Micro-grid optimal scheduling method, system and equipment considering flexible load

Abstract

The invention discloses a micro-grid optimal scheduling method, system and device considering flexible loads, and relates to the field of micro-grid optimal scheduling. Establishing a comprehensive objective function and constructing an operation load constraint set by taking micro-grid operation cost minimization and environmental benefit maximization as objectives; based on the operation load constraint set, solving the comprehensive objective function to obtain day-ahead scheduling information and intra-day rolling information; issuing a flexible load regulation and control instruction to the micro-grid through the block chain smart contract, and collecting flexible load real-time operation feedback information; and in combination with a deep reinforcement learning emergency regulation and control strategy and a sensitivity analysis optimization strategy, determining a compensation optimization scheduling scheme according to the flexible load real-time operation feedback information and the flexible load regulation and control instruction. The method can give consideration to both environmental protection and economy, improves the reasonability of load regulation and control, and effectively deals with the uncertainty of the source load.

Description

Technical Field

[0001] This application relates to the field of microgrid optimal dispatch, and in particular to a microgrid optimal dispatch method, system and equipment that takes into account flexible loads. Background Technology

[0002] Currently, for the optimization of microgrid dispatch, dynamic shiftable loads, seasonally dynamic transferable loads, and equipment-limited loads that can be reduced can be introduced to achieve dynamic optimization management of energy resources and load demand. However, the above methods still have the following limitations: they only target specific loads such as temperature control equipment and electric vehicle charging, without considering more complex load characteristics; and they cannot meet economic and environmental requirements by simply calculating pollutant emission coefficients. Summary of the Invention

[0003] The purpose of this application is to provide a microgrid optimization scheduling method, system, and equipment that takes into account flexible loads, which can balance environmental protection and economy, improve the rationality of load regulation, and effectively handle source-load uncertainty.

[0004] To achieve the above objectives, this application provides the following solution:

[0005] Firstly, this application provides a microgrid optimal scheduling method considering flexible loads, including:

[0006] Acquire operational data of distributed power sources, flexible loads, and energy storage within the microgrid;

[0007] Based on the distributed power source operation data, the flexible load operation data, and the energy storage data, a comprehensive objective function is established with the goal of minimizing microgrid operating costs and maximizing environmental benefits, and an operating load constraint set is constructed.

[0008] Based on the set of operating load constraints, the comprehensive objective function is solved to obtain the day-ahead scheduling information and intraday rolling information;

[0009] Based on the day-ahead scheduling information and the intraday rolling information, flexible load control instructions are issued to the microgrid through a blockchain smart contract, and real-time operation feedback information of the flexible load is collected.

[0010] By combining deep reinforcement learning emergency control strategies and sensitivity analysis optimization strategies, a compensation optimization scheduling scheme is determined based on the real-time operation feedback information of the flexible load and the control instructions of the flexible load.

[0011] Secondly, this application provides a microgrid optimized dispatching system considering flexible loads, comprising:

[0012] The data acquisition module is used to acquire operating data of distributed power sources, flexible loads, and energy storage within the microgrid.

[0013] The objective function and constraint establishment module is used to: establish a comprehensive objective function based on the distributed power source operation data, the flexible load operation data, and the energy storage data, with the goal of minimizing the microgrid operating cost and maximizing environmental benefits, and construct a set of operating load constraints;

[0014] The objective function solving module is used to: solve the comprehensive objective function based on the set of operating load constraints to obtain day-ahead scheduling information and intraday rolling information;

[0015] The feedback information collection module is used to: issue flexible load control instructions to the microgrid through a blockchain smart contract based on the day-ahead scheduling information and the intraday rolling information, and collect real-time operation feedback information of the flexible load;

[0016] The scheduling scheme optimization module is used to: combine deep reinforcement learning emergency control strategies and sensitivity analysis optimization strategies, and determine a compensation optimization scheduling scheme based on the real-time operation feedback information of the flexible load and the control instructions of the flexible load.

[0017] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the microgrid optimization scheduling method considering flexible loads.

