Microgrid dispatching control method, device and equipment based on strategy optimization and medium

Abstract

The invention provides a micro-grid dispatching control method, device and equipment based on strategy optimization and a medium, and belongs to the technical field of power grid dispatching, and the method comprises the steps: collecting the real-time operation data of new energy power generation equipment, power utilization load equipment and an energy storage system; receiving power generation and power utilization plans of a user, extracting multi-resource dynamic response characteristics through an LSTM model in combination with the real-time operation data and the historical operation data, and outputting an adjustment capability evaluation matrix; receiving a power grid dispatching plan, matching an optimal operation mode from the knowledge base based on the adjustment capability evaluation matrix, and generating a dispatching instruction set; issuing a scheduling instruction set to the target device and executing the scheduling instruction set, collecting execution data in real time and calculating operation deviation; when the operation deviation exceeds a threshold value, an online Q learning algorithm is adopted to update the instruction weight, and an optimization strategy is fed back to the knowledge base. Through real-time data acquisition and an intelligent optimization algorithm, fine scheduling of the micro-grid is realized, the resource utilization rate and the operation flexibility are improved, and the operation cost is reduced.

Description

Technical Field

[0001] The present application belongs to the field of power grid dispatching technology, and specifically relates to a microgrid dispatching control method, device, equipment and medium based on strategy optimization. Background Art

[0002] With the development of distributed renewable energy (such as solar photovoltaic and wind power) and adjustable loads (electric vehicles), a new type of power system has emerged. However, the large-scale access of these distributed power sources and adjustable loads has significantly increased the random volatility on both the source and load sides of the power system, posing huge challenges to the traditional power grid that uses a source-follow-load dynamic regulation mode.

[0003] As a carrier of a new type of power grid, smart microgrids promote energy structure transformation through energy management, improve renewable energy absorption capacity, and enhance the flexibility and reliability of the power system. Traditional microgrid scheduling has the following defects: First, the scheduling model is single-minded. It focuses solely on peak shaving and valley filling and peak-valley arbitrage, without developing tailored scheduling strategies tailored to the characteristics of different electricity users and resource scenarios, resulting in limited resource utilization. For example, in industrial park scenarios, production schedules and electricity consumption characteristics vary significantly across enterprises. Traditional, single-minded scheduling models cannot fully leverage the potential of distributed power sources and adjustable loads. Second, resource coordination is insufficient. Restricting renewable energy, energy storage, or loads in isolation, without dynamic coupling of multiple resources, results in microgrids lacking flexibility and reliability in addressing complex power demand scenarios, making them prone to power shortages and wind and solar curtailment. Finally, scheduling strategies are rigid and rigid. They often rely on manual day-ahead instructions, lack a closed-loop feedback loop for execution, and have a high rate of deviation in execution results.

[0004] Although virtual power plants can currently carry out microgrid dispatching, they rely too much on macro-dispatching and ignore users' personalized needs (such as comfort and production scheduling), which may lead to reduced users' enthusiasm for participating in virtual power plant dispatching and affect the collaborative interaction between microgrids and virtual power plants. At the same time, the evaluation indicators are single and only assess grid deviations. There is no strategy self-optimization mechanism, which makes it difficult for microgrids to adapt to the dynamic changes in market rules when participating in electricity market transactions. Summary of the Invention

[0005] In a first aspect, an embodiment of the present application provides a microgrid dispatching control method based on strategy optimization, comprising the following steps: S1. Collect real-time operating data of new energy power generation equipment, power load equipment, and energy storage systems, and store them in a comprehensive database; S2. Receive user power generation and consumption plans, combine real-time and historical operating data, extract dynamic response features of multiple resources using a pre-trained LSTM model, and output a regulation capability evaluation matrix. S3 receives the grid dispatch plan, matches the optimal operating mode from the knowledge base based on the adjustment capacity evaluation matrix, and generates a dispatch instruction set for new energy power generation equipment, power load equipment, and energy storage systems; S4. Send the scheduling instruction set to the target device and execute it, collect execution data in real time and calculate the operation deviation; S5. When the operation deviation exceeds the threshold, the online Q learning algorithm is used to update the instruction weight and the optimization strategy is fed back to the knowledge base.

[0006] Furthermore, the specific steps of step S1 are as follows: S11. Collect three types of resource data and environmental parameters through a multi-protocol interface: Obtaining the active power of renewable energy power generation equipment through the Modbus / TCP protocol , power generation and operating status ; Collect the real-time power of power load equipment through the IOT network , electricity consumption and controllable identification ; Collect the charging and discharging power of the energy storage system through the CAN bus , SOC status and rated capacity ; Collect ambient temperature through sensors and irradiance ; S12. Store the collected real-time operation data into a comprehensive database according to the timestamp and perform spatiotemporal alignment: ; in, is the vector of real-time running data, t represents the timestamp; S13. Preprocess the collected real-time data, perform missing value filling and outlier processing.

