Distributed energy optimization management method and system for rapidly deploying mobile shelter hospital

Abstract

The invention provides a distributed energy optimization management method and system for rapidly deploying a mobile shelter hospital, and the method comprises the steps: carrying out the dynamic adjustment of energy distribution through constructing a multi-layer virtual energy network model of the mobile shelter hospital; predicting the energy consumption demand and the power generation potential by using a time sequence prediction technology, and generating an energy supply and demand balance plan; introducing a multi-objective optimization algorithm to intelligently schedule the charging and discharging process of the energy storage system to obtain an optimal charging and discharging plan; the operation data of the distributed energy system is collected and analyzed at a preset time node, fault risk prediction is performed on the operation data through a machine learning model, and an optimal energy distribution strategy and a multi-objective optimization algorithm are optimized according to a result, so that a self-adaptive and self-learning energy closed-loop management system is formed. According to the technical scheme provided by the invention, dynamic optimization management of mobile shelter hospital energy is realized, efficient, flexible and adaptive energy distribution and scheduling are ensured, and the overall performance and reliability of the system are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of energy management for mobile cabin hospitals, and in particular to a distributed energy optimization management method for rapidly deploying mobile cabin hospitals. Background Art

[0002] With the frequent occurrence of public health emergencies, the rapid deployment of mobile shelter hospitals has become an important means to deal with large-scale epidemics. Mobile shelter hospitals need to be built and put into use in a short period of time, which places extremely high demands on energy supply. Especially in remote areas or with limited resources, how to ensure that the energy supply of mobile shelter hospitals is both efficient and reliable has become an urgent problem to be solved. In addition, mobile shelter hospitals are usually equipped with a large number of medical equipment, and the power demand of these equipment is highly uncertain and volatile, so a flexible and intelligent energy management system is needed to cope with this complex power consumption environment.

[0003] Although the existing energy management solutions have met the basic electricity needs of mobile cabin hospitals to a certain extent, there are still problems such as insufficient flexibility, low intelligence level and high maintenance cost. The traditional centralized energy supply method cannot dynamically adjust the energy allocation strategy according to the real-time electricity demand of medical equipment and the power generation of renewable energy, resulting in energy waste or supply shortage from time to time. The existing energy management system lacks intelligent scheduling capabilities and cannot fully utilize the advantages of renewable energy and energy storage systems. Especially in the face of complex and changing power consumption environments, the system's adaptability and robustness are poor. In addition, due to the lack of effective fault prediction and self-healing functions, the existing energy management system is prone to failure during operation, requiring frequent manual intervention and maintenance, which increases operation and maintenance costs and workload. Summary of the invention

[0004] The embodiments of the present invention provide a distributed energy optimization management method and system for rapidly deploying mobile cabin hospitals, so as to solve the problem that the traditional centralized energy supply method in the prior art cannot dynamically adjust the energy allocation strategy according to the real-time power demand of medical equipment and the power generation of renewable energy, resulting in energy waste or frequent shortage of energy.

[0005] In a first aspect, an embodiment of the present invention provides a distributed energy optimization management method for rapidly deploying a mobile cabin hospital, including:

[0006] Construct a multi-layer virtual energy network model for mobile cabin hospitals;

[0007] Based on the multi-layer virtual energy network model, a deep reinforcement learning algorithm is applied to dynamically adjust energy distribution and generate an optimal energy distribution strategy;

[0008] Combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, using time series forecasting technology to predict energy consumption demand and power generation potential, and generate an energy supply and demand balance plan;

[0009] Based on the energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system to obtain the optimal charging and discharging plan;

[0010] Based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted, the operating data of the distributed energy system is collected and analyzed at preset time nodes, the fault risk of the operating data is predicted through a machine learning model, and an evaluation result is generated. The optimal energy allocation strategy and multi-objective optimization algorithm are iteratively optimized according to the evaluation results to form an adaptive and self-learning energy closed-loop management system.

[0011] Optionally, based on the energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system to obtain an optimal charging and discharging plan, including:

[0012] Using the current state information of the energy storage system and combining it with the energy supply and demand balance plan, a dynamic mathematical model of the energy storage system is constructed;

[0013] Using the dynamic mathematical model, combined with the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, a multi-objective optimization function is designed;

[0014] Applying the evolutionary computing method in combination with the fuzzy logic controller, the multi-objective optimization function is solved to obtain a Pareto frontier consisting of a set of non-inferior solutions;

[0015] A multi-criteria decision support system is used to integrate expert knowledge and user preferences to select the optimal charging and discharging plan from the Pareto frontier.

[0016] Optionally, the dynamic mathematical model is used to design a multi-objective optimization function in combination with the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, including:

[0017] Using the dynamic mathematical model, combined with the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, the target variables of the multi-objective optimization function are defined;

[0018] Designing a specific expression of a multi-objective optimization function according to the target variables, wherein the expression includes each target variable and a relative importance weight of each target variable;

[0019] Using the specific expression in combination with historical data and real-time data, calibrate the parameters in the multi-objective optimization function to obtain a calibrated multi-objective optimization function;

[0020] By utilizing the calibrated multi-objective optimization function and combining the external environment changes and the internal demand changes, the weight coefficients in the multi-objective optimization function are dynamically adjusted to generate a multi-objective optimization function that adapts to different operating conditions.

[0021] Optionally, it is characterized in that the weight coefficients in the multi-objective optimization function are dynamically adjusted by using the calibrated multi-objective optimization function in combination with external environment changes and internal demand changes to generate a multi-objective optimization function adapted to different operating conditions, including:

[0022] Using the calibrated multi-objective optimization function, combined with external environment changes and internal demand changes, the importance change trend of each target variable is analyzed to obtain the importance change trend analysis results;

[0023] Designing a weight adjustment rule based on the importance change trend analysis result, wherein the weight adjustment rule includes the importance difference of each target variable in different time periods;

[0024] Dynamically adjusting the weight coefficients in the multi-objective optimization function using the weight adjustment rule to obtain dynamic weight coefficients;

[0025] The dynamic weight coefficient is applied to the multi-objective optimization function to generate a multi-objective optimization function that is adaptable to different operating conditions.