[0018] According to the specific embodiments provided in this application, this application has the following technical effects: This application establishes a comprehensive objective function based on distributed power source operation data, flexible load operation data, and energy storage data, taking into account both minimizing operating costs and maximizing environmental benefits, thus balancing economic and environmental needs. The rationality of load regulation is ensured by setting a set of operating load constraints. The comprehensive objective function is solved based on the set of operating load constraints to obtain day-ahead scheduling information and intraday rolling information. Through the hierarchical architecture of the aforementioned day-ahead scheduling and intraday rolling optimization, flexible load regulation instructions are determined, enabling real-time processing of source-load uncertainties. Combining deep reinforcement learning emergency regulation strategies and sensitivity analysis optimization strategies, a compensation-optimized scheduling scheme is determined based on real-time operating feedback information of flexible loads and flexible load regulation instructions. This allows for rapid correction of the scheduling scheme when power deviation exceeds limits, enhancing system stability. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a microgrid optimization scheduling method considering flexible loads, provided as an embodiment of this application.

[0021] Figure 2 This is a flowchart illustrating a microgrid optimization scheduling method considering flexible loads, provided as another embodiment of this application.

[0022] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] To make the objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] In one exemplary embodiment, such as Figure 1 As shown, a microgrid optimization scheduling method considering flexible loads is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, it includes the following steps 101 to 105.

[0026] Step 101: Obtain distributed power source operation data, flexible load operation data, and energy storage data within the microgrid.

[0027] The distributed power generation operation data includes the power generation capacity, efficiency, and operating status data of each distributed power source. This data is collected once per second, ensuring real-time performance and accuracy. Power generation capacity is used to calculate environmental benefits. Environmental benefits are measured by the sum of the products of the power generation capacity of each distributed power source and the pollutant emission coefficient per unit of power generation, reflecting the goal of maximizing environmental benefits in the comprehensive objective function. Power generation capacity also participates in the calculation of power balance constraints, ensuring real-time balance between distributed power sources, energy storage, purchased power, and load demand. Power generation efficiency affects power generation costs. When calculating the power generation cost of distributed power sources, efficiency and power generation capacity jointly determine the cost, thus affecting the microgrid operating cost. Power source operating status data (such as normal operation or fault status) helps determine whether a distributed power source can generate power normally. If it is in a fault state, its power generation capacity is 0, which will affect power balance constraints and the entire dispatch plan.

[0028] The flexible load operation data includes the power demand, time range, electricity value, priority, and user satisfaction parameters for each flexible load. Power demand is key data for calculating power balance constraints, ensuring real-time balance between the demands of distributed power sources, energy storage, purchased power, and flexible loads. When considering constraints for transferable, reduceable, and adjustable loads, power demand determines the basic quantity for load regulation. For transferable loads, the time range determines the time interval for load transfer when establishing dynamic spatiotemporal transfer constraints based on the load transfer characteristic matrix and time period shift probability factors. For adjustable loads, the time range affects the collaborative optimization of power time shifts across multiple time periods. In the regulation of adjustable loads, the shift priority coefficient matrix is ​​formulated based on load priority, with higher priority loads being prioritized for satisfaction or adjustment to ensure power supply for important loads. For reduceable loads, fuzzy chance constraints are used, quantifying the nonlinear relationship between load reduction and user satisfaction through membership functions to ensure the rationality of load regulation and balance user electricity experience with microgrid operational efficiency. Electricity value can be collected hourly.

[0029] The energy storage data includes the state of charge (SOC) and charging / discharging power of each energy storage system. The SOC is used in the energy storage constraint calculation. Based on the current SOC, charging / discharging power, charging / discharging efficiency, energy storage capacity, and time step, the SOC at the next moment is determined. It also affects the formulation of energy storage charging and discharging strategies. If the SOC is low, charging may be prioritized; if it is high and it is during a period of low load and low electricity price, discharging may be prioritized.

[0030] In addition, in practical applications, it can also acquire state information such as voltage, current and weather forecast data, providing a data foundation for the subsequent construction of comprehensive objective functions and operating load constraint sets.

[0031] In another practical application, for the initially collected multiple data sets, data cleaning, outlier handling, and normalization can be performed to form optimized data, namely the data obtained in step 101 of this application. Data cleaning involves removing missing or invalid values ​​caused by sensor malfunctions or communication interruptions to ensure the integrity and reliability of the dataset. Normalization can employ a min-max normalization method to convert the collected raw values ​​into standardized values ​​within the [0,1] interval. The normalization formula is:

[0032]

[0033] Where, x norm For standardized values, x min and x max These are the minimum and maximum values ​​for the corresponding data types, respectively, where x is the original value.