[0007] Furthermore, the specific steps of step S2 are as follows: S21. Build LSTM model: Set the LSTM model input to a feature sequence with a time window length of k ; Among them, the eigenvector at time t ; The LSTM network architecture includes the input layer, the first LSTM layer, the second LSTM layer, the fully connected layer, and the output layer connected in sequence; Set the output to the regulation capability evaluation matrix represented by the normalized regulation capability indicator vector:

[0008] in, Indicates the power regulation capability, that is, the power response speed regulation capability index. Indicates the power regulation capability, that is, the continuous regulation time regulation capability indicator, It indicates the SOC state regulation capability, that is, the SOC change rate regulation capability index; S22. Divide the historical operation data into feature sequences and construct a training data set; Set the loss function and pre-train the LSTM model until the loss function converges or the maximum number of iterations is met; The loss function is as follows:

[0009] Where N represents the number of samples in the training data set, is the actual power change rate, is the actual energy regulation amount, is the SOC change rate, is the power conversion time, is the typical adjustment time, is the response time; S23. Divide the real-time operation data into time windows k to obtain real-time data windows , and input the pre-trained LSTM model, perform forward propagation, and obtain the output adjustment evaluation ability matrix:

[0010] The rows of the regulation evaluation capability matrix represent renewable energy power generation equipment, energy storage systems, and power load equipment, respectively, and the columns represent power regulation, power regulation, and SOC regulation, respectively. S24. Calculate the resource integration adjustment index and quantify the response capability by the following method:

[0011] in, To adjust the preset weights corresponding to the elements in the evaluation capability matrix, represents the energy storage system regulation index, represents the new energy regulation index, Indicates the load regulation index.

[0012] Furthermore, the specific steps of step S3 are as follows: S31. Power constraints at the grid connection point based on grid dispatching plan and daily power consumption constraints , combined with the adjustment capability evaluation matrix C, the optimal operation mode is selected through the decision tree preset in the knowledge base; The specific decisions are as follows: If the grid dispatch plan requires that the grid connection point power does not exceed the grid connection point power constraint , then select the power limited operation mode; If the grid dispatch plan requires that the daily power consumption does not exceed the daily power consumption constraint , then select the power-limited operation mode; If the grid dispatch plan requires zero-carbon operation, select the zero-carbon operation mode; If the grid dispatch plan is unconstrained, the economic operation mode is selected; If the grid is disconnected from the main grid, the microgrid autonomous operation mode is selected; Otherwise, select normal operation mode; S32. Based on the selected optimal operation mode, combined with the regulation capability evaluation matrix C and the resource fusion regulation index , generate a scheduling instruction set ; in, It is the new energy power generation equipment directive, It is the energy storage system instruction. It is the instruction for electrical load equipment; S33. Store the generated scheduling instruction set in the knowledge base, and record the basis for selecting the optimal operation mode and the instruction generation parameters.

[0013] Furthermore, the specific generation process of the scheduling instruction set in step S32 is as follows: S321. New Energy Power Generation Equipment Directive The generation process is as follows: In power-limited operation mode, set ; in, Indicates the maximum available power of new energy power generation equipment, represents the new energy regulation index, is the maximum available power allowed by the grid; In zero-carbon operation mode, the goal is to maximize new energy ; In the power limit operation mode, economic operation mode, microgrid autonomous operation mode and normal operation mode, the active power priority is set as the goal. ; S322. Energy storage system instructions The generation process is as follows: In power-limited operation mode settings ; in, Indicates the actual power of the load. represents the energy storage system regulation index, is the maximum available power allowed by the grid; Setting in zero carbon operation mode ; in, is the total load demand, is the amount of electricity generated by renewable energy, It is the purchase of green electricity. represents the energy storage system regulation index, It is the security verification time window for instruction execution; In the economic operation mode, according to the time-of-use electricity price, the valley period charging setting , peak period discharge setting ; in, Indicates the charging power of the energy storage system, Indicates the discharge power of the energy storage system; S323. Electrical load equipment instructions The generation process is as follows: In the power-limited operation mode, the controllable load is adjusted according to the load regulation index. The power is proportionally reduced, that is, ; in, Indicates the actual power of the load; In zero-carbon operation mode, only non-essential loads are adjusted; In the economic operation mode, load is shifted from peak period to valley period; No power load device instructions are generated in the power limit operation mode, microgrid autonomous operation mode, and normal operation mode.

[0014] Furthermore, the specific steps of step S4 are as follows: S41. Verify the instructions in the scheduling instruction set using security verification rules: For new energy power generation equipment, the command power is limited to not exceed its maximum available power, that is, ; in, Indicates the maximum available power of new energy power generation equipment, It is the New Energy Power Generation Equipment Directive; For the energy storage system, the charging and discharging power is set not to exceed the rated power, and the SOC change is within the allowable range, that is:

[0015]

[0016] in, Indicates the rated power of the energy storage system, Indicates the rated capacity of the energy storage system, Indicates the minimum allowable SOC of the energy storage system, Indicates the maximum allowable SOC of the energy storage system. Indicates the energy storage system instruction, It is the security verification time window for instruction execution; For power load equipment, the command power is set not to exceed the current actual power and does not change the production key load, that is, ; in, Indicates the actual power of the load. Indicates the power load equipment instruction; S42. Real-time collection of data after the execution of instructions in the scheduling instruction set and calculation of running deviation :

[0017] in, Indicates the deviation of renewable energy power generation power, , Indicates the actual power generation of new energy equipment, Indicates the rated power of the new energy power generation system, It is the New Energy Power Generation Equipment Directive; Indicates the power deviation of the energy storage system, , Indicates the actual energy storage system power, Indicates the rated power of the energy storage system, Indicates energy storage system instructions; Indicates the load power deviation, , Indicates the actual load power, Indicates the base power of the electrical load. Indicates the power load equipment instruction; The weight coefficient satisfies ; S43. Determine whether there is an SOC control target; If yes, go to step S44; If not, the deviation will be run As the final running deviation , proceed to step S5; S44. Calculate SOC deviation term :

[0018] in, Indicates the actual SOC, Indicates the target SOC; Final running deviation ,in, is the weight of the SOC deviation term, It is the operating deviation.

