Multi-source heterogeneous energy storage cooperative control method considering multi-coupling influence factors

Abstract

The invention relates to the technical field of multi-source heterogeneous energy storage cooperative control considering multi-coupling influence factors, and discloses a multi-source heterogeneous energy storage cooperative control method considering multi-coupling influence factors. According to the method, an output and adjustment characteristic model which accurately reflects the dynamic characteristics of the multi-type energy storage system is constructed, on the basis, a multi-constraint optimization model is established, and a cooperative charging and discharging plan is generated by adopting a multi-scale economic dispatching algorithm. The method has the beneficial effects that the problems of single factor consideration and insufficient model precision in the prior art are effectively solved, advantage complementation and collaborative optimization of energy storage resources with different characteristics are realized, and the economy of microgrid operation is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of multi-source heterogeneous energy storage collaborative control technology that considers multiple coupling factors, and particularly to a multi-source heterogeneous energy storage collaborative control method that considers multiple coupling factors. Background Technology

[0002] With the increasing penetration rate of renewable energy, distributed generation, represented by photovoltaic and wind power, is being used more and more widely in microgrids. However, the intermittency and volatility of its output pose significant challenges to the stable and economical operation of microgrids. Energy storage systems, especially heterogeneous energy storage systems that combine energy-type energy storage (such as flow batteries) and power-type energy storage (such as lithium batteries), are considered key technologies for smoothing out fluctuations and improving absorption capacity. Currently, energy storage collaborative control strategies for microgrids mostly focus on optimizing a single objective or limited factors, failing to fully consider the complex interactions of multiple coupled influencing factors inside and outside the system. This results in poor economic efficiency of the formulated charging and discharging plans, making it difficult to achieve globally optimal operation across multiple time scales while ensuring system safety.

[0003] Therefore, there is an urgent need for a collaborative control method that can deeply integrate multi-source heterogeneous data, accurately model energy storage characteristics, and perform multi-scale economic optimization under complex and multi-constraints, so as to fully tap the complementary potential of various types of energy storage and improve the overall operating efficiency of microgrids. Summary of the Invention

[0004] Based on this, it is necessary to propose a method, device, electronic equipment and storage medium for the coordinated control of multi-source heterogeneous energy storage that considers multiple coupling factors, in order to address the existing problem of coordinated control of multi-source heterogeneous energy storage considering multiple coupling factors.

[0005] A method for coordinated control of multi-source heterogeneous energy storage considering multiple coupling factors, the method comprising: Real-time acquisition of multi-source heterogeneous operation data of a specified microgrid system; wherein, the multi-source heterogeneous operation data includes photovoltaic power generation data, status data of at least two heterogeneous energy storage systems, user electricity consumption mode data, and external environment data; The multi-source heterogeneous operating data is standardized using a preset processing method to generate standardized operating data. A power output and regulation characteristic model of the multi-source heterogeneous energy storage system is established based on the standardized operating data. Based on the power output and regulation characteristic model, the user electricity consumption pattern data, and the external environment data, multiple constraints related to the operation of the specified microgrid system are generated. An optimization model is established based on the aforementioned multiple constraints, and a multi-scale economic dispatch algorithm is used to solve the problem with the objective function of optimizing the economic operation of the specified microgrid system, thereby generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems. The coordinated charge and discharge plan is executed in the designated microgrid system.

[0006] Furthermore, the at least two heterogeneous energy storage systems include power-type energy storage batteries and energy-type energy storage batteries. Before the step of establishing an optimization model based on the multiple constraints and using a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charge-discharge plan for the at least two heterogeneous energy storage systems, the method further includes: Obtain the real-time status of the specified microgrid system; When the real-time state requires a rapid power response, the priority of the power-type energy storage battery is set to be higher than that of the energy-type energy storage battery. When the real-time state requires long-term energy support, the priority of the energy-type energy storage battery is set to be higher than that of the power-type energy storage battery.

[0007] Furthermore, the power-type energy storage battery is a lithium iron phosphate battery system, and the energy-type energy storage battery is a vanadium redox flow battery system. The step of obtaining the real-time status of the designated microgrid system includes: Determine whether the external ambient temperature is greater than the preset temperature; If the external ambient temperature is greater than the preset temperature, the real-time state is determined to require a rapid power response.

[0008] Furthermore, the step of establishing an optimization model based on the multiple constraints and using a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems, includes: Output the Pareto solution set based on the optimization model; The strategies in the Pareto solution set are encoded to obtain the initial population; Genetic operations are performed on the initial population to generate a new generation population; wherein the genetic operations include selection, crossover, and mutation; The initial population and the new generation population are merged, and non-dominated sorting and crowding calculations are performed to select the next generation population. Determine whether the merged population meets the preset termination condition. If the preset termination condition is not met, use the next generation population as the initial population and repeat the target step and the steps after the target step until the preset termination condition is met to obtain the final population. The target step is to perform genetic operations on the initial population to generate a new generation population. The optimal solution is selected from the final population according to preset rules to generate the cooperative charging and discharging strategy.

[0009] Furthermore, before the step of establishing an optimization model based on the multiple constraints and using a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems, the method further includes: Obtain the time limit for each constraint in the multiple constraint conditions; Based on the time limit, time tags are added to each constraint to obtain the updated multiple constraints; wherein, the time tag is the effective time of the constraint.

[0010] Furthermore, prior to the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: The collaborative charging and discharging plan is compared with the problem strategies in the historical problem strategy library; Determine whether the comparison result is greater than a set threshold; If the comparison result is greater than the set threshold, the collaborative charging and discharging plan will be regenerated.

