magnetic disk, or optical disk.

[0197] Computing devices also include other components such as input / output interfaces, display components, and communication components.

[0198] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices or input devices.

[0199] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0200] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0201] This invention also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown presents an optimal discharge control method and system for a container based on hybrid energy storage.

[0202] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0203] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0204] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0205] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimal discharge control of a mobile cabin based on hybrid energy storage, characterized in that, include: The sensor network is used to monitor and record the status parameters of different types of energy storage units in the cabin in real time, so as to obtain detailed status information of each energy storage unit. The detailed status information includes: state of charge, temperature, voltage, current and internal resistance. The different types of energy storage units include: lithium-ion batteries, lead-acid batteries and supercapacitors. The deep reinforcement learning algorithm in the deep reinforcement model is used to comprehensively analyze the detailed state information, external electricity market price fluctuations, user load forecasts and weather forecasts to generate a discharge strategy. The discharge strategy is optimized using a genetic algorithm to comprehensively calculate economic cost, health status, response speed, and environmental impact, resulting in multiple candidate discharge strategies. The fuzzy logic decision support system is used to evaluate each of the multiple candidate discharge strategies. Different weight coefficients are set to reflect the priority under different scenarios, and the optimal discharge scheme is selected. The evaluation process of each candidate discharge strategy using the fuzzy logic decision support system includes: using a multi-level fuzzy logic model combined with a set of weight coefficients to quantify the performance of each candidate discharge strategy on each evaluation index, and obtaining a fuzzy membership function value set for each candidate discharge strategy; applying the fuzzy comprehensive evaluation method combined with the fuzzy membership function value set and the set of weight coefficients to calculate the comprehensive performance of each candidate discharge strategy, and obtaining a comprehensive evaluation value for each candidate discharge strategy. Using an intelligent scheduling algorithm, discharge commands are output to the different types of energy storage units according to the optimal discharge scheme, and the charging and discharging behavior of each energy storage unit is dynamically adjusted to obtain the optimal operating state of the container power supply system. Based on the optimal operating state of the modular power supply system, a machine learning algorithm is used to learn the deviation between the actual operation of the system and the preset target during the discharge process, automatically adjust the model parameters, optimize the discharge strategy, and form an adaptive closed-loop control mechanism.

2. The method according to claim 1, characterized in that, Also includes: By using scenario demand analysis to analyze the specific needs of the current application scenario, multiple evaluation indicators are obtained, including: economic cost, health status, response speed, environmental impact, and user satisfaction. The multiple evaluation indicators are defined based on the specific requirements of the current application scenario, resulting in a set of evaluation indicators; By using preset standards, real-time environmental changes, and user preferences, an adjustable weight coefficient is set for each evaluation indicator in the evaluation indicator set, resulting in a dynamically adjusted set of weight coefficients. By setting different weighting coefficients to reflect the priorities in different scenarios, the optimal discharge scheme is selected, including: A threshold judgment mechanism is used to filter the comprehensive evaluation value to obtain a candidate discharge strategy set. By comparing the comprehensive evaluation value of each strategy in the candidate discharge strategy set, the discharge strategy with the best comprehensive performance is selected to obtain the optimal discharge scheme.

3. The method according to claim 2, characterized in that, The fuzzy comprehensive evaluation method is applied, combining the fuzzy membership function value set and the weight coefficient set, to calculate the comprehensive performance of each candidate discharge strategy, obtaining the comprehensive evaluation value of each candidate discharge strategy, including: The weighted membership value of each evaluation index is obtained by multiplying each value in the set of fuzzy membership function values ​​with the weight coefficients in the set of corresponding weight coefficients using the fuzzy comprehensive evaluation method. The weighted membership values ​​of all evaluation indicators for each candidate discharge strategy are summed to obtain a preliminary comprehensive evaluation value; The preliminary comprehensive evaluation value is normalized to the interval [0, 1] using a normalization method to obtain the normalized comprehensive evaluation value. A multi-objective optimization algorithm is introduced to optimize the normalized comprehensive evaluation value in multiple dimensions, calculating the balance between economic cost, health status, response speed, environmental impact and user satisfaction, and obtaining the target comprehensive evaluation value.