[0026] Optionally, it is characterized in that based on the multi-layer virtual energy network model, a deep reinforcement learning algorithm is applied to dynamically adjust energy distribution to generate an optimal energy distribution strategy, including:

[0027] The multi-layer virtual energy network model is used to integrate the energy consumption forecast information of the mobile cabin hospital, the real-time power generation of renewable energy, the price fluctuation of the external power grid, and the urgency of the mobile cabin hospital. At the same time, the uncertainty of weather changes is calculated, and the possible impact of weather changes is simulated through a probability distribution model to obtain the input state;

[0028] Based on the input state, the priority of electricity demand of the mobile cabin hospital, the availability of renewable energy, the price signal of the external power grid and the uncertainty of weather changes are set as a multidimensional reward function, where the stable power supply and cost saving of the emergency medical unit are the main optimization goals, and a multidimensional reward function is obtained;

[0029] Processing the input state and the multidimensional reward function using a deep reinforcement learning model based on an Actor-Critic architecture to obtain an initial energy allocation strategy;

[0030] Based on the initial energy allocation strategy, during the training process, the experience replay mechanism is used to randomly extract historical data for replay learning, thereby improving the generalization ability and stability of the model and obtaining an enhanced energy allocation strategy;

[0031] Based on the enhanced energy allocation strategy, a specific emergency response strategy is designed for a preset abnormal situation to obtain an energy allocation strategy that adapts to the abnormal situation;

[0032] Based on the energy allocation strategy that adapts to abnormal situations, during actual operation, online learning is used to collect the latest energy consumption data of mobile cabin hospitals, real-time power generation of renewable energy and electricity price fluctuations of external power grids, dynamically adjust the input state, and use the deep reinforcement learning model to generate an energy allocation strategy that adapts to the current situation in real time. At the same time, the parameters of the deep reinforcement learning model are adjusted according to the actual situation to obtain the optimal energy allocation strategy.

[0033] Optionally, it is characterized in that, in combination with the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, time series forecasting technology is used to predict energy consumption demand and power generation potential, and an energy supply and demand balance plan is generated, including:

[0034] Combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, multi-source data fusion is performed to obtain a multi-dimensional data set;

[0035] Performing data preprocessing and feature engineering on the multi-dimensional data set to obtain a preprocessed multi-dimensional data set;

[0036] Based on the preprocessed multidimensional data set, a time series prediction model is constructed, and a long short-term memory network deep learning model is adopted to capture the long-term dependency and periodicity in the time series data to obtain a trained time series prediction model;

[0037] The trained time series prediction model is used to perform multi-step predictions on energy consumption demand and power generation potential within multiple preset time windows, and a rolling update mechanism is used to add actual observations to the training data set, and the time series prediction model is retrained and multi-step predictions are performed again to obtain updated prediction results;

[0038] Based on the updated forecast results, the electricity demand of the medical unit, the power generation capacity of renewable energy and the power supply price of the external power grid are comprehensively calculated to generate a detailed energy supply and demand balance plan.

[0039] Optionally, based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted, the operation data of the distributed energy system is collected and analyzed at preset time nodes, the failure risk of the operation data is predicted by a machine learning model, an evaluation result is generated, and the optimal energy allocation strategy and multi-objective optimization algorithm are iteratively optimized according to the evaluation result to form an adaptive and self-learning energy closed-loop management system, including:

[0040] Using the optimal charging and discharging plan, determine the operating parameter range of the distributed energy system, wherein the operating parameters include: the charging and discharging state of the energy storage system, the energy consumption of each medical unit, and the power generation of renewable energy;

[0041] At a preset time node, collect and analyze the operating data of the distributed energy system within the operating parameter range, the operating data including: energy efficiency ratio, carbon emissions, system reliability and response speed;

[0042] Based on the operation data, predict the failure risk of the distributed energy system through a machine learning model to generate a failure risk prediction report;

[0043] Using the fault risk prediction report and combining it with key indicators of the operating data, a performance evaluation result is generated;

[0044] Using the performance evaluation results, iteratively optimizing the optimal energy allocation strategy and the multi-objective optimization algorithm to obtain optimized optimal energy allocation strategy and the multi-objective optimization algorithm;

[0045] The optimized optimal energy allocation strategy and multi-objective optimization algorithm are applied again to the distributed energy system to form an adaptive and self-learning energy closed-loop management system.

[0046] In a second aspect, the embodiment of the present application provides a distributed energy optimization management system for rapidly deploying a mobile cabin hospital, including:

[0047] A building module for constructing a multi-layer virtual energy network model of a mobile shelter hospital;

[0048] An adjustment module, for dynamically adjusting energy allocation based on the multi-layer virtual energy network model and applying a deep reinforcement learning algorithm to generate an optimal energy allocation strategy;

[0049] A forecasting module, for forecasting energy demand and power generation potential using time series forecasting technology in combination with the optimal energy allocation strategy, real-time weather forecast information, and historical energy consumption data and medical activity patterns, and generating an energy supply and demand balance plan;

[0050] A scheduling module, which is used to intelligently schedule the charging and discharging process of the energy storage system based on the energy supply and demand balance plan and introduce a multi-objective optimization algorithm to obtain an optimal charging and discharging plan;

[0051] The generation module, based on the optimal charging and discharging plan, adopts a performance evaluation mechanism based on big data analysis, collects and analyzes the operating data of the distributed energy system at preset time nodes, predicts the fault risk of the operating data through a machine learning model, generates evaluation results, and iteratively optimizes the optimal energy allocation strategy and multi-objective optimization algorithm based on the evaluation results to form an adaptive and self-learning energy closed-loop management system.

[0052] In a third aspect, an embodiment of the present invention provides a computing device, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a distributed energy optimization management method for rapidly deploying a mobile cabin hospital as described in any one of the first aspects.

[0053] In a fourth aspect, an embodiment of the present invention provides a computer storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement a distributed energy optimization management method for rapidly deploying a mobile cabin hospital as described in any one of the first aspects.

[0054] In an embodiment of the present invention, a multi-layer virtual energy network model of a mobile cabin hospital is constructed; based on the multi-layer virtual energy network model, a deep reinforcement learning algorithm is applied to dynamically adjust energy distribution to generate an optimal energy distribution strategy; combined with the optimal energy distribution strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, time series prediction technology is used to predict energy consumption demand and power generation potential, and an energy supply and demand balance plan is generated; based on the energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system to obtain the optimal charging and discharging plan; based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted, and the operating data of the distributed energy system is collected and analyzed at preset time nodes, and the operating data is predicted for fault risk through a machine learning model to generate an evaluation result, and the optimal energy distribution strategy and multi-objective optimization algorithm are iteratively optimized according to the evaluation results to form an adaptive and self-learning energy closed-loop management system. The technical solution provided by the present invention realizes efficient utilization, flexible scheduling and stable supply of energy by constructing a multi-layer virtual energy network model and combining deep reinforcement learning algorithms, time series prediction technologies and multi-objective optimization algorithms. Specifically, this method can dynamically adjust energy allocation strategies, accurately predict future energy consumption demand and power generation potential, intelligently schedule the charging and discharging process of the energy storage system, and through performance evaluation mechanisms based on big data analysis and machine learning models, monitor the system status in real time and predict potential failures, thereby improving the overall energy efficiency, adaptability and reliability of the system and reducing maintenance costs and workload.