[0011] Furthermore, the step of standardizing the multi-source heterogeneous operating data using a preset processing method to generate standardized operating data includes: Obtain the data type of each running data; Obtain the corresponding standardized method based on the data type; The standardized data of each data type is processed according to the standardized processing method to obtain standardized data of each data type.

[0012] Furthermore, after the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: Obtain power grid early warning information to identify impending extreme scenarios; Invoke a reinforcement learning agent pre-matched to the extreme scenario; wherein the agent is obtained through offline training using historical operational data under extreme scenarios; The reinforcement learning agent outputs a temporary charging and discharging strategy to cope with the extreme scenario, which overrides the original coordinated charging and discharging plan.

[0013] Furthermore, after the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: Collect actual operating data of the at least two heterogeneous energy storage systems; Calculate the deviation between the actual operating data and the theoretical operating data in the coordinated charge and discharge plan; Determine whether the deviation value is greater than a preset deviation value; If the deviation exceeds the preset value, a new collaborative charging and discharging plan will be generated based on the actual operating data.

[0014] Furthermore, prior to the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: The coordinated charging and discharging plan is input into a digital twin model constructed based on the specified microgrid system for simulation execution; Obtain the preset performance indicators of the simulated execution by the digital twin model; Determine whether the preset performance indicators meet the preset requirements; If the preset performance indicators meet the preset requirements, then it is determined that the conditions for executing the coordinated charging and discharging plan in the specified microgrid system are met.

[0015] The beneficial effects of this invention are as follows: By deeply integrating and standardizing multi-source heterogeneous operating data, an output and regulation characteristic model that accurately reflects the dynamic characteristics of various types of energy storage systems is constructed. Based on this, a multi-constraint optimization model is established, and a multi-scale economic scheduling algorithm is used to generate a coordinated charging and discharging plan. This effectively solves the problems of single factor consideration and insufficient model accuracy in the existing technology, realizes the complementary advantages and coordinated optimization of energy storage resources with different characteristics, significantly improves the economic efficiency of microgrid operation, and improves the operating efficiency and safety rate of microgrids and reduces energy consumption through multi-source data fusion and optimized scheduling. Attached Figure Description

[0016] 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] in: Figure 1 This is a flowchart of a multi-source heterogeneous energy storage collaborative control method that considers multiple coupling influencing factors in one embodiment. Detailed Implementation

[0018] 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.

[0019] like Figure 1 As shown, in one embodiment, a multi-source heterogeneous energy storage collaborative control method considering multiple coupling factors is provided. This method can be applied to both terminals and servers; this embodiment illustrates its application to a terminal. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling factors specifically includes the following steps: S1: Real-time acquisition of multi-source heterogeneous operation data of a specified microgrid system; wherein, the multi-source heterogeneous operation data includes photovoltaic power generation data, status data of at least two heterogeneous energy storage systems, user electricity consumption mode data, and external environment data; S2: The multi-source heterogeneous operating data is standardized using a preset processing method to generate standardized operating data; S3: Establish a power output and regulation characteristic model of the multi-source heterogeneous energy storage system based on the standardized operating data; S4: Based on the power output and regulation characteristic model, the user power consumption pattern data, and the external environment data, generate multiple constraints related to the operation of the specified microgrid system; S5: Based on the multiple constraints, establish an optimization model and use a multi-scale economic dispatch algorithm to solve the problem with the objective function of optimizing the economic operation of the specified microgrid system, thereby generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems. S6: Execute the coordinated charge and discharge plan in the designated microgrid system.

[0020] As described in step S1 above, multi-source heterogeneous operation data related to the specified microgrid system is acquired through real-time monitoring. This multi-source heterogeneous operation data includes photovoltaic power generation data, status data of at least two different types of heterogeneous energy storage systems, user electricity consumption pattern data, and external environmental data. "Multiple coupled influencing factors" refers to a set of multiple factors that are interdependent and interact with each other during the operation of the specified microgrid system. These factors may include environmental conditions, technical parameters, economic factors, user behavior, etc., and in this application specifically refers to multi-source heterogeneous operation data. Specifically, photovoltaic power generation data mainly reflects the current output power of the photovoltaic system, which is directly affected by changes in weather and sunlight. Heterogeneous energy storage system status data includes the charging status, remaining capacity, and health status of each energy storage system, providing real-time information on the operation of the energy storage system. User electricity consumption pattern data involves users' daily electricity consumption habits; by analyzing user electricity consumption patterns, system demand can be better predicted. External environmental data includes meteorological information such as ambient temperature, humidity, and wind speed, as well as price fluctuations in the electricity market; these factors directly affect the charging and discharging strategies of energy storage. In addition, during the data acquisition process, real-time performance may be affected by communication delays or data loss. Therefore, if data is missing, the nearest neighbor interpolation method is used to supplement it.

[0021] As described in step S2 above, the multi-source heterogeneous operating data is standardized using a preset processing method to generate standardized operating data. Since this operating data comes from different sources, it may have different data formats, dimensions, and precisions, which may sometimes affect the accuracy of subsequent analysis. The purpose of standardization is to unify all data into a standard format, making it comparable and processable. Specific steps may include noise removal, filling in missing values, and unifying units. For photovoltaic power generation data, it needs to be converted to power (e.g., kW) under a unified time base, while energy storage system status data needs to be evaluated for charging and discharging status, capacity, etc., to reflect the actual situation. User electricity consumption pattern data also needs to be converted to ensure it can be clearly correlated with photovoltaic power generation data and energy storage status data. Through standardization, the generated standardized operating data will provide accurate and consistent basic data for subsequent characteristic modeling and constraint generation, thereby ensuring that the subsequent calculation process can reflect the true situation of the system.