4. The method according to claim 2, characterized in that, A threshold judgment mechanism is used to filter the comprehensive evaluation values ​​to obtain a candidate discharge strategy set. By comparing the comprehensive evaluation values ​​of each strategy in the candidate discharge strategy set, the discharge strategy with the best comprehensive performance is selected to obtain the optimal discharge scheme, including: The comprehensive evaluation value is filtered using a threshold judgment mechanism. Multiple preset thresholds are set, and the comprehensive evaluation value is classified according to the preset thresholds to obtain multiple candidate discharge strategy subsets. The multiple candidate discharge strategy subsets are screened layer by layer using a multi-level screening algorithm. Based on the comprehensive evaluation value of the candidate discharge strategies in each subset, candidate discharge strategies below a preset threshold are eliminated, and candidate discharge strategies above a preset threshold are retained to obtain an initial set of candidate discharge strategies. Based on the set of weight coefficients, each candidate discharge strategy in the initial screening of candidate discharge strategies is re-evaluated. Combining real-time environmental changes and user preferences, the comprehensive evaluation value of each candidate discharge strategy is dynamically adjusted to obtain a dynamic comprehensive evaluation value set. By using the fuzzy comprehensive evaluation method in conjunction with the dynamic comprehensive evaluation value set, the comprehensive performance of each candidate discharge strategy is recalculated to obtain the target comprehensive evaluation value set; The target comprehensive evaluation value set is sorted using a sorting algorithm. Based on the sorting results, the discharge strategy with the best comprehensive performance is selected as the optimal discharge scheme.

5. The method according to claim 3, characterized in that, A multi-objective optimization algorithm is introduced to optimize the normalized comprehensive evaluation value from multiple dimensions, calculating the balance between economic cost, health status, response speed, environmental impact, and user satisfaction, to obtain the target comprehensive evaluation value, including: The normalized comprehensive evaluation value is optimized in multiple dimensions using a multi-objective optimization algorithm. The balance between economic cost, health status, response speed, environmental impact and user satisfaction is comprehensively calculated to obtain the multi-dimensional optimization result. Based on the multidimensional optimization results, a multi-objective optimization objective function is defined, and a dynamic adjustment factor is introduced to dynamically adjust the weights of each indicator according to real-time environmental changes and user needs, thereby obtaining the optimization objective function. The objective function includes indicators such as economic cost, health status, response speed, environmental impact, and user satisfaction. A multi-objective optimization algorithm is run, and the objective function is optimized by combining historical data and real-time data to obtain an initial comprehensive evaluation value; Scenario simulation technology is introduced to conduct simulation tests on the discharge strategy corresponding to the initial comprehensive evaluation value under various scenarios, evaluate its performance under different environments and user needs, and obtain simulation test results. The various scenarios include: normal operation scenario, extreme weather scenario and sudden failure scenario. The simulation test results are analyzed using machine learning algorithms to identify key influencing factors and potential risk points, and a risk assessment report is generated. Based on the risk assessment report and simulation test results, the initial comprehensive evaluation value is corrected to obtain the target comprehensive evaluation value.

6. The method according to claim 1, characterized in that, Using an intelligent scheduling algorithm, based on the optimal discharge scheme, discharge commands are output to the different types of energy storage units, dynamically adjusting the charging and discharging behavior of each energy storage unit to obtain the optimal operating state of the modular power supply system, including: Based on the optimal discharge scheme, an intelligent scheduling algorithm is used to generate discharge instructions, which include the charging and discharging power, time, and sequence of each energy storage unit. The discharge command is sent to the controller of the different types of energy storage units using a real-time communication module to obtain the execution result of the discharge command; The charging and discharging behavior of each energy storage unit is dynamically adjusted based on the execution results, and the adjusted charging and discharging behavior is obtained by combining the real-time monitored energy storage unit status parameters. By monitoring the operating status of the energy storage unit in real time and using a feedback control mechanism, the adjusted charging and discharging behavior is optimized to obtain the optimized system operating status. Adaptive control algorithm and multi-objective optimization algorithm are introduced. Based on the deviation between the optimized system operating state and the preset target, the discharge strategy is automatically adjusted. The balance between economic cost, health status, response speed, environmental impact and user satisfaction is calculated. By combining historical data and real-time data, long-term trend prediction of the discharge strategy is made, and the discharge strategy is adjusted in advance to obtain the optimal operating state of the container power supply system.

