Self-adaptive optimization operation method, system and device for grid-connected micro-grid and storage medium

Abstract

The invention discloses a self-adaptive optimization operation method, system and equipment for a grid-connected micro-grid and a storage medium, and relates to the field of power system automation and new energy application, and the method comprises the steps: carrying out the multi-source fusion prediction based on the historical and real-time weather, equipment output, user load, industrial plan and other data; an operation scene is defined, and power deviation is identified; solving an optimal power distribution scheme by using the multi-objective decision model; after executing the instruction, quantifying subject contribution apportionment interests; acquiring output data of equipment, checking typical scene indexes, and finely adjusting power until the power reaches the standard; according to the method, the power impact on the main grid is remarkably inhibited, the new energy consumption efficiency is greatly improved, the balance and win-win situation of multi-subject benefits is realized, the adaptability and the execution effect of an operation strategy are effectively enhanced, and the operation level of the micro-grid is comprehensively improved.

Description

Technical Field

[0001] This invention relates to the fields of power system automation and new energy applications, and in particular to an adaptive optimization operation method, system, device and storage medium for grid-connected microgrids. Background Technology

[0002] With the increasing penetration rate of renewable energy, grid-connected microgrids, as a key carrier integrating distributed photovoltaic, wind power, energy storage, and diverse loads, play an important role in improving energy utilization efficiency and power supply reliability. However, in complex scenarios involving a high proportion of wind and solar renewable energy, multiple types of electrochemical energy storage systems, controllable industrial loads, and the coexistence of multiple stakeholders (such as distributed power operators, energy storage operators, and users), microgrid operation faces prominent problems such as power fluctuations impacting the main grid, persistently high wind and solar curtailment rates, and uncoordinated interests among multiple stakeholders. These issues severely restrict the safe, economical, and sustainable operation of microgrids.

[0003] Existing technologies often focus on a single dimension in defining operational status (e.g., only distinguishing between grid-connected and islanded modes), failing to comprehensively consider the relationship between the dynamic operating conditions of the main grid (peak and valley loads, real-time electricity prices) and the status of microgrid equipment (energy storage SOC, distributed power generation operating status) from a system perspective. They simply divide operating scenarios according to fixed rules. Regarding the coordination of interests among multiple stakeholders, they do not fully consider the different revenue demands and contributions of distributed power generation, energy storage, and users. Profit sharing relies on simple methods such as installed capacity or fixed subsidies, without establishing a matching mechanism between contribution and revenue. This results in rigid operating strategies that are difficult to adapt to complex dynamic operating conditions. The risks of main grid power surges (fluctuations exceeding ±8%) coexist with conflicts of interest among multiple stakeholders (energy storage revenue decline of 15%-20%, low user participation). Related optimization measures can only alleviate current problems and cannot meet the long-term operational needs of the microgrid. The strategy execution rate is less than 50%, resulting in insufficient economic efficiency and sustainability. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention provides an adaptive optimization operation method, system, device and storage medium for grid-connected microgrids.

[0005] Therefore, the technical problem solved by this invention is: how to provide an adaptive operation optimization method for grid-connected microgrids that can effectively suppress power surges in the main grid, reduce wind and solar curtailment rates, and achieve a balance of interests among multiple stakeholders.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an adaptive optimization operation method for grid-connected microgrids, comprising: Based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans, multi-source data fusion forecasting is performed. By utilizing the results of multi-source data fusion prediction, the current operating scenario of the system can be dynamically defined and power deviations can be identified. Based on the defined scenario and deviation, the optimal power allocation scheme is solved through a multi-objective decision model; Output commands are executed based on the power allocation scheme, and the contributions of each entity are quantified based on the operation results to achieve a balanced distribution of benefits. By collecting equipment output data, we calculate and verify the target index values ​​for typical scenarios, and then fine-tune the output power until the target index values ​​for typical scenarios are met.