[0022] As described in step S3 above, a power output and regulation characteristic model of the multi-source heterogeneous energy storage system is established based on the standardized operating data. This model can reflect the mathematical model of the energy storage system performance, taking into account various influencing factors, such as the charging and discharging efficiency, response speed, and characteristic curves of the energy storage system. Specifically, this characteristic model involves nonlinear function logic to better reflect the behavior of the energy storage system under different operating conditions. During the modeling process, the system output, charging speed, discharging speed, and losses will be analyzed and modeled based on historical data, taking into account the dynamic changes during the charging and discharging process. In addition, the model also needs to describe the cooperative characteristics between heterogeneous energy storage systems, such as how they work together to achieve the predetermined power transmission. The power output model is fitted based on historical data, and the formula is P=f(SOC,T), where P is the output power, SOC is the battery state of charge, T is the temperature, and f(SOC,T) is the calculation formula based on temperature and battery state of charge. Regulation characteristics refer to the system or device's response and regulation capabilities to input and output, including its behavior under various operating conditions and changes in external factors.

[0023] As described in step S4 above, based on the output and regulation characteristic model, the user electricity consumption pattern data, and the external environment data, multiple constraints related to the operation of the specified microgrid system are generated. These constraints are an important component of the optimization model, reflecting the system's requirements and limitations on various operating indicators. Specifically, the constraints may include, but are not limited to, minimum and maximum charging and discharging power limits of the energy storage system, maximum acceptable power demand limits for users, and limitations on the impact of environmental conditions on energy storage performance. For example, external environmental conditions such as temperature changes may affect the charging and discharging efficiency of the battery, thus requiring the setting of appropriate derating standards in the constraints. In addition, upper and lower limits of power consumption may be set according to the user's electricity consumption pattern to balance user demand and energy storage system performance. These generated multiple constraints will provide a constraint basis for the subsequent optimization model, ensuring that the solution obtained in the optimization process not only meets economic requirements but also conforms to the actual operating requirements of the system in terms of safety and reliability.

[0024] As described in step S5 above, an optimization model is established based on the multiple constraints, and a multi-scale economic dispatch algorithm is used to solve the problem with the objective function of optimizing the economic operation of the specified microgrid system, generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems. Setting the objective function is crucial to the entire optimization process. When constructing the optimization model, factors such as the energy storage system's dispatch requirements, energy utilization efficiency, and operating costs must be considered. The multi-scale economic dispatch algorithm can combine day-ahead, intraday, and real-time dispatch. This multi-scale economic dispatch algorithm refers to an optimization algorithm that combines day-ahead, intraday, and real-time dispatch, thereby comprehensively planning the charging and discharging strategy of the energy storage system while ensuring system safety and equipment effectiveness. Specific measures may include methods for finding the global optimum, improved heuristic algorithms, or evolutionary algorithms to ensure coverage of all possible variables and responses.

[0025] As described in step S6 above, the coordinated charging and discharging plan is executed in the designated microgrid system. Each energy storage device is scheduled to charge and discharge according to the generated plan to ensure that user needs are met while effectively handling fluctuations in renewable energy. This process may also involve real-time monitoring and adjustment. If the microgrid's operating status deviates from the expected value during execution, the original plan needs to be adjusted and optimized based on real-time monitoring data. Furthermore, the energy management system needs to comprehensively track the actual execution to ensure the accuracy of operations and the rationality of funding. Ultimately, through the effective execution of the coordinated charging and discharging plan, the economic efficiency, safety, and service quality of the microgrid are comprehensively improved, meeting user needs and maximizing the utilization potential of renewable energy. Through multi-source data fusion and optimized scheduling, the operating efficiency and safety rate of the microgrid are improved, and energy consumption is reduced.

[0026] In one embodiment, the at least two heterogeneous energy storage systems include power-type energy storage batteries and energy-type energy storage batteries. Before step S5, which involves establishing an optimization model based on the multiple constraints and employing a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charge-discharge plan for the at least two heterogeneous energy storage systems, the method further includes: S401: Obtain the real-time status of the specified microgrid system; S402: When the real-time state requires a fast power response, the priority of the power-type energy storage battery is set to be higher than that of the energy-type energy storage battery; S403: When the real-time state requires long-term energy support, the priority of the energy-type energy storage battery is set to be higher than that of the power-type energy storage battery.

[0027] As described in step S401 above, the real-time status related to the specified microgrid system is obtained. The real-time status of the microgrid system directly affects the subsequent scheduling decisions. The real-time status includes the need for rapid power response and the need for long-term energy support. Specifically, it can be obtained through multi-dimensional information analysis, such as current power demand, renewable energy output, the charging and discharging status of energy storage facilities, and changes in any external environmental factors (such as weather conditions). This information can be obtained through sensors, monitoring equipment, and data acquisition systems to achieve comprehensive monitoring of the internal and external environment of the microgrid. Specifically, the system will integrate data information from multiple signals such as photovoltaic power generation, wind power generation, and energy storage devices to form a dynamic overall status description.