[0055] These and other aspects of the present invention will become more apparent from the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 A flow chart of a distributed energy optimization management method for rapidly deploying a mobile cabin hospital provided by an embodiment of the present invention;

[0058] Figure 2 A structural schematic diagram of a distributed energy optimization management system for rapid deployment of mobile cabin hospitals provided by an embodiment of the present invention;

[0059] Figure 3 A schematic diagram of the structure of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0061] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0063] In the prior art, the traditional centralized energy supply method cannot dynamically adjust the energy allocation strategy according to the real-time electricity demand of medical equipment and the power generation of renewable energy, resulting in energy waste or supply shortage from time to time. The existing energy management system lacks intelligent scheduling capabilities and cannot fully utilize the advantages of renewable energy and energy storage systems. Especially in the face of complex and changeable power consumption environments, the system's adaptability and robustness are poor. In addition, due to the lack of effective fault prediction and self-healing functions, the existing energy management system is prone to failure during operation and requires frequent manual intervention and maintenance, which increases operation and maintenance costs and workloads. Based on this, the present invention provides a distributed energy optimization management method for rapid deployment of mobile cabin hospitals, such as Figure 1 , the specific steps are as follows:

[0064] Step 101: Construct a multi-layer virtual energy network model of a mobile cabin hospital;

[0065] In this step, a multi-level virtual energy network model is constructed, which comprehensively considers the energy flow and mutual relationship between each medical unit, energy storage system, renewable energy power generation unit and external grid access point in the mobile cabin hospital. By simulating the operating status and mutual influence of different equipment and systems, the infrastructure is provided for subsequent energy optimization management. This model not only includes physical connection relationships, but also covers information such as energy demand forecast and power generation capacity evaluation of each node to ensure that the actual operation is fully reflected.

[0066] Step 102: Based on the multi-layer virtual energy network model, a deep reinforcement learning algorithm is applied to dynamically adjust energy allocation to generate an optimal energy allocation strategy;

[0067] In this step, a deep reinforcement learning algorithm (such as Ql earning or DQN) is used to dynamically adjust the energy allocation strategy based on real-time data provided by the multi-layer virtual energy network model. The algorithm learns through trial and error and selects the optimal action, i.e., the energy allocation scheme, under different system states to maximize the long-term cumulative rewards. These rewards can be based on a comprehensive evaluation of multiple objectives such as energy consumption cost, renewable energy utilization rate, and power supply reliability of key medical facilities. Ultimately, the optimal energy allocation strategy that can adapt to various working conditions is generated.

[0068] Step 103: combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, using time series forecasting technology to forecast energy consumption demand and power generation potential, and generating an energy supply and demand balance plan;

[0069] In this step, time series forecasting technology is used to predict future energy demand and power generation potential by combining the best energy allocation strategy, real-time weather forecast information, historical energy consumption data, and medical activity patterns. Specifically, the time series model analyzes past data patterns and combines current external conditions (such as weather changes) and internal needs (such as changes in the power priority of medical equipment) to accurately predict energy demand and renewable energy generation in the future. The result of this process is to generate a detailed energy supply and demand balance plan to guide subsequent energy scheduling decisions.

[0070] Step 104: Based on the energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system to obtain an optimal charging and discharging plan;

[0071] In this step, based on the generated energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system. The multi-objective optimization algorithm aims to simultaneously consider multiple optimization objectives, such as minimizing the operating cost of the energy storage system, maximizing the utilization rate of renewable energy, and ensuring the reliability of power supply to critical medical facilities. By solving the multi-objective optimization problem, a Pareto frontier consisting of a set of non-inferior solutions is found, from which the best charging and discharging plan that best suits the current operating conditions is selected to ensure that the energy storage system can achieve efficient and smooth operation under different operating conditions.

[0072] Step 105: Based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted to collect and analyze the operation data of the distributed energy system at preset time nodes, and a fault risk prediction is performed on the operation data through a machine learning model to generate an evaluation result, and the optimal energy allocation strategy and multi-objective optimization algorithm are iteratively optimized according to the evaluation result to form an adaptive and self-learning energy closed-loop management system;

[0073] In this step, based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted to collect and analyze the operating data of the distributed energy system at the preset time nodes. The machine learning model is used to predict the fault risk of these data and generate evaluation results. Based on the evaluation results, the optimal energy allocation strategy and multi-objective optimization algorithm are iteratively optimized to ensure that the system can continuously improve and adapt to new operating conditions. This adaptive and self-learning energy closed-loop management system makes the entire energy management system more intelligent and efficient.

[0074] Based on this, the present invention provides a specific embodiment, wherein step 104, based on the energy supply and demand balance plan, introduces a multi-objective optimization algorithm to intelligently schedule the charging and discharging process of the energy storage system to obtain an optimal charging and discharging plan, specifically comprising the following steps:

[0075] Step 201: constructing a dynamic mathematical model of the energy storage system by using the current state information of the energy storage system and combining the energy supply and demand balance plan;

[0076] In this step, based on the current status information of the energy storage system (such as remaining capacity, maximum charge and discharge power limit, charge and discharge efficiency, etc.), combined with the future energy consumption demand and power generation potential forecast provided by the energy supply and demand balance plan, a dynamic mathematical model is constructed. The model can reflect the operating status of the energy storage system at different time points in real time and provide the necessary input data for the subsequent optimization algorithm to ensure that the model can accurately simulate the actual behavior of the energy storage system.

[0077] Step 202: Design a multi-objective optimization function by using the dynamic mathematical model and combining the power demand priority of medical equipment, the real-time power generation of renewable energy, and the price fluctuation of the external power grid.

[0078] In this step, a multi-objective optimization function is designed by using the dynamic mathematical model that has been constructed, combining the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid. This function comprehensively considers multiple optimization goals, such as minimizing the operating cost of the energy storage system, maximizing the utilization rate of renewable energy, and ensuring the reliability of power supply for key medical facilities. In this way, it is ensured that the optimization process can achieve efficient utilization and smooth supply of energy under various constraints.