[0007] As a preferred scheme for an adaptive optimization operation method for grid-connected microgrids, wherein: The multi-source data fusion prediction based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans includes: Using wind speed, irradiance, industrial user production plans, and residential electricity consumption peak-valley habits as input features, we can carry out wind and solar power output and load forecasting. For wind and solar power output forecasting, meteorological data and historical equipment output data are input, and the model weights are adjusted through real-time error feedback. Meteorological and equipment features are extracted, and time-series patterns are captured to obtain the total wind and solar power output forecast. For load forecasting, user type data, historical load data, and industrial production plans are input to classify the load. For each type of load, a forecast is made, and the weights are dynamically adjusted according to the proportion of each type of load to obtain the total load forecast value.

[0008] As a preferred scheme for an adaptive optimization operation method for grid-connected microgrids, wherein: The method of dynamically defining the current operating scenario of the system and identifying power deviations using the results of multi-source data fusion prediction includes: The power supply and demand deviation of the microgrid grid-connected lines is calculated using the results of multi-source data fusion prediction. The deviation is obtained by calculating the predicted output of wind and solar power and the predicted load.

[0009] As a preferred scheme for an adaptive optimization operation method for grid-connected microgrids, wherein: The method of dynamically defining the current operating scenario of the system and identifying power deviation by utilizing the results of multi-source data fusion prediction also includes: Based on the calculated power deviation value, the priority of equipment compensation resources is selected, and corresponding compensation strategies are adopted for excess and shortage scenarios according to different triggering conditions.

[0010] As a preferred scheme for an adaptive optimization operation method for grid-connected microgrids, wherein: The optimal power allocation scheme, based on the defined scenario and deviation, is solved using a multi-objective decision model, including: With the dual optimization objectives of minimizing wind and solar curtailment rates and minimizing grid power impact, a multi-stakeholder interest constraint is introduced, and the priority of objectives is dynamically balanced through weighted coefficients. Among them, the wind and solar curtailment rate is defined as the proportion of unabsorbed wind and solar power output to the total power output, and is only calculated in the case of power surplus. The grid power deviation impact is defined as the deviation between the predicted power and the planned power at the grid connection point.

[0011] The beneficial effects of this preferred technical solution are as follows: by setting dual optimization objectives and introducing multi-stakeholder interest constraints, it is possible to ensure the stable operation of the microgrid while taking into account the consumption of new energy sources and the stability of the main grid. By dynamically balancing the priority of objectives through weighted coefficients, the optimization focus can be flexibly adjusted according to different operating conditions, thereby improving the overall operating performance of the microgrid.

[0012] As a preferred scheme for an adaptive optimization operation method for grid-connected microgrids, wherein: The method for finding the optimal power allocation scheme based on the defined scenario and deviation, using a multi-objective decision model, also includes: The weighting coefficients are dynamically adjusted in priority for different time periods, and each scheme is scored accordingly. An improved particle swarm optimization algorithm is used to solve the problem by taking photovoltaic power output, wind turbine power output, energy storage charging and discharging, and adjustable load reduction as decision variables, and quickly generate equipment output commands. Set a fitness function and use a penalty coefficient to handle constraint violations; Using wind and solar curtailment rates, deviation rates, and the final revenue of each entity as target indicators, we ensure that the corresponding requirements are met, thereby deriving the optimal power allocation scheme.

[0013] The beneficial effects of this preferred technical solution are as follows: adjusting the weighting coefficients according to different time periods can better adapt to changes in the electricity market and the actual operating needs of the microgrid; the improved particle swarm optimization algorithm can quickly and accurately solve the optimal power allocation scheme, improving decision-making efficiency; setting fitness functions and target indicators can ensure that the power allocation scheme achieves multi-objective optimization while meeting various constraints, thus guaranteeing the efficient and stable operation of the microgrid and the interests of all stakeholders.

[0014] As a preferred scheme for an adaptive optimization operation method for grid-connected microgrids, wherein: The process of executing output commands based on a power allocation scheme and quantifying the contributions of each entity based on the operational results to achieve a balanced distribution of benefits includes: Output commands are executed based on the power allocation scheme, and the contribution of each entity, including distributed power sources, energy storage, users, and the grid, is quantified by the Shapley value to ensure that the contribution and benefits of each entity are matched. Calculate the total daily revenue and total daily cost, and then derive the net profit; The incremental contribution of a particular entity to the net revenue of the microgrid is calculated using the Shapley value. This ensures that the final benefits for all stakeholders are higher than those of the traditional model, and achieves a balanced distribution of benefits.