[0028] As described in step S402 above, when the real-time state requires a rapid power response, the priority of the power-type energy storage battery is set higher than that of the energy-type energy storage battery. When the real-time state indicates that the microgrid is facing power shortages or a sudden increase in power demand, resources capable of providing rapid response need to be quickly mobilized. In this situation, power-type energy storage batteries (such as lithium batteries) typically have rapid charging and discharging capabilities and are therefore set to a higher priority. By increasing the priority of power-type energy storage batteries, additional power support can be provided in the shortest possible time to meet instantaneous load demands. Conversely, although energy-type energy storage batteries (such as flow batteries) have significant energy storage capabilities, their charging and discharging response speed is relatively slow. Therefore, their priority is reduced when a rapid power response is required. This decision-making process ensures that in emergency situations, the microgrid system can quickly adjust to cope with sudden changes, maintain system stability, and minimize the risk of power outages.

[0029] As described in step S403 above, when the real-time state requires long-term energy support, the priority of the energy-type energy storage battery is set higher than that of the power-type energy storage battery. When insufficient power supply from the grid is detected, and users' power consumption demand remains stable or increases, the system needs to provide power support for a longer period. In this case, energy-type energy storage batteries (such as flow batteries) become a more suitable choice due to their larger energy storage capacity. By increasing the priority of energy-type energy storage batteries, the microgrid can effectively utilize their longer discharge duration to meet user needs. For example, during peak electricity consumption periods and periods of insufficient renewable energy output, calling upon energy-type energy storage batteries will help maintain grid stability and ensure the continuity of power supply. This priority adjustment ensures that energy storage resources can be rationally utilized under specific circumstances, and optimal scheduling strategies can be adopted for different operational needs, thereby improving the overall operating efficiency and reliability of the microgrid under different scenarios.

[0030] In one embodiment, the power-type energy storage battery is a lithium iron phosphate battery system, and the energy-type energy storage battery is a vanadium redox flow battery system. Step S401, which involves obtaining the real-time status of the specified microgrid system, includes: S4011: Determine whether the external ambient temperature is greater than a preset temperature; S4012: If the external ambient temperature is greater than the preset temperature, the real-time state is determined to require a fast power response.

[0031] As described in step S4011 above, the operating status of the microgrid is assessed by real-time monitoring of the external ambient temperature. External environmental conditions, especially temperature, have a significant impact on the performance and charge / discharge efficiency of the energy storage system. For lithium iron phosphate battery systems, excessively high operating temperatures can lead to decreased charge / discharge efficiency and may even shorten battery life. Furthermore, high temperatures can cause the battery temperature to rise further under heavy loads, increasing safety hazards. Therefore, the current ambient temperature is acquired in real-time using temperature sensors or meteorological monitoring equipment and compared with a preset temperature. The preset temperature is generally set according to the battery manufacturer's recommendations and operating manual to ensure it effectively reflects the battery's safe operating range, for example, set to 35°C. During the assessment process, if the current temperature exceeds the preset value, it indicates that environmental conditions may affect battery performance, thereby affecting the stability and reliability of the power grid. If the temperature does not exceed the preset value, the implementation status can be further determined based on other acquired parameters.

[0032] As described in step S4012 above, if the external ambient temperature is higher than the preset temperature, the real-time state is determined to require a rapid power response. When the external ambient temperature is determined to be higher than the preset temperature, the system marks the real-time state as requiring a rapid power response. Under high-temperature environmental conditions, the risks and challenges faced by energy storage devices (especially power-type energy storage batteries such as lithium iron phosphate batteries) are particularly evident. For example, the efficiency of batteries decreases at high temperatures, and there is a risk of thermal runaway, leading to a decline in battery performance or even a safety accident. In this situation, the necessity of rapid response becomes obvious. By prioritizing power-type energy storage batteries, the system can quickly call upon these resources to provide the required rapid power support, ensuring that load requirements can be quickly met when grid demand surges. This rapid response mechanism helps microgrids reduce the risk of power outages due to insufficient power supply, improves the resilience and reliability of the system, and ensures that microgrids can maintain efficient operation under complex environmental and load changes, maximizing the utilization rate of renewable energy.

[0033] In another embodiment, the determination can be made based on both the ambient temperature and load demand. For example, when the ambient temperature is higher than a preset temperature and the load demand suddenly increases, the real-time state is determined to require a rapid power response; when the ambient temperature is higher than the preset temperature and the load demand does not suddenly increase, power storage is prioritized. Specifically, a threshold value can be set for the sudden increase in load demand. When the increase in load demand exceeds this threshold value, the real-time state is determined to require a rapid power response.

[0034] In one embodiment, step S5, which involves establishing an optimization model based on the multiple constraints and employing a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems, includes: S501: Output the Pareto solution set based on the optimization model; S502: Encode the strategies in the Pareto solution set to obtain the initial population; S503: Perform genetic operations on the initial population to generate a new generation population; wherein the genetic operations include selection, crossover, and mutation; S504: Merge the initial population with the new generation population, and perform non-dominated sorting and crowding calculation to select the next generation population; S505: Determine whether the merged population meets the preset termination condition. If the preset termination condition is not met, use the next generation population as the initial population, and repeat the target step and the steps after the target step until the preset termination condition is met to obtain the final population. The target step is to perform genetic operations on the initial population to generate a new generation population. S506: Select the optimal solution from the final population according to preset rules to generate the cooperative charging and discharging strategy.