[0079] Step 203: applying an evolutionary computing method in combination with a fuzzy logic controller to solve the multi-objective optimization function and obtain a Pareto frontier consisting of a set of non-inferior solutions;

[0080] In this step, the multi-objective optimization function is solved by applying evolutionary computing methods (such as genetic algorithms or differential evolution algorithms) combined with fuzzy logic controllers. Evolutionary computing methods are used to explore the solution space and find a series of non-inferior solutions, namely the Pareto frontier. The fuzzy logic controller dynamically adjusts the parameters in the optimization process according to the current system state and environmental conditions to improve the efficiency of the solution and the quality of the solution. The final set of non-inferior solutions constitutes the Pareto frontier, and each solution represents a possible optimal charging and discharging plan.

[0081] Step 204: using a multi-criteria decision support system to integrate expert knowledge and user preferences, and select an optimal charging and discharging plan from the Pareto frontier surface;

[0082] In this step, a multi-criteria decision support system is used to integrate expert knowledge and user preferences to select the optimal charging and discharging plan from the Pareto front. The multi-criteria decision support system evaluates the performance of each non-inferior solution on different objectives through methods such as the analytic hierarchy process (AHP) or the technology selection and prioritization method (TOPS IS). Combining expert experience and the actual needs of users, an optimal charging and discharging plan that meets both performance requirements and operator preferences is finally selected to ensure its feasibility and effectiveness in practical applications.

[0083] Based on this, the present invention provides a specific embodiment, wherein the step 202 uses the dynamic mathematical model, combines the power demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, and designs a multi-objective optimization function, which specifically includes the following steps:

[0084] Step 301: using the dynamic mathematical model, combining the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, to define the target variables of the multi-objective optimization function;

[0085] In this step, based on the dynamic mathematical model of the energy storage system, combined with the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuations of the external power grid, the specific target variables of the multi-objective optimization function are determined. These target variables include but are not limited to the operating cost of the energy storage system, the utilization rate of renewable energy, the power supply reliability of key medical facilities, and carbon emissions. By clarifying these target variables, the foundation is laid for the subsequent design of specific optimization functions.

[0086] Step 302: designing a specific expression of a multi-objective optimization function according to the target variables, wherein the expression includes each target variable and a relative importance weight of each target variable;

[0087] In this step, a specific mathematical expression of the multi-objective optimization function is designed based on the defined target variables. This expression not only contains the various target variables, but also introduces the relative importance weight of each target variable. The weight coefficient reflects the priority and importance of different targets in the overall optimization process, ensuring that the optimization result can find the best balance between multiple targets. For example, the power supply reliability of critical medical facilities may be given a higher weight to ensure its stable operation.

[0088] Step 303: calibrating the parameters in the multi-objective optimization function by using the specific expression combined with historical data and real-time data to obtain a calibrated multi-objective optimization function;

[0089] In this step, specific expressions are used in combination with historical data and real-time data to calibrate the parameters in the multi-objective optimization function. Historical data provides the patterns and trends of past system operations, while real-time data reflects the current system status and environmental changes. Through comprehensive analysis of these data, the parameters in the optimization function are adjusted to more accurately reflect the actual operating conditions. The calibrated multi-objective optimization function can better adapt to different working conditions and improve the reliability and accuracy of the optimization results.

[0090] Step 304: using the calibrated multi-objective optimization function, in combination with external environment changes and internal demand changes, dynamically adjusting the weight coefficients in the multi-objective optimization function to generate a multi-objective optimization function that adapts to different operating conditions;

[0091] In this step, the weight coefficients in the multi-objective optimization function are dynamically adjusted using the calibrated multi-objective optimization function in combination with external environmental changes (such as weather conditions, fluctuations in external grid electricity prices) and internal demand changes (such as changes in the priority of electricity demand for medical equipment). This dynamic adjustment mechanism enables the optimization function to flexibly respond to different operating conditions, ensuring that the optimal energy management strategy can be found under any circumstances. The resulting multi-objective optimization function not only adapts to current operating conditions, but also remains efficient and stable in future environments.

[0092] Based on this, the present invention provides a specific embodiment, wherein step 304 uses the calibrated multi-objective optimization function, combines external environment changes and internal demand changes, dynamically adjusts the weight coefficients in the multi-objective optimization function, and generates a multi-objective optimization function that adapts to different operating conditions, specifically including the following steps:

[0093] Step 401: using the calibrated multi-objective optimization function, combined with external environment changes and internal demand changes, analyzing the importance change trend of each target variable to obtain importance change trend analysis results;

[0094] In this step, the calibrated multi-objective optimization function is used to analyze the importance change trend of each target variable in combination with external environmental changes (such as weather conditions, fluctuations in external grid electricity prices) and internal demand changes (such as changes in the electricity demand priority of medical equipment). Through comparative analysis of historical data and real-time data, the change pattern of each target variable in different time periods is identified, and finally the importance change trend analysis results are obtained.

[0095] Step 402: designing a weight adjustment rule for the importance change trend analysis result, wherein the weight adjustment rule includes the importance difference of each target variable in different time periods;

[0096] In this step, based on the analysis results of importance change trends, weight adjustment rules are designed. These rules take into account the differences in importance of each target variable in different time periods, ensuring that the optimization function can flexibly adapt to various operating conditions. For example, when renewable energy generation is high during the day, the weight of renewable energy utilization may be increased; while the importance of energy storage systems may be higher at night. In this way, a set of rules for dynamically adjusting weight coefficients is developed.

[0097] Step 403: dynamically adjusting the weight coefficients in the multi-objective optimization function using the weight adjustment rule to obtain dynamic weight coefficients;

[0098] In this step, the designed weight adjustment rules are applied to dynamically adjust the weight coefficients in the multi-objective optimization function. According to the current time period and actual operating conditions, the weight coefficients of each target variable are automatically updated to reflect the latest importance change trend. This process ensures that the optimization function can adapt to changes in the external environment and internal needs in real time, generate more accurate and effective energy management strategies, and finally obtain dynamic weight coefficients.

[0099] Step 404: applying the dynamic weight coefficient to the multi-objective optimization function to generate a multi-objective optimization function that is adaptable to different operating conditions;

[0100] In this step, the dynamic weight coefficient is applied to the multi-objective optimization function to generate a multi-objective optimization function that adapts to different operating conditions. By introducing the dynamic weight coefficient, the optimization function can find the optimal solution under different working conditions, ensuring that the charging and discharging process of the energy storage system and the overall energy management strategy are always in the best state. This adaptive optimization mechanism improves the flexibility and reliability of the system and ensures that the energy supply of the mobile cabin hospital is both efficient and stable.