[0015] The beneficial effects of this preferred technical solution are as follows: quantifying the contributions of each entity using the Shapley value enables fair and reasonable distribution of benefits, ensuring that the contributions and benefits of each entity match, and increasing the enthusiasm of each entity to participate in the operation of the microgrid; calculating the benefits and costs provides a clear understanding of the economic operation of the microgrid; and ensuring that the final benefits of each entity are improved is conducive to promoting the sustainable development of the microgrid and cooperation among the entities.

[0016] Secondly, the present invention provides an adaptive optimization operation system for grid-connected microgrids, comprising: The multi-source data fusion prediction module is used to perform multi-source data fusion prediction based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans. The operating scenario definition and deviation identification module is used to dynamically define the current operating scenario of the system and identify power deviations by utilizing the results of multi-source data fusion prediction. The optimal power allocation scheme solution module is used to solve for the optimal power allocation scheme based on the defined scenario and deviation through a multi-objective decision model. The instruction execution and benefit sharing module is used to execute output instructions based on the power allocation scheme, and to quantify the contributions of each entity based on the operation results, and to distribute benefits in a balanced manner. The typical scenario index verification and power fine-tuning module is used to calculate and verify the target index values ​​for typical scenarios by collecting equipment output data, and to fine-tune the output power until the target index values ​​for typical scenarios are met.

[0017] Thirdly, the present invention provides a computer device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of an adaptive optimization operation method for grid-connected microgrids.

[0018] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of an adaptive optimization operation method for a grid-connected microgrid.

[0019] The beneficial effects of this invention are as follows: By employing a hierarchical dynamic scenario definition and power compensation priority strategy, combined with dual-objective collaborative decision-making, this invention controls the grid connection point power deviation rate to ≤3%, far superior to the ±8% fluctuation level of existing technologies. This effectively reduces the impact risk of microgrids on the main grid and improves the safety and stability of grid-connected operation. Furthermore, relying on a CNN-LSTM hybrid prediction model (24h output / load prediction error ≤6%) and K-means+XGBoost load classification prediction, coupled with a weighted coefficient strategy (λ=0.6) for prioritizing grid absorption during off-peak hours, the wind and solar curtailment rate is controlled to ≤3%, significantly reducing renewable energy waste and improving energy utilization. Efficiency: By quantifying the contribution of each entity through the Shapley value, a benefit-sharing mechanism that "matches contribution with benefit" is established, ensuring that the benefits of entities such as distributed power sources, energy storage operators, and power users are increased by ≥10% compared to the traditional model. This solves the problems of low user participation and declining energy storage benefits caused by the uneven distribution of benefits in the original model, and enhances the sustainability of microgrid operation. Breaking the traditional fixed scenario division model, the operation strategy is dynamically adjusted by combining the main grid operating conditions and the microgrid equipment status. With the fast response of closed-loop feedback and improved PSO algorithm, the strategy execution rate is increased from less than 50% to more than 85%, which can flexibly adapt to complex scenarios with high proportion of wind and solar power, multiple equipment types, and multiple stakeholders. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is an overall flowchart of an adaptive optimization operation method for grid-connected microgrids provided by the present invention. Detailed Implementation

[0022] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0023] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides an adaptive optimization operation method for grid-connected microgrids, including: S1: Based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans, multi-source data fusion forecasting is performed; S2: Utilize the results of multi-source data fusion prediction to dynamically define the current operating scenario of the system and identify power deviations; S3: Based on the defined scenario and deviation, the optimal power allocation scheme is solved through a multi-objective decision model; S4: Execute output commands based on the power allocation scheme, and quantify the contributions of each entity based on the operation results to achieve a balanced distribution of benefits; S5: By collecting the power output data of the equipment, calculate and verify the target index values ​​for typical scenarios, and fine-tune the power output until the target index values ​​for typical scenarios are met.