[0035] As described in step S501 above, based on the established optimization model, a Pareto solution set is output. The core objective of this process is to identify the optimal solution set that balances different decisions. These solutions are not a single optimal solution, but a set of solutions that represent the equilibrium between various objects. They are generated in a multi-objective optimization problem. Each solution in the Pareto solution set is a solution that cannot further optimize any objective under specific constraints. For example, in the economic dispatch of microgrids, it may be necessary to consider multiple objectives such as economic efficiency, environmental impact, and battery life. By using methods such as non-dominated sorting algorithms and NSGA-II algorithms, these Pareto front solutions can be identified in the multi-objective space. Compared with single-objective optimization, multi-objective methods can more comprehensively reflect the trade-offs between various objectives in the optimization process, providing rich alternatives for subsequent decoding and decision-making, and enhancing the flexibility and adaptability of microgrids in actual operation.

[0036] As described in step S502 above, the strategies in the Pareto solution set are encoded to obtain the initial population. The encoding process is a crucial step in the genetic algorithm. A specific data structure (such as binary encoding) is used to convert the solutions into a computationally suitable algorithmic representation. The essence of this stage is to convert the solution into a format that a computer can understand and process, thereby generating new candidate solutions through genetic operations (such as selection, crossover, and mutation). In microgrid system applications, encoding can involve multiple aspects such as the charging and discharging strategies of energy storage devices, scheduling time, and power values. During the conversion process, it is essential to ensure that the encoding method fully expresses all decision parameters, enabling subsequent genetic operations to proceed effectively. The quality of the initial population directly affects the effectiveness of subsequent genetic operations. Therefore, a reasonable solution structure design must be considered during encoding to fully utilize the potential of these solutions in genetic operations and promote the progress of the optimization process.

[0037] As described in step S503 above, genetic operations are performed on the initial population to generate a new generation population. These genetic operations include selection, crossover, and mutation. Selection typically involves evaluating an individual's performance in the fitness function and selecting those with better performance for reproduction, thereby improving the superior traits of the offspring. Crossover combines the characteristics of two or more selected individuals to generate new individuals. In this way, the system can integrate the characteristics of superior individuals, hoping to generate a better solution in the next generation. Finally, mutation randomly alters certain genes in the new individuals to increase population diversity, thereby avoiding getting trapped in local optima. Generally, the combination of these steps can effectively enhance the search space, improve the algorithm's optimization ability, and explore new and better solutions among multiple candidate solutions. Through these genetic operations, the continuously evolving population will gradually approach the optimal solution, forming a more reasonable collaborative charging and discharging strategy for the energy storage devices within the system.

[0038] As described in step S504 above, the initial population is merged with the newly generated first-generation population to form a larger population. The merged population is used for subsequent non-dominated sorting and crowding calculation. The purpose of this process is to select the next generation of the population to ensure that subsequent generations can continuously evolve towards the Pareto optimal solution. During the non-dominated sorting process, the algorithm evaluates the relative merits of each individual in the solution space, grouping solutions that can dominate other individuals into the same level. Next, through crowding calculation, the distribution of each solution within its non-dominated level is evaluated to ensure a uniform distribution of solutions, avoid concentration in a specific area, and increase diversity. This combined use allows the genetic algorithm to consider not only the merits of individuals but also the diversity of the population, providing a rich set of potential solutions for the optimization process.

[0039] As described in step S505 above, it is determined whether the merged population meets the preset termination conditions. These termination conditions are typically based on the number of generations, the fitness stability of the solution, or the target value reached. For example, the number of merged populations might be greater than 300 to determine whether the optimization process should end. If these conditions are not met, the currently generated next-generation population will be used as the new initial population, and the target steps and subsequent steps will be repeated. This process reflects the iterative optimization characteristic of genetic algorithms, ensuring that the continuously evolving population can better adapt to the solution environment. In each iteration, the system gradually improves the quality of the solution through population selection, genetics, and evaluation. Ultimately, with multiple rounds of population evolution, the system can converge to a satisfactory solution, achieving the optimal configuration of the microgrid's coordinated charging and discharging strategy.

[0040] As described in step S506 above, the optimal solution is selected from the final population according to preset rules to generate the collaborative charging and discharging strategy. After multiple iterations and optimizations, the final population contains a series of well-performing solutions. The best solution needs to be selected from these solutions according to specific criteria. The preset rules may include fitness values, minimization or maximization of the objective function, or other actual conditions for strategy execution. After the optimal solution is selected, it is transformed into a charging and discharging strategy that meets the actual operational requirements, including explicit charging and discharging time, power settings, and priorities. The final generated collaborative charging and discharging strategy aims to maximize economic benefits while meeting user needs and improving the reliability and flexibility of the microgrid.

[0041] In one embodiment, before step S5, which involves establishing an optimization model based on the multiple constraints and employing a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems, the method further includes: S411: Obtain the time limit for each constraint in the multiple constraint conditions; S412: Add time tags to each constraint condition according to the time limit to obtain the updated multiple constraint conditions; wherein, the time tag is the effective time of the constraint condition.

[0042] As described in step S411 above, the time limits of each constraint in the multiple constraints are obtained. In order to make full use of energy storage resources and meet users' electricity demand, the allowed constraints usually change over time and are affected by market conditions, environmental factors and user behavior. Therefore, before establishing the optimization model, the effective time and timeliness of each constraint must be recorded in detail. For example, some constraints (such as electricity price constraints) may be effective within a set time period, while other conditions (such as battery charging and discharging limits) may also be affected by environmental changes (such as temperature changes). These changes may become more stringent within a specific time. Specifically, the time information associated with each constraint can be extracted by accessing real-time monitoring data and historical records. This can be achieved through an event-driven method, by setting a series of condition triggers to dynamically update the timeliness of the constraints. The process of obtaining these time limit information provides a basis for labeling the subsequent constraints, ensuring that the constraints used in the optimization model are based on the current real operating conditions and market status, thereby improving the effectiveness of the overall scheduling strategy.