[0034] Step 102: Based on the distributed power generation operation data, the flexible load operation data, and the energy storage data, a comprehensive objective function is established with the goals of minimizing microgrid operating costs and maximizing environmental benefits, and an operating load constraint set is constructed. The microgrid operating costs include distributed power generation costs, electricity purchase costs, and flexible load compensation costs; the environmental benefits are determined based on the quantification of pollutant emissions from distributed power generation.

[0035] In one application example, the process of determining the comprehensive objective function includes:

[0036] (21) Based on the fluctuation range of electricity value, a dynamic weight adjustment mechanism is adopted to determine the weights corresponding to the distributed power generation cost, the electricity purchase cost, the flexible load compensation cost and the environmental benefits; by adopting the dynamic weight adjustment mechanism, the weight coefficients are dynamically adjusted according to the real-time fluctuation range of electricity price, which can balance economic and environmental protection needs.

[0037] (22) Based on the distributed power source operation data, the flexible load operation data and the energy storage data, determine the distributed power source generation cost, power purchase cost and flexible load compensation cost;

[0038] (23) Based on the distributed power generation cost and its corresponding weight, the electricity purchase cost and its corresponding weight, the flexible load compensation cost and its corresponding weight, and the environmental benefits and their corresponding weight, a weighted calculation is performed to determine the comprehensive objective function.

[0039] In a practical application, the formula for calculating the environmental benefits is as follows:

[0040]

[0041] Among them, E envFor environmental benefits (unit: kgCO2 equivalent / hour), P gem，i (t) represents the power generation (kW) of the i-th distributed power source at time t, where n is the number of distributed power sources, and η is the power generation of the i-th distributed power source. emission，i The pollutant emission coefficient for the i-th distributed power generation unit is pre-calibrated based on the equipment type and fuel type.

[0042] The formula for calculating the weights corresponding to the environmental benefits is as follows:

[0043]

[0044] Where α(t) is the weight corresponding to the environmental benefits at time t, α base Let ΔP be the initial value, which can be 0.4. price (t) represents the difference between the current electricity price and the average electricity price, γ is the adjustment coefficient, and P price,avg This represents the average electricity price.

[0045] When the weighting of environmental benefits changes, the overall objective function's emphasis on environmental benefits shifts, thus affecting the optimization results. Distributed generation costs, electricity purchase costs, and flexible load compensation costs, as components of operating costs, are also affected by the optimization of the overall objective function. Increasing the weighting of environmental benefits may encourage dispatch schemes to favor distributed generation and reduce electricity purchases, thereby impacting generation and purchase costs. Simultaneously, the regulation strategy for flexible loads may also change, affecting flexible load compensation costs.

[0046] During the optimization of dispatching, the weights of the three costs should be adjusted accordingly based on changes in real-time electricity prices, distributed generation, and flexible load control needs. Using a dynamic weight adjustment mechanism, when electricity prices are high, the cost of electricity purchase increases. To reduce the overall cost, the weight of the electricity purchase cost can be appropriately reduced, while the proportion of distributed generation and flexible load control can be increased—that is, the weights of generation costs and flexible load compensation costs can be relatively increased. Conversely, when electricity prices are low, the weight corresponding to the electricity purchase cost may increase. Environmental benefits must also be considered. If maximizing environmental benefits is prioritized, the use of distributed generation may be increased, raising the weight of generation costs, while simultaneously adjusting the weights of electricity purchase costs and flexible load compensation costs to achieve the optimal overall goal.

[0047] In one application example, the set of operating load constraints includes transferable load constraints, load reduction constraints, adjustable load constraints, power balance constraints, energy storage constraints, and distributed power source operating constraints.

[0048] The transferable load constraint is a dynamic spatiotemporal transfer constraint established based on the load transfer characteristic matrix and the time-slot shift probability factor. The load transfer characteristic matrix is ​​determined according to the power demand, time range, and priority of the flexible load. Different flexible loads have different power demands at different times. Combining this with their transferable time range, a matrix describing the load transfer characteristics can be constructed, reflecting information such as the probability of each load transferring at different times and the magnitude of the transferred power. The time-slot shift probability factor is determined based on the time range of the flexible load and user satisfaction. If a certain type of flexible load has a higher probability of transfer during a specific time period, it may be because users have greater flexibility in using it during that time period. This is related to the collected time range of the flexible load and users' habits of using the load (which can be reflected in user satisfaction parameters, etc.). Through analysis and statistics of these preprocessed data, the time-slot shift probability factor can be determined to more accurately establish the dynamic spatiotemporal transfer constraints of the transferable load.