[0101] Since the existing methods usually adopt static weight coefficients, it is impossible to dynamically adjust the importance of each target variable according to changes in the external environment, internal demand and time period. This leads to the fact that the energy management strategy of the system is not flexible enough under different working conditions and cannot effectively cope with the fluctuation of medical equipment electricity demand and the uncertainty of renewable energy generation, thus affecting the overall performance and stability of the system. Therefore, the present invention introduces a weight adjustment rule for dynamically adjusting the weight coefficient. The expression of the weight adjustment rule for dynamically adjusting the weight coefficient is:

[0102]

[0103] Among them, w i (t) represents the weight coefficient of the i-th target variable at time t; represents the initial weight coefficient of the i-th objective variable, which is determined by the calibrated multi-objective optimization function; α i represents the weight adjustment coefficient of the i-th target variable over time, reflecting the difference in importance of each target variable in different time periods; Δt represents the time interval, which is used to represent the time difference from the last adjustment to the current time; β i It represents the weight adjustment coefficient of the i-th target variable as the external environment changes, reflecting the impact of external environment changes on the importance of the target variable; ΔE i (t) represents the external environmental change of the i-th target variable at time t, which can be obtained by analyzing external environmental data such as real-time weather forecast information and external power grid electricity price fluctuations; γ irepresents the weight adjustment coefficient of the i-th target variable as the internal demand changes, reflecting the impact of the internal demand change on the importance of the target variable; ΔD i (t) represents the change in internal demand of the i-th target variable at time t, which can be obtained by analyzing the internal demand data such as the change in the priority of electricity demand of medical equipment; δ i represents the weight adjustment coefficient of the i-th target variable as the system performance changes, reflecting the impact of system performance changes on the importance of the target variable; F i (t) represents the change in system performance of the i-th target variable at time t, which is obtained by analyzing the operating data of the distributed energy system; η i represents the weight adjustment coefficient of the i-th target variable over time, reflecting the importance of different time periods (such as daytime and nighttime) on the target variable; T(t) represents the time period factor, such as the number of hours in a day or the number of days in a week, which is used to reflect the weight changes in different time periods; θ i The weight adjustment coefficient of the i-th target variable with the trend of importance change reflects the degree of weight adjustment according to the analysis results of importance change trend; φ i (t) represents the importance change trend value of the i-th target variable at time t, which is obtained by analyzing historical data and real-time data;

[0104] Through the above complex formula, the present invention can more finely and dynamically adjust the weight coefficients in the multi-objective optimization function, so that the system can better adapt to changes in the external environment, internal demand and time cycle. This not only improves energy utilization efficiency and ensures stable power supply for key medical facilities, but also enhances the flexibility and reliability of the system, reduces maintenance costs and workload, and thus significantly improves the overall performance of the mobile cabin hospital energy management system.

[0105] Based on this, the present invention provides a specific embodiment, wherein step 102, based on the multi-layer virtual energy network model, applies a deep reinforcement learning algorithm to dynamically adjust energy allocation to generate an optimal energy allocation strategy, specifically comprising the following steps:

[0106] Step 501: using the multi-layer virtual energy network model, integrating the energy consumption forecast information of the mobile cabin hospital, the real-time power generation of renewable energy, the price fluctuation of the external power grid, and the urgency of the mobile cabin hospital, and calculating the uncertainty of weather changes, simulating the possible impact of weather changes through a probability distribution model, and obtaining an input state;

[0107] In this step, based on the multi-layer virtual energy network model, the energy consumption forecast information of the mobile cabin hospital, the real-time power generation of renewable energy, the electricity price fluctuations of the external power grid and the urgency of the mobile cabin hospital are integrated. At the same time, the uncertainty of weather changes is calculated, and the possible impact of weather changes is simulated through the probability distribution model. These data and simulation results together constitute the input state for subsequent optimization.

[0108] Step 502: Based on the input state, the priority of electricity demand of the mobile cabin hospital, the availability of renewable energy, the price signal of the external power grid and the uncertainty of weather changes are set as a multidimensional reward function, wherein the stable power supply and cost saving of the emergency medical unit are the main optimization goals, and a multidimensional reward function is obtained;

[0109] In this step, based on the input state, the priority of electricity demand of the mobile cabin hospital, the availability of renewable energy, the price signal of the external power grid and the uncertainty of weather changes are set as the multidimensional reward function. The stable power supply and cost savings of the emergency medical unit are set as the main optimization goals. In this way, a multidimensional reward function that comprehensively considers multiple factors is constructed.

[0110] Step 503: Processing the input state and the multidimensional reward function using a deep reinforcement learning model based on an Actor-Critic architecture to obtain an initial energy allocation strategy;

[0111] In this step, a deep reinforcement learning model based on the Actor-Critic architecture is used to process the input state and multidimensional reward function. The Actor is responsible for selecting actions (i.e., energy allocation strategy), while Critic evaluates the value of the selected action. Through continuous trial and error learning, an initial energy allocation strategy that can adapt to the current input state is finally obtained.

[0112] Step 504: Based on the initial energy allocation strategy, during the training process, the experience replay mechanism is used to randomly extract historical data for replay learning, thereby improving the generalization ability and stability of the model and obtaining an enhanced energy allocation strategy;

[0113] In this step, based on the initial energy allocation strategy, an experience replay mechanism is introduced in the training process. By randomly extracting historical data for replay learning, the model can be trained in more diverse scenarios, thereby improving its generalization ability and stability, and finally obtaining an enhanced energy allocation strategy through reinforcement learning.

[0114] Step 505: Based on the enhanced energy allocation strategy, a specific emergency response strategy is designed for the preset abnormal situation to obtain an energy allocation strategy that adapts to the abnormal situation;

[0115] In this step, based on the enhanced energy allocation strategy, specific emergency response strategies are designed for preset abnormal situations (such as equipment failure or extreme weather). These strategies ensure that the system can still maintain stable operation under abnormal circumstances, and ultimately form an energy allocation strategy that adapts to abnormal situations.