[0043] As described in step S412 above, time tags are added to each constraint condition according to the time limit to obtain updated multiple constraint conditions. The time tag defines the effective time and effective period of each constraint condition, so that the optimization algorithm can take into account the time characteristics of these constraints during operation. By combining the time tag with the constraint condition, the model can flexibly adjust the effective constraints in different time periods, making it more adaptable. For example, the constraints of certain environmental conditions (such as temperature) may only be applicable in specific seasons and time periods, while the electricity price is usually dynamic and may have different values ​​at different times of the day. By attaching these time tags to the constraint conditions, the optimization model can automatically call the corresponding effective constraints according to the current time period when constructing the scheduling strategy, thereby improving the accuracy and scientific nature of the decision. In addition, this logical flexibility ensures that the microgrid can maximize the use of various heterogeneous energy storage resources under constantly changing external conditions and market environment, ensuring that user needs are met satisfactorily while optimizing power consumption and achieving safe, economical and efficient operation.

[0044] In one embodiment, prior to step S6 of executing the coordinated charge / discharge plan in the designated microgrid system, the method further includes: S511: Compare the collaborative charging and discharging plan with the problem strategies in the historical problem strategy library; S512: Determine whether the comparison result is greater than a set threshold; S513: If the comparison result is greater than the set threshold, then the collaborative charging and discharging plan is regenerated.

[0045] As described in step S511 above, the system compares the generated collaborative charging and discharging plan with past problem strategies in the historical problem strategy library to assess their similarity. The historical problem strategy library contains schemes that have appeared in the past operation of microgrids. These schemes may have exhibited poor economy, stability, or safety during execution, leading to operational problems or equipment failures. Through this comparison, the system can identify whether the newly generated charging and discharging strategy is highly similar to historically known failed or inefficient strategies. In fact, the main purpose of using similarity comparison is to avoid repeating mistakes and ensure that the new strategy will not cause the same problems due to similar conditions or decisions. Similarity calculation can be achieved through various methods, such as data-driven machine learning models, cosine similarity, Euclidean distance, and other metrics, to quantify the differences and similarities between schemes.

[0046] As described in step S512 above, the similarity comparison result obtained in the previous step will be judged, specifically, whether the similarity exceeds a preset threshold. The threshold is usually set based on historical data analysis, experience, and computational models, for example, 0.85. When the similarity comparison result is greater than the set threshold, it means that there is a significant similarity between the newly generated charging and discharging plan and the historical problem strategy. This may indicate that the currently generated plan has potential risks or defects. Therefore, in this case, further action is needed to ensure the reliability and stability of the microgrid. The charging and discharging plan will be regenerated to avoid potential safety hazards and economic losses. This dynamic judgment mechanism helps to achieve continuous optimization of the microgrid strategy, prevents exponential losses due to decision-making errors, and ensures the efficient, economical, and safe operation of the power system.

[0047] As described in step S513 above, when the judgment result shows that the comparison result is greater than the set threshold, the system will regenerate the coordinated charging and discharging plan. This process is to ensure the efficiency and reliability of the microgrid during operation, especially in decision-making adjustments under known risks. The process of regenerating the plan can be to reselect other generated schemes, or to re-examine multiple factors such as current operating data, user needs, external environment, and energy storage characteristics, in order to construct a new optimization strategy. The new plan is generated based on the latest real-time data and improved algorithms, striving to overcome the defects of the original problematic strategy when changing the strategy, thereby formulating a more optimized charging and discharging scheme. Specifically, this includes adjusting the energy storage priority, setting a new scheduling time, and changing the magnitude of charging and discharging power.

[0048] In one embodiment, step S2, which standardizes the multi-source heterogeneous operating data using a preset processing method to generate standardized operating data, includes: S201: Obtain the data type of each running data; S202: Obtain the corresponding standardized method based on the data type; S203: Process the running data of the corresponding data type according to the standardization processing method to obtain the standardized data of each running data.

[0049] As described in steps S201-S203 above, the collected multi-source heterogeneous operational data is classified to obtain the type information of each data type. Since the operational data sources involved in the microgrid system are diverse, including photovoltaic power generation data, energy storage system status data, user electricity consumption pattern data, and external environmental data, each type of data has different characteristics and meanings. Obtaining the data type involves analyzing the structure and content of the data source to identify the data format, unit, and physical meaning. For example, photovoltaic power generation data may be in units of power (kilowatts), while energy storage system status data may include multiple measurements such as battery voltage (volts), capacitance (kilowatt-hours), and temperature (degrees Celsius). Standardization methods need to be applied according to the different types of data being processed, using different standardization strategies. Different data types may require different standardization strategies. For example, for large-scale numerical data (such as photovoltaic power generation data), common standardization methods include max-min normalization or Z-score standardization; while for categorical data (such as user electricity consumption patterns), one-hot encoding or label encoding may be required. A predefined standardization method library can be used to dynamically match various data types and attach corresponding processing logic. Finally, the running data of the corresponding data type is processed according to the standardization processing method to obtain the standardized data of each running data.

[0050] In one embodiment, after step S6 of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: S701: Acquire power grid early warning information to identify impending extreme scenarios; S702: Invoke the reinforcement learning agent pre-matched with the extreme scenario; wherein, the agent is obtained through offline training using historical running data under extreme scenarios; S703: The reinforcement learning agent outputs a temporary charging and discharging strategy to cope with the extreme scenario, and overrides the original cooperative charging and discharging plan.