[0049] The load reduction constraint employs fuzzy theory to quantify the load reduction amount and user satisfaction constraints through membership functions. The load reduction amount is determined based on the power demand of flexible loads. The power demand of flexible loads and user satisfaction parameters play a crucial role here. The potential load reduction amount is determined based on power demand, while the user satisfaction parameters are used to construct membership functions to measure the impact of different reduction amounts on user satisfaction, thereby achieving reasonable load reduction decisions. Fuzzy chance constraints are introduced to quantify the nonlinear relationship between load reduction and user satisfaction, improving the rationality of load regulation.

[0050] The adjustable load constraint is defined as follows: based on the priority and power demand of flexible loads, a priority shifting coefficient matrix is ​​introduced, ensuring that high-priority loads are given priority consideration in power time-shift optimization. The adjustable load constraint, by introducing the priority shifting coefficient matrix, achieves collaborative optimization of power time-shift across multiple time periods. Utilizing the priority data of flexible loads, high-priority loads are given priority in power time-shift optimization, guaranteeing their electricity demand.

[0051] The power balance constraint is defined as follows: it ensures power balance based on the generation capacity of distributed power sources, the power demand of flexible loads, and the state of charge and charging / discharging power of energy storage systems. By setting the power balance constraint, the real-time balance between distributed power sources, energy storage, purchased power, and load demand is ensured, enabling the system to maintain stable operation at all times.

[0052] The energy storage constraints are: constraints that limit the range of the state of charge and charge / discharge power of the energy storage system. In a specific application, this is expressed by the following formula:

[0053] When P ess (t) < 0.

[0054] When P ess (t) < 0.

[0055] Among them, SOC ess (t) represents the state of charge of the energy storage system at time t, P ess (t) represents the energy storage power at time t (negative values ​​indicate charging, positive values ​​indicate discharging), η charge and η discharge These are the charging efficiency and discharging efficiency, respectively, E nom Let Δt be the energy storage capacity and Δt be the time step.

[0056] The operating constraints of the distributed power sources are: constraints that limit the upper and lower limits of the power output of each distributed power source.

[0057] Step 103: Based on the set of operating load constraints, solve the comprehensive objective function to obtain the day-ahead scheduling information and intraday rolling information.

[0058] In one application example, step 103 includes: based on the set of operating load constraints, using a mixed integer second-order cone programming algorithm to solve the comprehensive objective function, and introducing a dynamic penalty factor adjustment mechanism and a parallel computing mechanism in the solution process to obtain day-ahead scheduling information and intraday rolling information.

[0059] By introducing a dynamic penalty factor adjustment mechanism and a parallel computing mechanism, the solution efficiency and convergence speed can be improved. The dynamic penalty factor adjustment mechanism is used to: when the convergence speed of the integrated objective function is lower than a preset threshold, adjust the weight coefficient of the penalty term according to the degree of constraint violation in the runtime constraint set, thereby accelerating the convergence speed and improving the solution accuracy. Specifically, when the model convergence speed is lower than the preset threshold, the weight coefficient of the constraint violation penalty term is automatically increased to accelerate the convergence speed and improve the solution accuracy.

[0060] The parallel computing mechanism utilizes the computing power of multi-core processors to distribute the computational tasks of the day-ahead scheduling phase and the intraday optimization phase into different threads for parallel computation, further improving computational efficiency. The day-ahead scheduling phase involves generating a 24-hour baseline scheduling plan based on forecast data, including energy storage charge-discharge curves and load control baselines. This phase primarily considers long-term trends and periodic changes, providing a benchmark reference for intraday scheduling. The intraday rolling optimization phase involves updating ultra-short-term forecast data every 15 minutes, handling uncertainties on both the source and load sides through a model predictive control framework and dynamic constraint relaxation techniques, and dynamically adjusting optimization parameters based on the day-ahead scheduling results. This phase primarily considers real-time fluctuations and random changes, improving the flexibility and robustness of scheduling.

[0061] During the intraday rolling optimization phase, the Kalman filter algorithm is used to update the ultra-short-term forecast data. The state equation for the Kalman filter is:

[0062] x k =Fx k-1 +Gu k +w k .