7. The method according to claim 1, characterized in that, Based on the optimal operating state of the modular power supply system, a machine learning algorithm is used to learn the deviation between the actual system operation and the preset target during the discharge process, automatically adjust the model parameters, optimize the discharge strategy, and form an adaptive closed-loop control mechanism, including: The optimal operating state of the modular power supply system is monitored in real time using machine learning algorithms in a deep reinforcement learning model. The actual performance data of the system operation is collected, including: economic cost, health status, response speed, environmental impact and user satisfaction. The actual performance data is compared with the preset target, the deviation value of each indicator is calculated, and the deviation analysis results are obtained. Based on the results of deviation analysis, the parameters in the deep reinforcement learning model and intelligent scheduling algorithm are automatically adjusted to optimize the prediction and control capabilities of the deep reinforcement learning model and obtain the optimized parameters of the deep reinforcement learning model. An adaptive control algorithm is introduced to dynamically adjust the discharge strategy based on the optimized deep reinforcement learning model parameters, thereby obtaining an adaptively adjusted discharge strategy. By combining historical and real-time data, machine learning algorithms are used to predict the long-term trend of the adaptively adjusted discharge strategy, identify potential problems and risks in advance, and further optimize the discharge strategy to form an adaptive closed-loop control mechanism.

8. An optimal discharge control system for a container based on hybrid energy storage, characterized in that, include: The monitoring module uses a sensor network to monitor and record the status parameters of different types of energy storage units in the cabin in real time, and obtains detailed status information of each energy storage unit. The detailed status information includes: state of charge, temperature, voltage, current and internal resistance. The different types of energy storage units include: lithium-ion batteries, lead-acid batteries and supercapacitors. The analysis module uses the deep reinforcement learning algorithm in the deep reinforcement model to comprehensively analyze the detailed state information, external electricity market price fluctuations, user load forecasts and weather forecasts, and generate a discharge strategy. The optimization module uses a genetic algorithm to perform multi-objective optimization of the discharge strategy, comprehensively calculating economic cost, health status, response speed, and environmental impact to obtain multiple sets of candidate discharge strategies. The evaluation module utilizes a fuzzy logic decision support system to evaluate each of the multiple candidate discharge strategies. By setting different weight coefficients to reflect the priority under different scenarios, the optimal discharge scheme is selected. The process of evaluating each of the multiple candidate discharge strategies using the fuzzy logic decision support system includes: using a multi-level fuzzy logic model combined with a set of weight coefficients to quantify the performance of each candidate discharge strategy on each evaluation index, obtaining a fuzzy membership function value set for each candidate discharge strategy; and applying a fuzzy comprehensive evaluation method combined with the fuzzy membership function value set and the set of weight coefficients to calculate the comprehensive performance of each candidate discharge strategy, obtaining a comprehensive evaluation value for each candidate discharge strategy. The output module uses an intelligent scheduling algorithm to output discharge commands to the different types of energy storage units according to the optimal discharge scheme, dynamically adjusts the charging and discharging behavior of each energy storage unit, and obtains the optimal operating state of the container power supply system. The adjustment module, based on the optimal operating state of the modular power supply system, uses machine learning algorithms to learn the deviation between the actual operation of the system and the preset target during the discharge process, automatically adjusts the model parameters, optimizes the discharge strategy, and forms an adaptive closed-loop control mechanism.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the optimal discharge control method for a container based on hybrid energy storage as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements the optimal discharge control method for a container based on hybrid energy storage as described in any one of claims 1 to 7.

Images