[0024] It should be noted that, through steps S1-S5, the adaptive optimization operation method of the present invention realizes the effective utilization of multi-source data of grid-connected microgrids, can dynamically adapt to different operating scenarios, accurately identify power deviations and solve the optimal power allocation scheme, reasonably distribute the main interests, and at the same time ensure that the system operation meets the objectives of typical scenarios through index verification and power fine-tuning, thereby improving the stability, economy and reliability of microgrid operation.

[0025] Example 2, refer to Figure 1 Table 1 illustrates one embodiment of the present invention, providing a method based on the previous embodiment, comprising: In this embodiment, the multi-source data fusion prediction based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans in step S1 above includes: Wind speed, irradiance (meteorological data), industrial user production plans, and residential electricity consumption peak-valley habits (load curves) are used as input features to conduct wind and solar power output and load forecasting.

[0026] Specifically, the wind and solar power output forecast includes: Input data: Meteorological data (wind speed) Irradiance G Ambient temperature T Historical power output data of equipment ; A CNN-LSTM hybrid model is used to adjust the model weights through real-time error feedback, as shown in the following formula: + =1 in, This is the predicted total power output of wind and solar power. Output for CNN model (extracting meteorological and equipment features). Output for LSTM model (capturing temporal patterns). These represent the real-time prediction errors of the two types of models. They are making real contributions to the scenic area.

[0027] In another possible implementation, the random forest algorithm can be used for wind and solar power output prediction. Random forest is an ensemble learning model composed of multiple decision trees; it can handle high-dimensional data and has good resistance to overfitting. Meteorological data (such as wind speed and irradiance) and historical power output data of the equipment are used as input features to train the random forest model. During prediction, the model integrates the results of multiple decision trees and outputs a predicted total wind and solar power output.

[0028] In another possible implementation, a Kalman filter-based prediction method can be used for wind and solar power output prediction. Kalman filtering is a recursive optimal estimation algorithm that is highly effective for handling noisy dynamic systems. First, a state-space model of wind and solar power output is established, and meteorological data and historical equipment output data are input as observations into the Kalman filter algorithm. The algorithm then recursively calculates the predicted wind and solar power output for the current moment based on the current observations and the state estimate from the previous moment.

[0029] Load forecasting includes: Input data: User type data (industrial / residential), historical load data, industrial production plan; First, the load is divided into three categories: "stable industrial load", "flexible and adjustable load", and "residential peak-valley load" using K-means. Then, XGBoost is used to predict the load for each category, and the weights of each category are dynamically adjusted. , , (The sum is 1), the formula is as follows: + Among them, This is the total load forecast value. This is the industrial load forecast value. This is a flexible and adjustable load forecast value. This is the predicted residential load. The weighting is based on the proportion of industrial load.

[0030] In another possible implementation, multi-source data fusion prediction can also employ a combination of Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs) from deep learning. First, CNNs are used to extract features from multi-source data, including historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans. This is because CNNs excel at handling data with a grid structure and can effectively extract local features. Then, the extracted features are input into an LSTM, which can capture the temporal information of the data, thereby achieving accurate predictions of wind and solar power output and load.

[0031] In another possible implementation, multi-source data fusion prediction can also employ a combination of fuzzy logic systems and support vector machines (SVMs). Fuzzy logic systems can handle the uncertainty and fuzziness in multi-source data, fuzzifying the data to obtain fuzzy membership degrees for different data features. The fuzzy results are then input into the SVM, which has strong generalization capabilities and can build a prediction model based on the input data, thus achieving multi-source data fusion prediction.

[0032] In this embodiment, the step S2 above, which uses the results of multi-source data fusion prediction to dynamically define the current operating scenario of the system and identify power deviation, includes: The power deviation, i.e. the power supply and demand deviation of the microgrid grid-connected lines, is calculated using the results of multi-source data fusion prediction. , is represented as: in, Contribute to photovoltaic forecasting, To predict the output of wind turbines, For load forecasting.