[0063] z k =Hx k +v k .

[0064] Where, x k Let u be the state vector. k To control the input, z k For the observed value, w k and v k Let F represent process noise and observation noise, G represent the state transition matrix, G represent the control input matrix, and H represent the observation matrix.

[0065] Step 104: Based on the day-ahead scheduling information and the intraday rolling information, a flexible load control instruction is issued to the microgrid through a blockchain smart contract, and real-time operation feedback information of the flexible load is collected. The blockchain smart contract has the characteristics of decentralization, immutability and traceability, which can ensure the security and reliability of the control instruction and verify the execution result.

[0066] The blockchain smart contract includes a load response verification module and a smart settlement module. The load response verification module confirms, through hash value comparison, whether the deviation between the actual execution power of the flexible load (i.e., the real-time operation feedback information of the flexible load) and the flexible load control command is less than 5%, to ensure the accurate execution of the control command; the smart settlement module automatically calculates compensation fees based on user satisfaction and actual response volume, realizing a fair and reasonable incentive mechanism.

[0067] Specifically, the real-time operation feedback information of flexible loads includes real-time power data, execution timestamps, and equipment status reported during the actual operation of the flexible load equipment; the flexible load control instructions include the control power parameters, timestamps, and key user / equipment identification information. The hash value is generated by performing an encrypted hash operation on the above data to ensure data integrity and immutability. If the real-time operation feedback information of the flexible load is consistent with the data in the flexible load control instructions, the hash values ​​are the same; if there is a deviation (power deviation exceeding 5%), the hash values ​​are different, triggering a failed verification.

[0068] Furthermore, the day-ahead scheduling information includes a 24-hour baseline plan, while the intraday rolling information includes instructions updated every 15 minutes based on the day-ahead results. The final issued flexible load control instructions are real-time instructions optimized intraday, not individual day-ahead or intraday plans. The day-ahead scheduling information serves as a baseline reference, providing initial parameters for intraday optimization. The final executed instructions are flexible load control instructions dynamically adjusted intraday; therefore, verification only examines the consistency between actual execution and real-time instructions. Whether actual execution conforms to the current flexible load control instructions is not differentiated between day-ahead and intraday phases.

[0069] In a specific application, when flexible load control instructions are issued to various device terminals of the microgrid via blockchain smart contracts, the transmission is carried out through the OPCUA protocol, and timestamps and digital signatures are used to ensure communication security, with a communication cycle of no more than 1 second.

[0070] Step 105: Combining the deep reinforcement learning emergency control strategy and the sensitivity analysis optimization strategy, a compensation optimization scheduling scheme is determined based on the real-time operation feedback information of the flexible load and the flexible load control instructions.

[0071] In one specific application, step 105 includes:

[0072] (51) Based on the real-time operation feedback information of the flexible load and the flexible load control instructions, the load response is verified and the verification result is obtained through the blockchain smart contract, intelligent settlement based on sensitivity analysis is performed, and the corresponding adjustment scheme is obtained.

[0073] Sensitivity analysis quantifies the impact of different parameter changes on system performance, providing a scientific basis for dynamic adjustments. In short, sensitivity analysis optimizes system response and allows for real-time adjustments to distributed power supply and energy storage strategies. Sensitivity analysis can help determine:

[0074] If the source-load power deviation is caused by a sudden drop in photovoltaic output, sensitivity analysis can be used to quickly identify the impact of energy storage discharge and electricity purchase on costs, and the adjustment method with the lowest cost can be selected first. If the source-load power deviation is caused by the failure of transferable loads to be transferred as planned, sensitivity analysis can be used to evaluate the relationship between load compensation costs and deviation elimination effects, and the compensation strategy can be dynamically adjusted.

[0075] Source-load power deviation refers to the real-time difference between the total power supplied by the power source and the total power demanded by the load in a microgrid, reflecting the real-time power balance of the system. If the power supplied by the power source is greater than that of the load, the excess power needs to be absorbed through energy storage charging or by selling electricity to the grid; if the power of the load is greater than that of the power source, the gap needs to be filled through energy storage discharge or by purchasing electricity from the grid.

[0076] (52) When the verification result indicates that the source-load power deviation is within the first preset deviation range, the adjustment scheme is marked as a compensation optimization scheduling scheme. Specifically, the first preset deviation range indicates that the deviation has not exceeded the limit (less than or equal to 5%).