[0019] Furthermore, the specific steps of step S5 are as follows: S51. Determine the final running deviation Whether it exceeds the preset deviation threshold : like , then trigger Q learning optimization and go to step S52; like , then the current execution result is stored in the knowledge base as a successful case and ends; S52.Q learning optimization process: The state space is defined as follows:

[0020] Among them, mode is the operation mode code, represents the energy storage system regulation index, represents the new energy regulation index, represents the load regulation index, It is the running deviation; Define the action space as the set of actions that adjust the instruction weights; The rewards are defined as follows:

[0021] in, is the new running deviation, is the original running deviation; The Q table is updated as follows:

[0022] in, For the new state, is the new action, α is the learning rate, and γ is the discount factor; According to the updated Q value, select the action that maximizes the Q value To adjust the instruction weight; S53. Feedback the adjusted instruction weights as the optimized strategy to the knowledge base, and update the instruction generation rules for the corresponding operating mode.

[0023] In a second aspect, an embodiment of the present application further provides a microgrid dispatching control device based on strategy optimization, comprising: Aggregate resource information collection module, used to collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store it in a comprehensive database; The response capability assessment module receives user power generation and consumption plans, combines real-time and historical operation data, extracts dynamic response features of multiple resources through a pre-trained LSTM model, and outputs a regulation capability assessment matrix. A dispatch control strategy selection module is configured to receive a power grid dispatch plan, match an optimal operating mode from a knowledge base based on the regulation capability evaluation matrix, and generate a dispatch instruction set for renewable energy power generation equipment, power load equipment, and energy storage systems; The scheduling control instruction execution module is used to send the scheduling instruction set to the target device and execute it, collect execution data in real time and calculate the operation deviation; The strategy optimization module is used to update the instruction weight using the online Q learning algorithm when the operation deviation exceeds the threshold, and feed the optimization strategy back to the knowledge base.

[0024] In a third aspect, an embodiment of the present application further provides an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the steps of the microgrid dispatching and control method based on strategy optimization as described in the first aspect are implemented.

[0025] In a fourth aspect, an embodiment of the present application further provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the microgrid dispatching and control method based on strategy optimization as described in the first aspect.

[0026] It can be seen from the above technical solutions that this application has the following advantages: The microgrid dispatching and control method, device, equipment and medium based on strategy optimization provided in this application collects real-time operating data of new energy power generation equipment, power load equipment and energy storage system, combines the user's power generation and power consumption plan, uses a pre-trained LSTM model to extract the dynamic response characteristics of multiple resources, and generates a regulation capability evaluation matrix; based on the regulation capability evaluation matrix, the optimal operating mode is matched from the knowledge base, a dispatching instruction set is generated, and the operating deviation is optimized through an online Q learning algorithm, which can realize the refined dispatching of the microgrid, improve resource utilization, enhance the flexibility and reliability of the microgrid, and at the same time reduce operating costs and enhance the competitiveness of the microgrid in the power market. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for the description. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0028] Figure 1 Schematic diagram of the flow of the microgrid dispatching control method based on strategy optimization of the present invention.

[0029] Figure 2 Schematic diagram of the microgrid dispatching and control system based on strategy optimization of the present invention. DETAILED DESCRIPTION

[0030] The various embodiments of the present disclosure will be described in more detail below in the specific steps of the microgrid dispatch control method based on policy optimization. The present disclosure can have various embodiments, and adjustments and changes can be made therein. However, it should be understood that there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but rather that the present disclosure should be understood to cover all adjustments, equivalents, and / or alternatives that fall within the spirit and scope of the various embodiments of the present disclosure.

[0031] For example, the rapid development of distributed renewable energy sources (such as solar photovoltaics and wind power) and adjustable loads (such as electric vehicles) has led to the emergence of new power systems. However, the massive integration of these distributed power sources and adjustable loads has significantly increased the random fluctuations on both the source and load sides of the power system, posing unprecedented challenges to the traditional power grid, which primarily relies on "source follows load" regulation.

[0032] As an important component of the new power system, smart microgrids promote energy structure transformation and improve the absorption capacity of renewable energy through advanced energy management technologies, while enhancing the flexibility and reliability of the power system. However, traditional microgrid scheduling technology has the following major drawbacks: First, the dispatch model is relatively simple. Current dispatching primarily focuses on peak shaving and valley filling, and exploiting peak-valley electricity prices for arbitrage. It fails to develop targeted dispatching strategies based on the characteristics of different power users and resource scenarios, thus limiting resource utilization. For example, in industrial parks, production schedules and electricity usage characteristics vary significantly across enterprises. Traditional, single-source dispatching models cannot fully tap the potential of distributed power sources and adjustable loads.

[0033] Secondly, resource coordination capabilities are insufficient. Existing dispatching methods mostly regulate renewable energy generation, energy storage systems, or adjustable loads in isolation, lacking dynamic coupling and coordinated optimization between multiple resources. This makes microgrids lack flexibility and reliability when dealing with complex power demand scenarios, making them prone to power shortages and wind and solar power curtailment.

[0034] Finally, scheduling strategies are rigid and rigid. Traditional scheduling typically relies on manual day-ahead instructions and lacks a closed-loop feedback mechanism, resulting in a high rate of deviation in the execution of scheduling instructions.