[0033] according to The value determines the priority of equipment compensation resources (energy storage, controllable load, and main grid interaction), and the rules are shown in Table 1: Table 1 Equipment Power Compensation Priority Strategy

[0034] In another possible implementation, priority selection can be based on the response speed and cost of the equipment. Equipment with fast response and low cost, such as small energy storage devices, is given higher priority. When power deviations occur, these devices are prioritized for compensation. Conversely, equipment with slow response and high cost, such as large power generation equipment, is only activated when smaller devices cannot meet the compensation requirements.

[0035] In another possible implementation, priority can be given based on the remaining capacity and lifespan of the equipment. Equipment with large remaining capacity and long lifespan is prioritized as compensation resources. For example, for energy storage devices, those with large remaining power and few charge / discharge cycles are prioritized. This can meet power compensation needs while extending the lifespan of the equipment and reducing overall operating costs.

[0036] In this embodiment, step S3 above, based on the defined scenario and deviation, solves for the optimal power allocation scheme through a multi-objective decision model, including: With the dual optimization objectives of minimizing wind and solar curtailment rates and minimizing power impact on the main grid, a multi-stakeholder interest constraint is introduced, and the priority of objectives is dynamically balanced through weighted coefficients.

[0037] Among them, the wind and solar curtailment rate γ is defined as the proportion of unconsumed wind and solar power output to the total power output, and the formula is as follows: in, (t) represents the wind's contribution at time t. Let t be the load at time t. The energy storage charging power (>0 indicates charging). To ensure the power supplied to the main grid, max[0,]: ensures that only the curtailment of wind and solar power in the "excess power output" scenario is calculated (no curtailment when power output is insufficient), and the target value of γ is ≤5%.

[0038] Main grid power deviation impact Defined as the deviation between the predicted power and the planned power at the grid connection point, the target deviation rate must be ≤3%, as shown in the following formula: in, , For energy storage discharge power, The power output at time t, which was determined recently, is the planned power output for the main grid interaction and is issued by the power grid company.

[0039] By dynamically adjusting the weighting coefficients (Values ​​0-1) Assign priority and score each solution: For wind and solar curtailment rates, The deviation rate between the predicted curve and the planned curve.

[0040] Mainnet peak =0.4, prioritize suppressing main network impacts (avoiding main network overload); Mainnet Valley Time =0.6, prioritize reducing wind and solar curtailment rates (improving the consumption of new energy). Mainnet flat segment =0.5, dual-objective equilibrium optimization.

[0041] Furthermore, an improved particle swarm optimization (PSO) algorithm is used during the solution process to quickly generate equipment output commands, ensuring both response accuracy and speed.

[0042] Decision variable: Photovoltaic power output Photovoltaic power output Energy storage charging and discharging Adjustable load reduction .

[0043] Fitness function: ( =100 is the penalty coefficient) Target indicators: Wind and solar curtailment rate γ ≤ 5%. Deviation rate target value must be ≤ 3%, and the final revenue of each entity must be ≥ 10% higher than the traditional model. In this embodiment, step S4 above, which involves executing output commands based on a power allocation scheme and quantifying the contributions of each entity based on the operational results to achieve a balanced distribution of benefits, includes: The contribution of each entity (distributed power generation, energy storage, users, and the grid) is quantified using the Shapley Value to ensure that contribution matches benefit, with the following constraints: Daily Total Revenue : in, Main grid consumption subsidy (RMB / MWh) Revenue from grid connection (RMB / MWh, calculated based on grid connection electricity price). For ancillary service revenue (RMB / MWh, such as peak shaving subsidies).

[0044] Daily total cost : in, The cost of operation and maintenance of distributed power sources (RMB / MWh). Cost of energy storage charging and discharging losses (yuan / MWh, calculated based on the number of charge and discharge cycles, e.g., 0.1 yuan / kWh). Cost of purchasing electricity from the main grid (RMB / MWh, calculated based on the real-time electricity price of the main grid); Net income (10,000 yuan) The incremental contribution of a particular entity to the net revenue of a microgrid is quantified using the Shapley value, as shown in the following formula: in, The main types are (distributed power supply = 1, energy storage = 2, users = 3). For all subjects (3 categories in total); For without a main body subsets of (e.g.) When the distribution of power sources and energy storage is equal to {distributed power sources and energy storage}, calculate the incremental revenue after adding "users". Subset Add main body Net income after the transaction (ten thousand yuan). For subset Net income (ten thousand yuan).