[0077] (53) When the verification result indicates that the source-load power deviation is within the second preset deviation range, an emergency control strategy based on deep reinforcement learning is adopted to correct the scheduling scheme in real time and feed it back to the parameters of the comprehensive objective function and the parameters of the operating load constraint set, forming a closed-loop control. Specifically, the second preset deviation range indicates that the deviation exceeds the limit (greater than 5%). Through the emergency control strategy, it is possible to quickly respond to emergencies and abnormal situations, and improve the stability and reliability of the system.

[0078] The parameters fed back to the comprehensive objective function include the weight of environmental benefits, and the parameters fed back to the set of operating load constraints include the parameters of the membership function of the load reduction amount and the constraint threshold of the energy storage system's state of charge. The weight of environmental benefits is dynamically adjusted based on the real-time source-load power deviation (e.g., temporarily reducing the environmental weight when the deviation is too large, prioritizing power balance). The parameters of the fuzzy membership function of the load reduction amount are adjusted to control the user satisfaction threshold for load reduction, avoiding excessive reduction that could lead to user complaints. The constraint threshold of the energy storage system's state of charge is temporarily relaxed to quickly respond to power shortages.

[0079] In a specific application, the emergency control strategy based on deep reinforcement learning employs a dual-delay Q-learning algorithm, with the corresponding reward function being:

[0080] R(t) = -α|P imbalance (t)|-β·ΔC grid (t).

[0081] Where R(t) is the value of the reward function, P imbalance (t) represents the source-load power deviation, ΔC grid (t) represents the increment of the grid's electricity purchase cost, and α and β are weighting coefficients. The reward function comprehensively considers two optimization objectives: power deviation and electricity purchase cost. By adjusting the weighting coefficients α and β, the priority of different optimization objectives can be balanced. The weighting coefficients α and β can be dynamically adjusted according to real-time conditions to balance the optimization objectives between power deviation and electricity purchase cost. During peak electricity price periods, the weighting coefficient of β can be appropriately increased to prioritize reducing electricity purchase cost; when the power deviation is large, the weighting coefficient of α can be appropriately increased to prioritize reducing the power deviation. By dynamically adjusting the weighting coefficients, the emergency control strategy can make reasonable decisions in different scenarios.

[0082] In another exemplary embodiment, such as Figure 2 As shown, this application first collects and preprocesses multi-source data, including distributed power sources, flexible loads, and real-time electricity prices within the microgrid. Then, it establishes a comprehensive objective function with the goal of minimizing operating costs and maximizing environmental benefits. In this step, by dynamically adjusting the weight coefficients, the optimization model can make economic and environmental trade-off decisions under different electricity price levels. A dynamic penalty factor adjustment mechanism adaptively adjusts the weight coefficients of the penalty term based on the real-time situation of the model solution, ensuring good solution performance in different scenarios. The introduction of parallel computing mechanisms fully utilizes the multi-core processing capabilities of modern computers, decomposing complex computational tasks into multiple sub-tasks for parallel execution, thereby significantly shortening computation time. Next, hierarchical control and dynamic execution are implemented, issuing flexible load regulation commands and adjusting strategies. Based on the optimization results, commands are issued through blockchain smart contracts to regulate flexible loads and adjust power supply and energy storage strategies in real time. Then, it determines whether the source-load power deviation exceeds a threshold. If so, a deep reinforcement learning-based emergency regulation and correction scheduling scheme is initiated and feedback is provided, forming a closed-loop control; otherwise, the process ends. This application's method balances economics and environmental protection, rationally regulates loads, effectively handles source-load uncertainty, and enhances system stability.

[0083] Based on the same inventive concept, this application also provides a microgrid optimized dispatching system considering flexible loads for implementing the methods described above. The solution provided by this system is similar to the solution described in the above methods; therefore, the specific limitations in one or more system embodiments provided below can be found in the limitations of the methods described above, and will not be repeated here. The system includes:

[0084] The data acquisition module is used to acquire operational data of distributed power sources, flexible loads, and energy storage within the microgrid.

[0085] The objective function and constraint establishment module is used to: establish a comprehensive objective function based on the distributed power source operation data, the flexible load operation data, and the energy storage data, with the goal of minimizing the microgrid operating cost and maximizing environmental benefits, and construct a set of operating load constraints.

[0086] The objective function solving module is used to: solve the comprehensive objective function based on the set of operating load constraints to obtain the day-ahead scheduling information and intraday rolling information.