[0116] Step 506: Based on the energy allocation strategy that adapts to abnormal situations, during actual operation, the latest mobile cabin hospital energy consumption data, real-time power generation of renewable energy and electricity price fluctuations of the external power grid are collected by online learning, the input state is dynamically adjusted, and the energy allocation strategy that adapts to the current situation is generated in real time by using the deep reinforcement learning model. At the same time, the parameters of the deep reinforcement learning model are adjusted according to the actual situation to obtain the optimal energy allocation strategy.

[0117] In this step, based on the energy allocation strategy that adapts to abnormal situations, an online learning method is adopted during actual operation. By continuously collecting the latest energy consumption data, renewable energy generation conditions and external grid electricity price fluctuations, the input state is dynamically adjusted, and a deep reinforcement learning model is used to generate an energy allocation strategy that adapts to the current situation in real time. The model parameters are adjusted according to the actual situation to finally obtain the optimal energy allocation strategy.

[0118] Based on this, the present invention provides a specific embodiment, wherein step 103 specifically includes the following steps:

[0119] Step 601: Combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, multi-source data fusion is performed to obtain a multi-dimensional data set;

[0120] In this step, multi-source data fusion is performed by combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data, and medical activity patterns. These data sources include but are not limited to the charging and discharging status of the energy storage system, the energy consumption of each medical unit, and the power generation of renewable energy. By integrating data from these different sources, a comprehensive data set with multiple dimensions such as time, location, and equipment status is formed, providing a comprehensive foundation for subsequent analysis.

[0121] Step 602: performing data preprocessing and feature engineering on the multi-dimensional data set to obtain a preprocessed multi-dimensional data set;

[0122] In this step, data preprocessing and feature engineering are performed on the multi-dimensional data set. Data preprocessing includes operations such as removing outliers, filling missing values, standardization or normalization to ensure data quality. Feature engineering involves extracting useful features from the original data, such as calculating moving averages, generating lag features, extracting periodic components, etc. After these processes, a structured and easy-to-model preprocessed multi-dimensional data set is obtained.

[0123] Step 603: Based on the preprocessed multidimensional data set, a time series prediction model is constructed, and a long short-term memory network deep learning model is adopted to capture the long-term dependency and periodicity in the time series data to obtain a trained time series prediction model;

[0124] In this step, a time series prediction model is constructed based on the preprocessed multidimensional data set. The long short-term memory network (LSTM) is selected as the deep learning model because it is good at capturing long-term dependencies and periodic laws in time series data. By training this model, it can accurately predict future energy consumption demand and power generation potential, and finally obtain a trained time series prediction model.

[0125] Step 604: using the trained time series prediction model to perform multi-step predictions on energy consumption demand and power generation potential within multiple preset time windows, and using a rolling update mechanism to add actual observations to the training data set, retraining the time series prediction model and performing multi-step predictions again to obtain updated prediction results;

[0126] In this step, the trained time series prediction model is used to make multi-step predictions of energy consumption demand and power generation potential within multiple preset time windows. In order to improve the accuracy of the prediction, a rolling update mechanism is adopted to add the actual observations to the training data set, retrain the time series prediction model, and perform multi-step predictions again. This process ensures that the model always makes predictions based on the latest data, thereby obtaining more accurate and reliable updated prediction results.

[0127] Step 605: Based on the updated forecast results, the electricity demand of the medical unit, the power generation capacity of the renewable energy and the power supply price of the external power grid are comprehensively calculated to generate a detailed energy supply and demand balance plan;

[0128] In this step, based on the updated forecast results, the electricity demand of the medical unit, the power generation capacity of renewable energy and the power supply price of the external power grid are comprehensively considered. By calculating the balance relationship between these factors in detail, a detailed energy supply and demand balance plan is generated. This plan not only guides the charging and discharging strategy of the energy storage system, but also optimizes the energy management of the entire mobile cabin hospital, ensuring stable power supply and effective cost control for key medical facilities.

[0129] Based on this, the present invention provides a specific embodiment, in which step 105, based on the optimal charging and discharging plan, adopts a performance evaluation mechanism based on big data analysis, collects and analyzes the operating data of the distributed energy system at preset time nodes, predicts the fault risk of the operating data through a machine learning model, generates an evaluation result, and iteratively optimizes the optimal energy allocation strategy and the multi-objective optimization algorithm according to the evaluation result to form an adaptive and self-learning energy closed-loop management system, which specifically includes the following steps:

[0130] Step 701: using the optimal charging and discharging plan, determining the operating parameter range of the distributed energy system, wherein the operating parameters include: the charging and discharging state of the energy storage system, the energy consumption of each medical unit, and the power generation of renewable energy;

[0131] In this step, the operating parameter range of the distributed energy system is determined based on the optimal charging and discharging plan. These parameters include the charging and discharging status of the energy storage system, the energy consumption of each medical unit, and the power generation of renewable energy. By clarifying the range of these operating parameters, specific guidance is provided for subsequent data collection and analysis to ensure the accuracy and representativeness of the data.

[0132] Step 702: at a preset time node, collecting and analyzing the operation data of the distributed energy system within the operation parameter range, the operation data including: energy efficiency ratio, carbon emissions, system reliability and response speed;

[0133] In this step, the operating data of the distributed energy system within the operating parameter range is collected and analyzed at the preset time nodes. These operating data include key indicators such as energy efficiency ratio, carbon emissions, system reliability and response speed. By regularly collecting and analyzing these data, we can fully understand the actual operating status of the system and provide a basis for subsequent performance evaluation and optimization.

[0134] Step 703: Based on the operation data, predict the failure risk of the distributed energy system through a machine learning model to generate a failure risk prediction report;

[0135] In this step, the failure risk of the distributed energy system is predicted through the machine learning model based on the collected operating data. The machine learning model can identify various factors that may cause system failures and predict possible risks in the future, and finally generate a failure risk prediction report to help managers take preventive measures in advance and reduce the potential failure rate.

[0136] Step 704: Generate a performance evaluation result by using the fault risk prediction report and combining it with the key indicators of the operation data;

[0137] In this step, the fault risk prediction report is used in combination with key indicators in the operating data, such as energy efficiency ratio, carbon emissions, system reliability and response speed, to generate performance evaluation results. The performance evaluation results not only reflect the current operating status of the system, but also point out existing problems and improvement directions, providing clear goals and basis for subsequent optimization.

[0138] Step 705: using the performance evaluation result, iteratively optimizing the optimal energy allocation strategy and the multi-objective optimization algorithm to obtain an optimized optimal energy allocation strategy and the multi-objective optimization algorithm;

[0139] In this step, the optimal energy allocation strategy and multi-objective optimization algorithm are iteratively optimized using the performance evaluation results. By adjusting the parameters and weights in the optimization algorithm, the new strategy and algorithm can meet the performance requirements while further improving the overall efficiency and stability of the system. Finally, the optimized optimal energy allocation strategy and multi-objective optimization algorithm are obtained to ensure that the system can continue to improve and adapt to new operating conditions.