[0045] The final returns for each entity must be at least 10% higher than in the traditional model to ensure a balance of interests. in, as the main body The final profit (ten thousand yuan). The main body in the traditional model The revenue (in ten thousand yuan).

[0046] In another possible implementation, a cost-benefit analysis method can be used for balanced benefit allocation. First, a detailed calculation of all costs incurred during microgrid operation is performed, including equipment investment costs, operation and maintenance costs, and energy procurement costs. Simultaneously, various benefits are calculated, such as electricity sales revenue and additional revenue from reduced wind and solar curtailment. Then, based on each entity's cost-sharing ratio and contribution to benefits, the benefit allocation for each entity is determined, thereby achieving a balanced benefit allocation.

[0047] In another possible implementation, blockchain technology can be introduced for balanced benefit distribution. Utilizing the distributed ledger and smart contract features of blockchain, the contributions and revenue distribution rules of each entity are recorded on the blockchain. When the microgrid generates revenue, the smart contract automatically distributes the benefits according to the contributions of each entity based on pre-set rules, ensuring transparency and fairness in the distribution process and achieving balanced benefit distribution.

[0048] In this embodiment, step S5 above involves collecting equipment output data, calculating and verifying the target index value for a typical scenario, and fine-tuning the output power until the target index value for the typical scenario is met, including: The microgrid central controller (MGCC) sends output commands to the controllers of each device for execution via the Modbus-TCP / CAN bus; The system collects actual execution data such as actual wind and solar power output, actual load value, energy storage charging and discharging power, and main grid interaction power through hardware devices such as power sensors, smart meters, and energy storage management systems. Calculate the wind and solar curtailment rates and the main grid power deviation rate, and verify whether they meet the targets of ≤5% and ≤3% respectively; If the verification does not meet the target indicators, the output setpoint of the photovoltaic inverter, the pitch control system of the wind turbine, the charging and discharging power command of the energy storage converter and the variable load controller command are verified and fine-tuned by improving the particle swarm optimization algorithm or the direct control strategy. The fine-tuned output command is re-executed, and data is collected again to verify the indicators, forming a closed-loop feedback until the target indicator value meets the requirements. If the verification meets the target indicators, the actual operating data, including the prediction model weights, decision weighting coefficients, optimal equipment output, load power, and adjustable load power, will be used for model optimization in subsequent cycles.

[0049] Example 3: The above is an illustrative scheme of an adaptive optimization operation method for a grid-connected microgrid according to this embodiment. It should be noted that the technical solution of an adaptive optimization operation system for a grid-connected microgrid and the technical solution of the adaptive optimization operation method for a grid-connected microgrid described above belong to the same concept. Details not described in detail in the technical solution of the adaptive optimization operation system for a grid-connected microgrid in this embodiment can be found in the description of the technical solution of the adaptive optimization operation method for a grid-connected microgrid described above.

[0050] This embodiment also provides an adaptive optimization operation system for grid-connected microgrids, including: The multi-source data fusion prediction module is used to perform multi-source data fusion prediction based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans. The operating scenario definition and deviation identification module is used to dynamically define the current operating scenario of the system and identify power deviations by utilizing the results of multi-source data fusion prediction. The optimal power allocation scheme solution module is used to solve for the optimal power allocation scheme based on the defined scenario and deviation through a multi-objective decision model. The instruction execution and benefit sharing module is used to execute output instructions based on the power allocation scheme, and to quantify the contributions of each entity based on the operation results, and to distribute benefits in a balanced manner. The typical scenario index verification and power fine-tuning module is used to calculate and verify the target index values ​​for typical scenarios by collecting equipment output data, and to fine-tune the output power until the target index values ​​for typical scenarios are met.