[0035] Furthermore, while virtual power plants can dispatch microgrids, their current scheduling methods overly rely on macro-level dispatch instructions and ignore individual user needs (such as comfort and production scheduling). This can reduce user enthusiasm for participating in virtual power plant scheduling, further impacting the synergistic interaction between microgrids and virtual power plants. Furthermore, virtual power plant dispatch evaluation metrics are relatively simple, solely based on grid deviation, and an effective strategy self-optimization mechanism has yet to be established. This makes it difficult for microgrids to adapt to dynamic market changes when participating in electricity market transactions, preventing them from fully realizing their potential for flexibility and economic benefits.

[0036] To address the above issues, this embodiment provides a microgrid dispatching and control method based on strategy optimization. Through real-time data acquisition and intelligent optimization algorithms, it realizes refined microgrid dispatching, improves resource utilization and operational flexibility, and reduces operating costs.

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] See also Figure 1 FIG. 1 is a flow chart of a microgrid dispatch control method based on strategy optimization in a specific embodiment, the method comprising the following steps: S1. Collect real-time operating data of new energy power generation equipment, power load equipment, and energy storage systems, and store them in a comprehensive database; It should be noted that data collection provides a data basis for subsequent regulation capability evaluation and dispatch instruction generation, ensuring the accuracy and completeness of the data and improving the reliability of dispatch control; S2. Receive user power generation and consumption plans, combine real-time and historical operating data, extract dynamic response features of multiple resources using a pre-trained LSTM model, and output a regulation capability evaluation matrix. It should be noted that by extracting the dynamic response characteristics of multiple resources through the LSTM model, it is possible to accurately evaluate the adjustment capabilities of each resource, provide a scientific basis for the generation of scheduling instructions, and improve the accuracy and adaptability of scheduling; S3 receives the grid dispatch plan, matches the optimal operating mode from the knowledge base based on the adjustment capacity evaluation matrix, and generates a dispatch instruction set for new energy power generation equipment, power load equipment, and energy storage systems; It should be noted that by selecting the optimal operation mode through the decision tree, it is possible to quickly respond to grid dispatching needs, generate reasonable dispatching instructions, and improve the operating efficiency and flexibility of the microgrid; S4. Send the scheduling instruction set to the target device and execute it, collect execution data in real time and calculate the operation deviation; It should be noted that by real-time monitoring of the execution effect of scheduling instructions and calculating the operating deviation, data support is provided for subsequent strategy optimization, thereby improving the accuracy and reliability of scheduling control; S5. When the running deviation exceeds the threshold, the online Q-learning algorithm is used to update the instruction weight and the optimization strategy is fed back to the knowledge base; It should be noted that by updating the instruction weights, the dispatching strategy can be dynamically adjusted according to the real-time operating conditions, continuously optimizing the dispatching instructions and improving the operating efficiency and stability of the microgrid.

[0039] This embodiment achieves refined scheduling of microgrids through real-time data collection, intelligent regulation capability evaluation and optimization algorithms, improves resource utilization and operational flexibility, reduces operating costs, and enhances the competitiveness of microgrids in the power market.

[0040] Furthermore, as a refinement and extension of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, another microgrid dispatching control method based on policy optimization is provided, which includes the following steps: S1. Collect real-time operating data of new energy power generation equipment, power load equipment, and energy storage systems, and store them in a comprehensive database; The specific steps of step S1 are as follows: S11. Collect three types of resource data and environmental parameters through a multi-protocol interface: Obtaining the active power of renewable energy power generation equipment through the Modbus / TCP protocol , power generation and operating status ; Collect the real-time power of power load equipment through the IOT network , electricity consumption and controllable identification ; Collect the charging and discharging power of the energy storage system through the CAN bus , SOC status and rated capacity ; Collect ambient temperature through sensors and irradiance ; S12. Store the collected real-time operation data into a comprehensive database according to the timestamp and perform spatiotemporal alignment: ; in, is the vector of real-time running data, t represents the timestamp; S13. Preprocess the collected real-time data, perform missing value filling and outlier processing; Missing value supplementation:

[0041] Outlier handling: If , then replace it with the sliding mean ; S2. Receive user power generation and consumption plans, combine real-time and historical operating data, extract dynamic response features of multiple resources using a pre-trained LSTM model, and output a regulation capability evaluation matrix. The specific steps of step S2 are as follows: S21. Build LSTM model: Set the LSTM model input to a feature sequence with a time window length of k ; Among them, the eigenvector at time t ; The LSTM network architecture includes the input layer, the first LSTM layer, the second LSTM layer, the fully connected layer, and the output layer connected in sequence; The input layer uses a fully connected layer architecture to map the 5×k dimensional input to a 5k output. The first LSTM layer includes 64 LSTM units with a forget gate to map the 5k input to a 64-dimensional output. The second LSTM layer includes 32 LSTM units to map the 64-dimensional input to a 32-dimensional output. The fully connected layer linearly maps the 32-dimensional input to the output dimension, mapping the 32-dimensional input to the 3-dimensional output. The output layer uses Sigmoid activation to generate a regulation capability vector. Set the output to the regulation capability evaluation matrix represented by the normalized regulation capability indicator vector:

[0042] in, Indicates the power regulation capability, that is, the power response speed regulation capability index. Indicates the power regulation capability, that is, the continuous regulation time regulation capability indicator, It indicates the SOC state regulation capability, that is, the SOC change rate regulation capability index; S22. Divide the historical operation data into feature sequences and construct a training data set; Set the loss function and pre-train the LSTM model until the loss function converges or the maximum number of iterations is met; The loss function is as follows:

[0043] Where N represents the number of samples in the training data set, is the actual power change rate, is the actual energy regulation amount, is the SOC change rate, is the power conversion time, is the typical adjustment time, is the response time; S23. Divide the real-time operation data into time windows k to obtain real-time data windows , and input the pre-trained LSTM model, perform forward propagation, and obtain the output adjustment evaluation ability matrix:

[0044] The rows of the regulation evaluation capability matrix represent renewable energy power generation equipment, energy storage systems, and power load equipment, respectively, and the columns represent power regulation, power regulation, and SOC regulation, respectively. It should be noted that there is no SOC adjustment for new energy power generation equipment and power loads, so they are set to 0 in the matrix; Perform the forward pass:

[0045]

[0046] S24. Calculate the resource integration adjustment index and quantify the response capability by the following method:

[0047] in, To adjust the preset weights corresponding to the elements in the evaluation capability matrix, represents the energy storage system regulation index, represents the new energy regulation index, Indicates the load regulation index; For example, if , , ,but: , , ; S3 receives the grid dispatch plan, matches the optimal operating mode from the knowledge base based on the adjustment capacity evaluation matrix, and generates a dispatch instruction set for new energy power generation equipment, power load equipment, and energy storage systems; The specific steps of step S3 are as follows: S31. Power constraints at the grid connection point based on grid dispatching plan and daily power consumption constraints , combined with the adjustment capability evaluation matrix C, the optimal operation mode is selected through the decision tree preset in the knowledge base; The specific decisions are as follows: If the grid dispatch plan requires that the grid connection point power does not exceed the grid connection point power constraint , then select the power limited operation mode; If the grid dispatch plan requires that the daily power consumption does not exceed the daily power consumption constraint , then select the power-limited operation mode; If the grid dispatch plan requires zero-carbon operation, select the zero-carbon operation mode; If the grid dispatch plan is unconstrained, the economic operation mode is selected; If the grid is disconnected from the main grid, the microgrid autonomous operation mode is selected; Otherwise, select normal operation mode; It should be noted that zero-carbon operation means that the amount of electricity generated by new energy + the amount of green electricity purchased ≥ electricity consumption; S32. Based on the selected optimal operation mode, combined with the regulation capability evaluation matrix C and the resource fusion regulation index , generate a scheduling instruction set ; in, It is the new energy power generation equipment directive, It is the energy storage system instruction. It is the instruction for electrical load equipment; The specific generation process of the scheduling instruction set in step S32 is as follows: S321. New Energy Power Generation Equipment Directive The generation process is as follows: In power-limited operation mode, set ; in, Indicates the maximum available power of new energy power generation equipment, represents the new energy regulation index, is the maximum available power allowed by the grid; In zero-carbon operation mode, the goal is to maximize new energy ; In the power limit operation mode, economic operation mode, microgrid autonomous operation mode and normal operation mode, the active power priority is set as the goal. ; S322. Energy storage system instructions The generation process is as follows: In power-limited operation mode settings ; in, Indicates the actual power of the load. represents the energy storage system regulation index, is the maximum available power allowed by the grid; Setting in zero carbon operation mode ; in, is the total load demand, is the amount of electricity generated by renewable energy, It is the purchase of green electricity. represents the energy storage system regulation index, It is the security verification time window for instruction execution; In the economic operation mode, according to the time-of-use electricity price, the valley period charging setting , peak period discharge setting ; in, Indicates the charging power of the energy storage system, Indicates the discharge power of the energy storage system; S323. Electrical load equipment instructions The generation process is as follows: In the power-limited operation mode, the controllable load is adjusted according to the load regulation index. The power is proportionally reduced, that is, ; in, Indicates the actual power of the load; in zero-carbon operation mode, only non-essential loads are adjusted (such as delaying charging of charging piles); In the economic operation mode, load is shifted from peak period to valley period; In the power-limited operation mode, microgrid autonomous operation mode, and normal operation mode, no power load device instructions are generated; S33. The generated scheduling instruction set is stored in the knowledge base, and the optimal operating mode selection basis and instruction generation parameters are recorded; S4. Send the scheduling instruction set to the target device and execute it, collect execution data in real time and calculate the operation deviation; The specific steps of step S4 are as follows: S41. Verify the instructions in the scheduling instruction set using security verification rules: For new energy power generation equipment, the command power is limited to not exceed its maximum available power, that is, ; in, Indicates the maximum available power of new energy power generation equipment, It is the New Energy Power Generation Equipment Directive; For the energy storage system, the charging and discharging power is set not to exceed the rated power, and the SOC change is within the allowable range, that is:

[0048]

[0049] in, Indicates the rated power of the energy storage system, Indicates the rated capacity of the energy storage system, Indicates the minimum allowable SOC of the energy storage system, Indicates the maximum allowable SOC of the energy storage system. Indicates the energy storage system instruction, It is the security verification time window for instruction execution; For power load equipment, the command power is set not to exceed the current actual power and does not change the production key load, that is, ; in, Indicates the actual power of the load. Indicates the power load equipment instruction; S42. Real-time collection of data after the execution of instructions in the scheduling instruction set and calculation of running deviation :

[0050] in, Indicates the deviation of renewable energy power generation power, , Indicates the actual power generation of new energy equipment, Indicates the rated power of the new energy power generation system, It is the New Energy Power Generation Equipment Directive; Indicates the power deviation of the energy storage system, , Indicates the actual energy storage system power, Indicates the rated power of the energy storage system, Indicates energy storage system instructions; Indicates the load power deviation, , Indicates the actual load power, Indicates the base power of the electrical load. Indicates the base power of the electrical load. Indicates the power load equipment instruction; The weight coefficient satisfies ; S43. Determine whether there is an SOC control target; If yes, go to step S44; If not, the deviation will be run As the final running deviation , proceed to step S5; Exemplarily, there is an SOC control target in the zero-carbon operation mode; S44. Calculate SOC deviation term :