[0140] Step 706: Apply the optimized optimal energy allocation strategy and multi-objective optimization algorithm to the distributed energy system again to form an adaptive and self-learning energy closed-loop management system.

[0141] In this step, the optimized optimal energy allocation strategy and multi-objective optimization algorithm are applied again to the distributed energy system. By continuously looping this process, the system can automatically adjust and optimize its own operation mode according to the latest operation data and performance evaluation results, forming an adaptive and self-learning energy closed-loop management system. This mechanism ensures that the system can always maintain efficient and stable operation under different working conditions.

[0142] Figure 2 A structural diagram of a distributed energy optimization management system for rapidly deploying a mobile shelter hospital is provided for the present application embodiment. Figure 2 As shown, the system includes:

[0143] Building module 21, used to build a multi-layer virtual energy network model of a mobile cabin hospital;

[0144] An adjustment module 22, for dynamically adjusting energy allocation based on the multi-layer virtual energy network model by applying a deep reinforcement learning algorithm to generate an optimal energy allocation strategy;

[0145] A prediction module 23 is used to combine the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, use time series prediction technology to predict energy consumption demand and power generation potential, and generate an energy supply and demand balance plan;

[0146] A scheduling module 24 is used to introduce a multi-objective optimization algorithm to intelligently schedule the charging and discharging process of the energy storage system based on the energy supply and demand balance plan to obtain an optimal charging and discharging plan;

[0147] The generation module 25, based on the optimal charging and discharging plan, adopts a performance evaluation mechanism based on big data analysis, collects and analyzes the operating data of the distributed energy system at preset time nodes, predicts the fault risk of the operating data through a machine learning model, generates an evaluation result, and iteratively optimizes the optimal energy allocation strategy and multi-objective optimization algorithm based on the evaluation result to form an adaptive and self-learning energy closed-loop management system.

[0148] Figure 2 The distributed energy optimization management system for rapidly deploying mobile shelter hospitals can be implemented Figure 1 The implementation principle and technical effect of the xx method described in the embodiment shown will not be repeated. For a distributed energy optimization management system for rapid deployment of mobile cabin hospitals in the above embodiment, the specific way in which each module and unit performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0149] Figure 2 A distributed energy optimization management system for rapidly deploying a mobile cabin hospital in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0150] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0151] The processing component 32 is used to construct a multi-layer virtual energy network model of the mobile cabin hospital;

[0152] Based on the multi-layer virtual energy network model, a deep reinforcement learning algorithm is applied to dynamically adjust energy distribution to generate an optimal energy distribution strategy; in combination with the optimal energy distribution strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, time series prediction technology is used to predict energy consumption demand and power generation potential to generate an energy supply and demand balance plan; based on the energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system to obtain an optimal charging and discharging plan; based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted, and the operating data of the distributed energy system is collected and analyzed at preset time nodes. The fault risk of the operating data is predicted through a machine learning model to generate an evaluation result, and the optimal energy distribution strategy and the multi-objective optimization algorithm are iteratively optimized according to the evaluation results to form an adaptive and self-learning energy closed-loop management system.

[0153] The processing component 32 includes one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be implemented by one or more application-specific integrated circuits (AICs), digital signal processors (DPs), digital signal processing devices (DPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0154] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (RAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0155] Computing devices also include other components, such as input / output interfaces, display components, and communication components.

[0156] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above-mentioned peripheral interface module can be an output device or an input device.

[0157] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0158] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0159] The embodiment of the present invention further provides a computer storage medium storing a computer program, which can achieve the above-mentioned Figure 1 The illustrated embodiment provides a distributed energy optimization management method and system for rapidly deploying mobile cabin hospitals.

[0160] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0161] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0162] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0163] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A distributed energy optimization management method for rapid deployment of mobile cabin hospitals, characterized in that: include: Construct a multi-layer virtual energy network model for mobile cabin hospitals; Based on the multi-layer virtual energy network model, a deep reinforcement learning algorithm is applied to dynamically adjust energy distribution and generate an optimal energy distribution strategy; Combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, using time series forecasting technology to predict energy consumption demand and power generation potential, and generate an energy supply and demand balance plan; Based on the energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system to obtain the optimal charging and discharging plan; Based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted, the operating data of the distributed energy system is collected and analyzed at preset time nodes, the fault risk of the operating data is predicted through a machine learning model, and an evaluation result is generated. The optimal energy allocation strategy and multi-objective optimization algorithm are iteratively optimized according to the evaluation results to form an adaptive and self-learning energy closed-loop management system.

2. The method according to claim 1, characterized in that: Based on the energy supply and demand balance plan, a multi-objective optimization algorithm is introduced to intelligently schedule the charging and discharging process of the energy storage system to obtain the optimal charging and discharging plan, including: Using the current state information of the energy storage system and combining it with the energy supply and demand balance plan, a dynamic mathematical model of the energy storage system is constructed; Using the dynamic mathematical model, combined with the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, a multi-objective optimization function is designed; Applying the evolutionary computing method in combination with the fuzzy logic controller, the multi-objective optimization function is solved to obtain a Pareto frontier consisting of a set of non-inferior solutions; A multi-criteria decision support system is used to integrate expert knowledge and user preferences to select the optimal charging and discharging plan from the Pareto frontier.

3. The method according to claim 2, characterized in that Using the dynamic mathematical model, combined with the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, a multi-objective optimization function is designed, including: Using the dynamic mathematical model, combined with the electricity demand priority of medical equipment, the real-time power generation of renewable energy, and the electricity price fluctuation of the external power grid, the target variables of the multi-objective optimization function are defined; Designing a specific expression of a multi-objective optimization function according to the target variables, wherein the expression includes each target variable and a relative importance weight of each target variable; Using the specific expression in combination with historical data and real-time data, calibrate the parameters in the multi-objective optimization function to obtain a calibrated multi-objective optimization function; By utilizing the calibrated multi-objective optimization function and combining the external environment changes and the internal demand changes, the weight coefficients in the multi-objective optimization function are dynamically adjusted to generate a multi-objective optimization function that adapts to different operating conditions.