[0051] As described in steps S701-S703 above, by integrating various data sources, early warning information of the power grid is obtained in real time to identify impending extreme scenarios. This early warning information is typically provided by power monitoring systems, meteorological services, and market dispatch centers, covering various factors that may affect the balance of power supply and demand. For example, power grid early warning information may include weather forecasts (such as extreme weather events), warnings of a surge in power demand, and fault alarms for renewable energy equipment. A reinforcement learning agent related to the identified extreme scenarios is invoked. This reinforcement learning agent is trained offline by analyzing historical operational data under extreme scenarios. Specifically, a deep Q-network (DQN) can be used, taking historical extreme scenario data as input and minimizing power deficit as the reward function. It possesses a certain degree of intelligence and adaptability. Through training on historical data, the agent learns the optimal response strategy under extreme conditions. Specifically, this agent processes various past extreme operational scenario data, such as grid overload, equipment failure, and rapid changes in renewable energy output, understanding which strategy is most effective in dealing with these challenges. Based on the input information from the previous step, the reinforcement learning agent outputs a temporary charging and discharging strategy for the identified extreme scenarios. This strategy is a decision made by the agent after learning and analyzing various historical extreme scenarios. It aims to ensure that energy storage resources can be mobilized in the shortest time and with the best efficiency when the power grid faces emergencies. Unlike the original coordinated charging and discharging plan, this temporary strategy may be more inclined to quickly meet power demand or minimize system load in order to cope with unconventional events, such as sudden load spikes or a sharp drop in renewable energy.

[0052] In one embodiment, after step S6 of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: S711: Collect actual operating data of the at least two heterogeneous energy storage systems; S712: Calculate the deviation between the actual operating data and the theoretical operating data in the coordinated charge and discharge plan; S713: Determine whether the deviation value is greater than a preset deviation value; S714: If the deviation is greater than the preset deviation value, then the collaborative charging and discharging plan is regenerated based on the actual operating data.

[0053] As described in steps S711-S712 above, the system is responsible for collecting real-time operational data from at least two heterogeneous energy storage systems. This operational data includes various information such as the energy storage system's charging status, discharge power, operating time, internal temperature, overall health status, and battery cycle count. The system then calculates the deviation between the actual operational data and the theoretical operational data in the coordinated charge-discharge plan. The theoretical operational data is generated through an optimization model and represents the expected operational state under ideal conditions based on previously set indicators and strategies. In contrast, the actual operational data is the real-time state obtained from the physical energy storage devices and may be affected by various factors, leading to differences from the theoretical values. The deviation function can use absolute deviation, relative deviation, or mean square error to quantify the gap between the actual output and the expected target.

[0054] As described in steps S713-S714 above, the calculated deviation value needs to be judged to determine whether it exceeds the preset deviation value. The preset deviation value is usually a reasonable range determined during the system design phase or based on past operating data and experience, for example, set to 0.2. The purpose of comparing the deviation value with the preset value is to evaluate the accuracy and stability of the current system operation. If the deviation value is small, it indicates that the actual operation is relatively stable and the system can operate well according to the coordinated charging and discharging plan; conversely, if the deviation value is greater than the preset deviation value, it means that there may be operational problems in the system, and further measures need to be taken. If it is determined that the deviation value between the actual operating data and the theoretical operating data exceeds the preset value, the system will regenerate the coordinated charging and discharging plan based on the current actual operating data. In some embodiments, a backup plan can be generated in advance. Before the coordinated charging and discharging plan is regenerated, the backup plan is activated first, and the optimized plan is regenerated at the same time.

[0055] In one embodiment, prior to step S6 of executing the coordinated charge / discharge plan in the designated microgrid system, the method further includes: S521: Input the coordinated charging and discharging plan into a digital twin model constructed based on the specified microgrid system for simulation execution; S522: Obtain the preset performance indicators of the simulated execution of the digital twin model; S523: Determine whether the preset performance indicators meet the preset requirements; S524: If the preset performance indicators meet the preset requirements, then it is determined that the conditions for executing the coordinated charging and discharging plan in the specified microgrid system are met.

[0056] As described in steps S521-S522 above, the generated collaborative charging and discharging plan is input into a digital twin model built based on the microgrid system for simulation execution. The digital twin model is a virtual representation of the actual microgrid system, capable of reflecting the system's operating status and dynamic behavior in real time. By inputting the collaborative charging and discharging plan into this model, operators can test the plan's effectiveness in a safe virtual environment without affecting the actual system's operation. Specifically, if the specified microgrid system has a complex structure, a digital twin can be built based on a simplified physical model, focusing on simulating power flow. During the simulation, the system will reflect real-time changing parameters in the digital twin model, such as load demand, renewable energy output fluctuations, and energy storage response capabilities, ensuring the simulation results are realistic and reliable. Based on the executed digital twin model, a series of preset performance indicators are obtained. These performance indicators are typically key parameters for evaluating the actual effectiveness of the collaborative charging and discharging plan. For example, they may include the system's economic benefits (such as cost savings or revenue), power supply and demand balance, renewable energy absorption rate, the number of charge and discharge cycles of the energy storage system, equipment health status, and overall system stability.