[0051] in, Indicates the actual SOC, Indicates the target SOC; Final running deviation ,in, is the weight of the SOC deviation term, is the operating deviation; S5. When the running deviation exceeds the threshold, the online Q learning algorithm is used to update the instruction weight and the optimization strategy is fed back to the knowledge base. The specific steps of step S5 are as follows: S51. Determine the final running deviation Whether it exceeds the preset deviation threshold : like , then trigger Q learning optimization and go to step S52; like , then the current execution result is stored in the knowledge base as a successful case and ends; S52.Q learning optimization process: The state space is defined as follows:

[0052] Among them, mode is the operation mode code, represents the energy storage system regulation index, represents the new energy regulation index, represents the load regulation index, is the operating deviation; Define the action space as the set of actions that adjust the instruction weights; For example: To increase the weight of energy storage regulation , To reduce load adjustment weight , To adjust the limit ratio of new energy; The rewards are defined as follows:

[0053] in, is the new running deviation, is the original running deviation; The Q table is updated as follows:

[0054] in, For the new state, is the new action, α is the learning rate, and γ is the discount factor; According to the updated Q value, select the action that maximizes the Q value To adjust the instruction weight; S53. Feedback the adjusted instruction weights as the optimized strategy to the knowledge base, and update the instruction generation rules for the corresponding operating mode.

[0055] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0056] like Figure 2 As shown, the following is an embodiment of a microgrid dispatching and control device based on policy optimization provided by an embodiment of the present disclosure. The device and the microgrid dispatching and control method based on policy optimization of the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiment of the microgrid dispatching and control device based on policy optimization, please refer to the embodiment of the microgrid dispatching and control method based on policy optimization.

[0057] The device includes: Aggregate resource information collection module, used to collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store it in a comprehensive database; The response capability assessment module receives user power generation and consumption plans, combines real-time and historical operation data, extracts dynamic response features of multiple resources through a pre-trained LSTM model, and outputs a regulation capability assessment matrix. A dispatch control strategy selection module is configured to receive a power grid dispatch plan, match an optimal operating mode from a knowledge base based on the regulation capability evaluation matrix, and generate a dispatch instruction set for renewable energy power generation equipment, power load equipment, and energy storage systems; The scheduling control instruction execution module is used to send the scheduling instruction set to the target device and execute it, collect execution data in real time and calculate the operation deviation; The strategy optimization module is used to update the instruction weight using the online Q learning algorithm when the operation deviation exceeds the threshold, and feed the optimization strategy back to the knowledge base.

[0058] This embodiment realizes the refined scheduling of microgrids, improves resource utilization and operational flexibility, reduces operating costs, and enhances the competitiveness of microgrids in the power market by integrating the interactive collaboration of the resource information collection module, the response capability evaluation module, the scheduling control strategy selection module, the scheduling control instruction execution module, and the strategy optimization module.

[0059] The microgrid dispatch control method based on policy optimization provided in the embodiment of the present application can be applied to electronic devices. Those skilled in the art will understand that the electronic device structure involved in the embodiment of the present invention does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently. In the embodiment of the present invention, the electronic device includes but is not limited to a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0060] The electronic device may include a processor, an external memory interface, an internal memory, a universal serial bus (USB) interface, a charging management module, a power management module, a battery, a wireless communication module, an audio module, a speaker, a microphone, a sensor module, a button, a camera, a display, and a SIM card interface, etc.

[0061] It is understood that the structures illustrated in the embodiments of the present application do not constitute specific limitations on the electronic device. In other embodiments of the present application, the electronic device may include more or fewer components than shown, or combine or separate certain components, or arrange the components differently. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0062] A processor may include one or more processing units, such as a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU). Different processing units may be independent devices or integrated into one or more processors.

[0063] The processor can be the nerve center and command center of the electronic device. The controller can generate operation control signals based on the instruction opcode and timing signal to complete the control of instruction fetching and execution.

[0064] The processor may also include a memory for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. This memory can store instructions or data that the processor has just used or is reusing. If the processor needs to use the instruction or data again, it can directly call it from the memory. This avoids repeated accesses, reduces processor latency, and thus improves system efficiency.

[0065] The above-mentioned electronic device realizes the microgrid dispatching and control method based on strategy optimization of the present application to collect real-time operating data of new energy power generation equipment, power load equipment and energy storage system, and stores it in a comprehensive database; receives user power generation and power consumption plans, combines real-time operating data and historical operating data, extracts multi-resource dynamic response characteristics through a pre-trained LSTM model, and outputs a regulation capability evaluation matrix; receives the power grid dispatching plan, matches the optimal operating mode from the knowledge base based on the regulation capability evaluation matrix, and generates a dispatching instruction set for new energy power generation equipment, power load equipment and energy storage system; sends the dispatching instruction set to the target device and executes it, collects execution data in real time and calculates the operation deviation; when the operation deviation exceeds the threshold, uses an online Q learning algorithm to update the instruction weight, and feeds back the optimization strategy to the knowledge base. The technical solution achieves the beneficial effects of realizing refined dispatching of microgrids, improving resource utilization and operation flexibility, reducing operation costs, and enhancing the competitiveness of microgrids in the power market.

[0066] The storage medium provided in this application stores a program product that can implement a microgrid dispatching control method based on strategy optimization.