4. The method according to claim 3, characterized in that Using the calibrated multi-objective optimization function, combined with changes in the external environment and internal demand, dynamically adjust the weight coefficients in the multi-objective optimization function to generate a multi-objective optimization function that adapts to different operating conditions, including: Using the calibrated multi-objective optimization function, combined with external environment changes and internal demand changes, the importance change trend of each target variable is analyzed to obtain the importance change trend analysis results; Designing a weight adjustment rule based on the importance change trend analysis result, wherein the weight adjustment rule includes the importance difference of each target variable in different time periods; Dynamically adjusting the weight coefficients in the multi-objective optimization function using the weight adjustment rule to obtain dynamic weight coefficients; The dynamic weight coefficient is applied to the multi-objective optimization function to generate a multi-objective optimization function that is adaptable to different operating conditions.

5. The method according to claim 1, characterized in that Based on the multi-layer virtual energy network model, a deep reinforcement learning algorithm is applied to dynamically adjust energy distribution and generate an optimal energy distribution strategy, including: The multi-layer virtual energy network model is used to integrate the energy consumption forecast information of the mobile cabin hospital, the real-time power generation of renewable energy, the price fluctuation of the external power grid, and the urgency of the mobile cabin hospital. At the same time, the uncertainty of weather changes is calculated, and the possible impact of weather changes is simulated through a probability distribution model to obtain the input state; Based on the input state, the priority of electricity demand of the mobile cabin hospital, the availability of renewable energy, the price signal of the external power grid and the uncertainty of weather changes are set as a multidimensional reward function, where the stable power supply and cost saving of the emergency medical unit are the main optimization goals, and a multidimensional reward function is obtained; Processing the input state and the multidimensional reward function using a deep reinforcement learning model based on an Actor-Criti c architecture to obtain an initial energy allocation strategy; Based on the initial energy allocation strategy, during the training process, the experience replay mechanism is used to randomly extract historical data for replay learning, thereby improving the generalization ability and stability of the model and obtaining an enhanced energy allocation strategy; Based on the enhanced energy allocation strategy, a specific emergency response strategy is designed for a preset abnormal situation to obtain an energy allocation strategy that adapts to the abnormal situation; Based on the energy allocation strategy that adapts to abnormal situations, during actual operation, online learning is used to collect the latest energy consumption data of mobile cabin hospitals, real-time power generation of renewable energy and electricity price fluctuations of external power grids, dynamically adjust the input state, and use the deep reinforcement learning model to generate an energy allocation strategy that adapts to the current situation in real time. At the same time, the parameters of the deep reinforcement learning model are adjusted according to the actual situation to obtain the optimal energy allocation strategy.

6. The method according to claim 1, characterized in that Combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, time series forecasting technology is used to predict energy consumption demand and power generation potential, and an energy supply and demand balance plan is generated, including: Combining the optimal energy allocation strategy, real-time weather forecast information, historical energy consumption data and medical activity patterns, multi-source data fusion is performed to obtain a multi-dimensional data set; Performing data preprocessing and feature engineering on the multi-dimensional data set to obtain a preprocessed multi-dimensional data set; Based on the preprocessed multidimensional data set, a time series prediction model is constructed, and a long short-term memory network deep learning model is adopted to capture the long-term dependency and periodicity in the time series data to obtain a trained time series prediction model; The trained time series prediction model is used to perform multi-step predictions on energy consumption demand and power generation potential within multiple preset time windows, and a rolling update mechanism is used to add actual observations to the training data set, and the time series prediction model is retrained and multi-step predictions are performed again to obtain updated prediction results; Based on the updated forecast results, the electricity demand of the medical unit, the power generation capacity of renewable energy and the power supply price of the external power grid are comprehensively calculated to generate a detailed energy supply and demand balance plan.

7. The method according to claim 1, characterized in that Based on the optimal charging and discharging plan, a performance evaluation mechanism based on big data analysis is adopted, the operation data of the distributed energy system is collected and analyzed at preset time nodes, the failure risk of the operation data is predicted by a machine learning model, and an evaluation result is generated. The optimal energy allocation strategy and multi-objective optimization algorithm are iteratively optimized according to the evaluation result to form an adaptive and self-learning energy closed-loop management system, including: Using the optimal charging and discharging plan, determine the operating parameter range of the distributed energy system, wherein the operating parameters include: the charging and discharging state of the energy storage system, the energy consumption of each medical unit, and the power generation of renewable energy; At a preset time node, collect and analyze the operating data of the distributed energy system within the operating parameter range, the operating data including: energy efficiency ratio, carbon emissions, system reliability and response speed; Based on the operating data, predict the failure risk of the distributed energy system through a machine learning model to generate a failure risk prediction report; Using the fault risk prediction report and combining it with key indicators of the operating data, a performance evaluation result is generated; Using the performance evaluation results, iteratively optimizing the optimal energy allocation strategy and the multi-objective optimization algorithm to obtain optimized optimal energy allocation strategy and the multi-objective optimization algorithm; The optimized optimal energy allocation strategy and multi-objective optimization algorithm are applied again to the distributed energy system to form an adaptive and self-learning energy closed-loop management system.

8. A distributed energy optimization management system for rapid deployment of mobile cabin hospitals, characterized in that: include: A building module for constructing a multi-layer virtual energy network model of a mobile shelter hospital; An adjustment module, for dynamically adjusting energy allocation based on the multi-layer virtual energy network model and applying a deep reinforcement learning algorithm to generate an optimal energy allocation strategy; A forecasting module, for forecasting energy demand and power generation potential using time series forecasting technology in combination with the optimal energy allocation strategy, real-time weather forecast information, and historical energy consumption data and medical activity patterns, and generating an energy supply and demand balance plan; A scheduling module, which is used to intelligently schedule the charging and discharging process of the energy storage system based on the energy supply and demand balance plan and introduce a multi-objective optimization algorithm to obtain an optimal charging and discharging plan; The generation module, based on the optimal charging and discharging plan, adopts a performance evaluation mechanism based on big data analysis, collects and analyzes the operating data of the distributed energy system at preset time nodes, predicts the fault risk of the operating data through a machine learning model, generates evaluation results, and iteratively optimizes the optimal energy allocation strategy and multi-objective optimization algorithm based on the evaluation results to form an adaptive and self-learning energy closed-loop management system.

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 used to be called and executed by the processing component to implement a distributed energy optimization management method for rapidly deploying a mobile cabin hospital as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a distributed energy optimization management method for rapidly deploying a mobile cabin hospital as described in any one of claims 1 to 7 is implemented.

Images