[0087] The feedback information collection module is used to: issue flexible load control instructions to the microgrid through a blockchain smart contract based on the day-ahead scheduling information and the intraday rolling information, and collect real-time operation feedback information of the flexible load.

[0088] The scheduling scheme optimization module is used to: combine deep reinforcement learning emergency control strategies and sensitivity analysis optimization strategies, and determine a compensation optimization scheduling scheme based on the real-time operation feedback information of the flexible load and the control instructions of the flexible load.

[0089] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 3 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a microgrid optimal scheduling method that considers flexible loads.

[0090] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0091] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0092] In one exemplary embodiment, a computer device is provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program to implement steps of a microgrid optimal scheduling method that takes into account flexible loads.

[0093] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0094] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0095] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0096] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0097] 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.

[0098] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A microgrid optimal scheduling method considering flexible loads, characterized in that, The method includes: Acquire operational data of distributed power sources, flexible loads, and energy storage within the microgrid; Based on the distributed power source operation data, the flexible load operation data, and the energy storage data, a comprehensive objective function is established with the goal of minimizing microgrid operating costs and maximizing environmental benefits, and an operating load constraint set is constructed. Based on the set of operating load constraints, the comprehensive objective function is solved to obtain the day-ahead scheduling information and intraday rolling information; Based on the day-ahead scheduling information and the intraday rolling information, flexible load control instructions are issued to the microgrid through a blockchain smart contract, and real-time operation feedback information of the flexible load is collected. By combining deep reinforcement learning emergency control strategies and sensitivity analysis optimization strategies, a compensation optimization scheduling scheme is determined based on the real-time operation feedback information of the flexible load and the control instructions of the flexible load.

2. The microgrid optimal scheduling method considering flexible loads according to claim 1, characterized in that, The distributed power source operation data includes the power generation capacity, power generation efficiency, and power source operation status data of each distributed power source. The flexible load operation data includes the power demand, time range, electrical value, priority, and user satisfaction parameters of each flexible load. The energy storage data includes the state of charge and charging / discharging power of each energy storage system.

3. The microgrid optimal scheduling method considering flexible loads according to claim 2, characterized in that, The operating cost of the microgrid includes the generation cost of distributed power sources, the cost of purchasing electricity, and the cost of flexible load compensation; the environmental benefits are determined based on the quantification of pollutant emissions from distributed power sources. The process of determining the comprehensive objective function includes: Based on the fluctuation range of electricity value, a dynamic weight adjustment mechanism is used to determine the weights corresponding to the distributed power generation cost, the electricity purchase cost, the flexible load compensation cost, and the environmental benefits. Based on the distributed power source operation data, the flexible load operation data, and the energy storage data, the distributed power source generation cost, electricity purchase cost, and flexible load compensation cost are determined. The comprehensive objective function is determined by weighting the distributed power generation cost and its corresponding weight, the electricity purchase cost and its corresponding weight, the flexible load compensation cost and its corresponding weight, and the environmental benefits and their corresponding weights.

4. The microgrid optimal scheduling method considering flexible loads according to claim 3, characterized in that, The formula for calculating the environmental benefits is as follows: Among them, E env For environmental benefits, P gem，i (t) represents the power generation of the i-th distributed power source at time t, where n is the number of distributed power sources, and η is the power output of the i-th distributed power source. emission，i Let be the pollutant emission coefficient of the i-th distributed power generation unit; The formula for calculating the weights corresponding to the environmental benefits is as follows: Where α(t) is the weight corresponding to the environmental benefits at time t, α base As the initial value, ΔP price (t) represents the difference between the current electricity price and the average electricity price, γ is the adjustment coefficient, and P price,avg This represents the average electricity price.