[0067] The microgrid dispatching and control method based on strategy optimization includes: collecting real-time operating data of renewable energy power generation equipment, power load equipment and energy storage systems, and storing it in a comprehensive database; receiving user power generation and consumption plans, combining real-time operating data with historical operating data, extracting dynamic response characteristics of multiple resources through a pre-trained LSTM model, and outputting a regulation capability evaluation matrix; receiving the power grid dispatching plan, matching the optimal operating mode from the knowledge base based on the regulation capability evaluation matrix, and generating a dispatching instruction set for renewable energy power generation equipment, power load equipment and energy storage systems; issuing the dispatching instruction set to the target device and executing it, collecting execution data in real time and calculating the operation deviation; when the operation deviation exceeds the threshold, using the online Q learning algorithm to update the instruction weight and feed the optimization strategy back to the knowledge base.

[0068] In some possible embodiments, the microgrid dispatching control method based on strategy optimization disclosed herein can be implemented in the form of a program product, which includes program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps of various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0069] The storage medium of the present disclosure can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0070] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A microgrid dispatching control method based on strategy optimization, characterized in that: The steps include: S1. Collect real-time operating data of new energy power generation equipment, power load equipment, and energy storage systems, and store them in a comprehensive database; S2. Receive user power generation and consumption plans, combine real-time and historical operating data, extract dynamic response features of multiple resources using a pre-trained LSTM model, and output a regulation capability evaluation matrix. S3 receives the grid dispatch plan, matches the optimal operating mode from the knowledge base based on the adjustment capacity evaluation matrix, and generates a dispatch instruction set for new energy power generation equipment, power load equipment, and energy storage systems; S4. Send the scheduling instruction set to the target device and execute it, collect execution data in real time and calculate the operation deviation; S5. When the operation deviation exceeds the threshold, the online Q learning algorithm is used to update the instruction weight and the optimization strategy is fed back to the knowledge base.

2. The microgrid dispatching control method based on strategy optimization according to claim 1 is characterized in that: The specific steps of step S1 are as follows: S11. Collect three types of resource data and environmental parameters through a multi-protocol interface: Obtaining the active power of renewable energy power generation equipment through the Modbus / TCP protocol , power generation and operating status ; Collect the real-time power of power load equipment through the IOT network , electricity consumption and controllable identification ; Collect the charging and discharging power of the energy storage system through the CAN bus , SOC status and rated capacity ; Collect ambient temperature through sensors and irradiance ; S12. Store the collected real-time operation data into a comprehensive database according to the timestamp and perform spatiotemporal alignment: ； in, is the vector of real-time running data, t represents the timestamp; S13. Preprocess the collected real-time data, perform missing value filling and outlier processing.

3. The microgrid dispatching control method based on strategy optimization according to claim 2 is characterized in that: The specific steps of step S2 are as follows: S21. Build LSTM model: Set the LSTM model input to a feature sequence with a time window length of k ; Among them, the eigenvector at time t ; The LSTM network architecture includes the input layer, the first LSTM layer, the second LSTM layer, the fully connected layer, and the output layer connected in sequence; Set the output to the regulation capability evaluation matrix represented by the normalized regulation capability indicator vector: in, Indicates the power regulation capability, that is, the power response speed regulation capability index. Indicates the power regulation capability, that is, the continuous regulation time regulation capability indicator, It indicates the SOC state regulation capability, that is, the SOC change rate regulation capability index; S22. Divide the historical operation data into feature sequences and construct a training data set; Set the loss function and pre-train the LSTM model until the loss function converges or the maximum number of iterations is met; The loss function is as follows: Where N represents the number of samples in the training data set, is the actual power change rate, is the actual energy regulation amount, is the SOC change rate, is the power conversion time, is the typical adjustment time, is the response time; S23. Divide the real-time operation data into time windows k to obtain real-time data windows , and input the pre-trained LSTM model, perform forward propagation, and obtain the output adjustment evaluation ability matrix: The rows of the regulation evaluation capability matrix represent renewable energy power generation equipment, energy storage systems, and power load equipment, respectively, and the columns represent power regulation, power regulation, and SOC regulation, respectively. S24. Calculate the resource integration adjustment index and quantify the response capability by the following method: in, To adjust the preset weights corresponding to the elements in the evaluation capability matrix, represents the energy storage system regulation index, represents the new energy regulation index, Indicates the load regulation index.

4. The microgrid dispatching control method based on strategy optimization according to claim 3 is characterized in that: The specific steps of step S3 are as follows: S31. Power constraints at the grid connection point based on grid dispatching plan and daily power consumption constraints , combined with the adjustment capability evaluation matrix C, the optimal operation mode is selected through the decision tree preset in the knowledge base; The specific decisions are as follows: If the grid dispatch plan requires that the grid connection point power does not exceed the grid connection point power constraint , then select the power limited operation mode; If the grid dispatch plan requires that the daily power consumption does not exceed the daily power consumption constraint , then select the power-limited operation mode; If the grid dispatch plan requires zero-carbon operation, select the zero-carbon operation mode; If the grid dispatch plan is unconstrained, the economic operation mode is selected; If the grid is disconnected from the main grid, the microgrid autonomous operation mode is selected; Otherwise, select normal operation mode; S32. Based on the selected optimal operation mode, combined with the regulation capability evaluation matrix C and the resource fusion regulation index , generate a scheduling instruction set ; in, It is the new energy power generation equipment directive, It is the energy storage system instruction. It is the instruction for electrical load equipment; S33. Store the generated scheduling instruction set in the knowledge base, and record the basis for selecting the optimal operation mode and the instruction generation parameters.