[0057] As described in steps S523-S524 above, the preset performance indicators obtained from the digital twin model simulation are judged to determine whether these indicators meet the preset standards. The preset requirements are usually set based on historical data, best practices, and system performance management standards. These requirements may include economic benefits, load matching, renewable energy absorption capacity, and safe operating conditions of equipment. If the simulation results show that the performance indicators all exceed or meet these requirements, it means that the coordinated charging and discharging plan is theoretically reasonable and can effectively support the operation of the microgrid and further achieve the specified goals. Conversely, if one or more indicators fail to meet the preset standards, it indicates that the plan may lead to adverse consequences during implementation, such as system overload, economic losses, and equipment damage. If the judgment result shows that the preset performance indicators have met the preset requirements, the system will determine that the coordinated charging and discharging plan can be executed in the microgrid system.

[0058] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0059] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0060] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for coordinated control of multi-source heterogeneous energy storage considering multiple coupling influencing factors, characterized in that, The method includes: Real-time acquisition of multi-source heterogeneous operation data of a specified microgrid system; wherein, the multi-source heterogeneous operation data includes photovoltaic power generation data, status data of at least two heterogeneous energy storage systems, user electricity consumption mode data, and external environment data; The multi-source heterogeneous operating data is standardized using a preset processing method to generate standardized operating data. A power output and regulation characteristic model of the multi-source heterogeneous energy storage system is established based on the standardized operating data. Based on the power output and regulation characteristic model, the user electricity consumption pattern data, and the external environment data, multiple constraints related to the operation of the specified microgrid system are generated. An optimization model is established based on the aforementioned multiple constraints, and a multi-scale economic dispatch algorithm is used to solve the problem with the objective function of optimizing the economic operation of the specified microgrid system, thereby generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems. The coordinated charge and discharge plan is executed in the designated microgrid system.

2. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, The at least two heterogeneous energy storage systems include power-type energy storage batteries and energy-type energy storage batteries. Before the step of establishing an optimization model based on the multiple constraints and using a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charge and discharge plan for the at least two heterogeneous energy storage systems, the method further includes: Obtain the real-time status of the specified microgrid system; When the real-time state requires a rapid power response, the priority of the power-type energy storage battery is set to be higher than that of the energy-type energy storage battery. When the real-time state requires long-term energy support, the priority of the energy-type energy storage battery is set to be higher than that of the power-type energy storage battery.

3. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 2, characterized in that, The power-type energy storage battery is a lithium iron phosphate battery system, and the energy-type energy storage battery is a vanadium redox flow battery system. The step of obtaining the real-time status of the designated microgrid system includes: Determine whether the external ambient temperature is greater than the preset temperature; If the external ambient temperature is greater than the preset temperature, the real-time state is determined to require a rapid power response.

4. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, The step of establishing an optimization model based on the multiple constraints, and using a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, to generate a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems, includes: Output the Pareto solution set based on the optimization model; The strategies in the Pareto solution set are encoded to obtain the initial population; Genetic operations are performed on the initial population to generate a new generation population; wherein the genetic operations include selection, crossover, and mutation; The initial population and the new generation population are merged, and non-dominated sorting and crowding calculations are performed to select the next generation population. Determine whether the merged population meets the preset termination condition. If the preset termination condition is not met, use the next generation population as the initial population and repeat the target step and the steps after the target step until the preset termination condition is met to obtain the final population. The target step is to perform genetic operations on the initial population to generate a new generation population. The optimal solution is selected from the final population according to preset rules to generate the cooperative charging and discharging strategy.

5. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, Before the step of establishing an optimization model based on the multiple constraints and using a multi-scale economic dispatch algorithm to solve for the optimal economic operation of the specified microgrid system as the objective function, and generating a coordinated charging and discharging plan for the at least two heterogeneous energy storage systems, the method further includes: Obtain the time limit for each constraint in the multiple constraint conditions; Based on the time limit, time tags are added to each constraint to obtain the updated multiple constraints; wherein, the time tag is the effective time of the constraint.

6. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, Before the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: The collaborative charging and discharging plan is compared with the problem strategies in the historical problem strategy library; Determine whether the comparison result is greater than a set threshold; If the comparison result is greater than the set threshold, the collaborative charging and discharging plan will be regenerated.

7. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, The step of standardizing the multi-source heterogeneous operating data using a preset processing method to generate standardized operating data includes: Obtain the data type of each running data; Obtain the corresponding standardized method based on the data type; The standardized data of each data type is processed according to the standardized processing method to obtain standardized data of each data type.

8. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, After the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: Obtain power grid early warning information to identify impending extreme scenarios; Invoke a reinforcement learning agent pre-matched to the extreme scenario; wherein the agent is obtained through offline training using historical operational data under extreme scenarios; The reinforcement learning agent outputs a temporary charging and discharging strategy to cope with the extreme scenario, which overrides the original coordinated charging and discharging plan.

9. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, After the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: Collect actual operating data of the at least two heterogeneous energy storage systems; Calculate the deviation between the actual operating data and the theoretical operating data in the coordinated charge and discharge plan; Determine whether the deviation value is greater than a preset deviation value; If the deviation exceeds the preset value, a new collaborative charging and discharging plan will be generated based on the actual operating data.

10. The multi-source heterogeneous energy storage collaborative control method considering multiple coupling influencing factors according to claim 1, characterized in that, Before the step of executing the coordinated charge-discharge plan in the designated microgrid system, the method further includes: The coordinated charging and discharging plan is input into a digital twin model constructed based on the specified microgrid system for simulation execution; Obtain the preset performance indicators of the simulated execution by the digital twin model; Determine whether the preset performance indicators meet the preset requirements; If the preset performance indicators meet the preset requirements, then it is determined that the conditions for executing the coordinated charging and discharging plan in the specified microgrid system are met.

Images