5. The microgrid optimal scheduling method considering flexible loads according to claim 2, characterized in that, The set of operating load constraints includes transferable load constraints, load reduction constraints, adjustable load constraints, power balance constraints, energy storage constraints, and distributed power source operating constraints. The transferable load constraint is a dynamic spatiotemporal transfer constraint established based on the load transfer characteristic matrix and the time period shift probability factor. The load transfer characteristic matrix is ​​determined according to the power demand, time range, and priority of the flexible load, and the time period shift probability factor is determined according to the time range of the flexible load and user satisfaction. The load reduction constraint is as follows: using fuzzy theory, the constraint between load reduction amount and user satisfaction is quantified through membership functions; the load reduction amount is determined based on the power demand of flexible loads. The adjustable load constraint is as follows: based on the priority and power demand of the flexible load, a shift priority coefficient matrix is ​​introduced, so that the high priority load is given priority consideration in the power time shift optimization. The power balance constraint is: a constraint that ensures power balance based on the power generation capacity of distributed power sources, the power demand of flexible loads, the state of charge and charging / discharging power of energy storage systems. The energy storage constraints are: constraints that limit the range of the state of charge and the charging and discharging power of the energy storage system; The operating constraints of the distributed power sources are: constraints that limit the upper and lower limits of the power output of each distributed power source.

6. The microgrid optimal scheduling method considering flexible loads according to claim 1, characterized in that, Based on the aforementioned set of operating load constraints, the comprehensive objective function is solved to obtain day-ahead scheduling information and intraday rolling information, including: Based on the set of operating load constraints, a mixed integer second-order cone programming algorithm is used to solve the comprehensive objective function. A dynamic penalty factor adjustment mechanism and a parallel computing mechanism are introduced in the solution process to obtain day-ahead scheduling information and intraday rolling information. The dynamic penalty factor adjustment mechanism is used to: when the convergence speed of solving the comprehensive objective function is lower than a preset threshold, adjust the weight coefficient of the penalty term according to the degree of constraint violation in the set of operating load constraints; The parallel computing mechanism is used to: allocate the computing tasks of the day-ahead scheduling phase and the day-ahead optimization phase to different threads for parallel computing.

7. The microgrid optimal scheduling method considering flexible loads according to claim 1, characterized in that, Combining deep reinforcement learning-based emergency control strategies and sensitivity analysis optimization strategies, and based on the real-time operation feedback information of the flexible load and the flexible load control instructions, a compensation optimization scheduling scheme is determined, including: Based on the real-time operation feedback information of the flexible load and the flexible load control instructions, the load response is verified and the verification result is obtained through the blockchain smart contract; intelligent settlement based on sensitivity analysis is performed and the corresponding adjustment plan is obtained. When the verification result indicates that the source-load power deviation is within the first preset deviation range, the adjustment scheme is marked as a compensation optimization scheduling scheme. When the verification result indicates that the source-load power deviation is within the second preset deviation range, an emergency control strategy based on deep reinforcement learning is adopted to correct the scheduling scheme in real time and feed it back to the parameters of the comprehensive objective function and the parameters of the operating load constraint set to form closed-loop control. The parameters fed back to the comprehensive objective function include the weight of environmental benefits, and the parameters fed back to the set of operating load constraints include the parameters of the membership function of the load reduction amount and the constraint threshold of the state of charge of the energy storage system.

8. The microgrid optimal scheduling method considering flexible loads according to claim 7, characterized in that, The emergency control strategy based on deep reinforcement learning employs a dual-delay Q-learning algorithm, with the corresponding reward function being: R(t)=-α|P imbalance (t)|-β·ΔC grid (t); Where R(t) is the value of the reward function, P imbalance (t) represents the source-load power deviation, ΔC grid (t) represents the incremental cost of purchasing electricity from the power grid, and α and β are weighting coefficients.

9. A microgrid optimized dispatching system considering flexible loads, characterized in that, The system includes: The data acquisition module is used to acquire operating data of distributed power sources, flexible loads, and energy storage within the microgrid. The objective function and constraint establishment module is used to: establish a comprehensive objective function based on the distributed power source operation data, the flexible load operation data, and the energy storage data, with the goal of minimizing the microgrid operating cost and maximizing environmental benefits, and construct a set of operating load constraints; The objective function solving module is used to: solve the comprehensive objective function based on the set of operating load constraints to obtain day-ahead scheduling information and intraday rolling information; The feedback information collection module is used to: issue flexible load control instructions to the microgrid through a blockchain smart contract based on the day-ahead scheduling information and the intraday rolling information, and collect real-time operation feedback information of the flexible load; The scheduling scheme optimization module is used to: combine deep reinforcement learning emergency control strategies and sensitivity analysis optimization strategies, and determine a compensation optimization scheduling scheme based on the real-time operation feedback information of the flexible load and the control instructions of the flexible load.

10. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the microgrid optimal scheduling method considering flexible loads as described in any one of claims 1-8.

Images