5. The microgrid dispatching control method based on strategy optimization according to claim 4 is characterized in that: The specific generation process of the scheduling instruction set in step S32 is as follows: S321. New Energy Power Generation Equipment Directive The generation process is as follows: In power-limited operation mode, set ; in, Indicates the maximum available power of new energy power generation equipment, represents the new energy regulation index, is the maximum available power allowed by the grid; In zero-carbon operation mode, the goal is to maximize new energy ; In the power limit operation mode, economic operation mode, microgrid autonomous operation mode and normal operation mode, the active power priority is set as the goal. ; S322. Energy storage system instructions The generation process is as follows: In power-limited operation mode settings ; in, Indicates the actual power of the load. represents the energy storage system regulation index, is the maximum available power allowed by the grid; Setting in zero carbon operation mode ; in, is the total load demand, is the amount of electricity generated by renewable energy, It is the purchase of green electricity. represents the energy storage system regulation index, It is the security verification time window for instruction execution; In the economic operation mode, according to the time-of-use electricity price, the valley period charging setting , peak period discharge setting ; in, Indicates the charging power of the energy storage system, Indicates the discharge power of the energy storage system; S323. Electrical load equipment instructions The generation process is as follows: In the power-limited operation mode, the controllable load is adjusted according to the load regulation index. The power is proportionally reduced, that is, ; in, Indicates the actual power of the load; In zero-carbon operation mode, only non-essential loads are adjusted; In the economic operation mode, load is shifted from peak period to valley period; No power load device instructions are generated in the power limit operation mode, microgrid autonomous operation mode, and normal operation mode.

6. The microgrid dispatching control method based on strategy optimization according to claim 4 is characterized in that: The specific steps of step S4 are as follows: S41. Verify the instructions in the scheduling instruction set using security verification rules: For new energy power generation equipment, the command power is limited to not exceed its maximum available power, that is, ; in, Indicates the maximum available power of new energy power generation equipment, It is the New Energy Power Generation Equipment Directive; For the energy storage system, the charging and discharging power is set not to exceed the rated power, and the SOC change is within the allowable range, that is: in, Indicates the rated power of the energy storage system, Indicates the rated capacity of the energy storage system, Indicates the minimum allowable SOC of the energy storage system, Indicates the maximum allowable SOC of the energy storage system. Indicates the energy storage system instruction, It is the security verification time window for instruction execution; For power load equipment, the command power is set not to exceed the current actual power and does not change the production key load, that is, ; in, Indicates the actual power of the load. Indicates the power load equipment instruction; S42. Real-time collection of data after the execution of instructions in the scheduling instruction set and calculation of running deviation : in, Indicates the deviation of renewable energy power generation power, , Indicates the actual power generation of new energy equipment, Indicates the rated power of the new energy power generation system, It is the New Energy Power Generation Equipment Directive; Indicates the power deviation of the energy storage system, , Indicates the actual energy storage system power, Indicates the rated power of the energy storage system, Indicates energy storage system instructions; Indicates the load power deviation, , Indicates the actual load power, Indicates the base power of the electrical load. Indicates the power load equipment instruction; The weight coefficient satisfies ; S43. Determine whether there is an SOC control target; If yes, go to step S44; If not, the deviation will be run As the final running deviation , proceed to step S5; S44. Calculate SOC deviation term : in, Indicates the actual SOC, Indicates the target SOC; Final running deviation ,in, is the weight of the SOC deviation term, It is the operating deviation.

7. The microgrid dispatching control method based on strategy optimization according to claim 6 is characterized in that: The specific steps of step S5 are as follows: S51. Determine the final running deviation Whether it exceeds the preset deviation threshold : like , then trigger Q learning optimization and go to step S52; like , then the current execution result is stored in the knowledge base as a successful case and ends; S52.Q learning optimization process: The state space is defined as follows: Among them, mode is the operation mode code, represents the energy storage system regulation index, represents the new energy regulation index, represents the load regulation index, It is the running deviation; Define the action space as the set of actions that adjust the instruction weights; The rewards are defined as follows: in, is the new running deviation, is the original running deviation; The Q table is updated as follows: in, For the new state, is the new action, α is the learning rate, and γ is the discount factor; According to the updated Q value, select the action that maximizes the Q value To adjust the instruction weight; S53. Feedback the adjusted instruction weights as the optimized strategy to the knowledge base, and update the instruction generation rules for the corresponding operating mode.

8. A microgrid dispatching control device based on strategy optimization, characterized in that: include: Aggregate resource information collection module, used to collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store it in a comprehensive database; The response capability assessment module receives user power generation and consumption plans, combines real-time and historical operation data, extracts dynamic response features of multiple resources through a pre-trained LSTM model, and outputs a regulation capability assessment matrix. A dispatch control strategy selection module is configured to receive a power grid dispatch plan, match an optimal operating mode from a knowledge base based on the regulation capability evaluation matrix, and generate a dispatch instruction set for renewable energy power generation equipment, power load equipment, and energy storage systems; The scheduling control instruction execution module is used to send the scheduling instruction set to the target device and execute it, collect execution data in real time and calculate the operation deviation; The strategy optimization module is used to update the instruction weight using the online Q learning algorithm when the operation deviation exceeds the threshold, and feed the optimization strategy back to the knowledge base.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the microgrid dispatching control method based on strategy optimization as described in any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the microgrid dispatching control method based on strategy optimization as claimed in any one of claims 1 to 7 are implemented.

Images