[0051] This embodiment also provides an electronic device applicable to an adaptive optimization operation method for grid-connected microgrids, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement an adaptive optimization operation method for grid-connected microgrids as proposed in the above embodiments.

[0052] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements an adaptive optimization operation method for grid-connected microgrids as proposed in the above embodiments.

[0053] The storage medium proposed in this embodiment belongs to the same inventive concept as the adaptive optimization operation method for grid-connected microgrids proposed in the above embodiment. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0054] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An adaptive optimization operation method for grid-connected microgrids, characterized in that, include: Based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans, multi-source data fusion forecasting is performed. By utilizing the results of multi-source data fusion prediction, the current operating scenario of the system can be dynamically defined and power deviations can be identified. Based on the defined scenario and deviation, the optimal power allocation scheme is solved through a multi-objective decision model; Output commands are executed based on the power allocation scheme, and the contributions of each entity are quantified based on the operation results to achieve a balanced distribution of benefits. By collecting equipment output data, we calculate and verify the target index values ​​for typical scenarios, and then fine-tune the output power until the target index values ​​for typical scenarios are met.

2. The adaptive optimization operation method for grid-connected microgrids as described in claim 1, characterized in that, The multi-source data fusion prediction based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans includes: Using wind speed, irradiance, industrial user production plans, and residential electricity consumption peak-valley habits as input features, we can carry out wind and solar power output and load forecasting. For wind and solar power output forecasting, meteorological data and historical equipment output data are input, and the model weights are adjusted through real-time error feedback. Meteorological and equipment features are extracted, and time-series patterns are captured to obtain the total wind and solar power output forecast. For load forecasting, user type data, historical load data, and industrial production plans are input to classify the load. For each type of load, a forecast is made, and the weights are dynamically adjusted according to the proportion of each type of load to obtain the total load forecast value.

3. The adaptive optimization operation method for grid-connected microgrids as described in claim 2, characterized in that, The method of dynamically defining the current operating scenario of the system and identifying power deviations using the results of multi-source data fusion prediction includes: The power supply and demand deviation of the microgrid grid-connected lines is calculated using the results of multi-source data fusion prediction. The deviation is obtained by calculating the predicted output of wind and solar power and the predicted load.

4. The adaptive optimization operation method for grid-connected microgrids as described in claim 3, characterized in that, The method of dynamically defining the current operating scenario of the system and identifying power deviation by utilizing the results of multi-source data fusion prediction also includes: Based on the calculated power deviation value, the priority of equipment compensation resources is selected, and corresponding compensation strategies are adopted for excess and shortage scenarios according to different triggering conditions.

5. The adaptive optimization operation method for grid-connected microgrids as described in claim 4, characterized in that, The optimal power allocation scheme, based on the defined scenario and deviation, is solved using a multi-objective decision model, including: With the dual optimization objectives of minimizing wind and solar curtailment rates and minimizing grid power impact, a multi-stakeholder interest constraint is introduced, and the priority of objectives is dynamically balanced through weighted coefficients. Among them, the wind and solar curtailment rate is defined as the proportion of unabsorbed wind and solar power output to the total power output, and is only calculated in the case of power surplus. The grid power deviation impact is defined as the deviation between the predicted power and the planned power at the grid connection point.

6. The adaptive optimization operation method for grid-connected microgrids as described in claim 5, characterized in that, The method for finding the optimal power allocation scheme based on the defined scenario and deviation, using a multi-objective decision model, also includes: The weighting coefficients are dynamically adjusted in priority for different time periods, and each scheme is scored accordingly. An improved particle swarm optimization algorithm is used to solve the problem by taking photovoltaic power output, wind turbine power output, energy storage charging and discharging, and adjustable load reduction as decision variables, and quickly generate equipment output commands. Set a fitness function and use a penalty coefficient to handle constraint violations; Using wind and solar curtailment rates, deviation rates, and the final revenue of each entity as target indicators, we ensure that the corresponding requirements are met, thereby deriving the optimal power allocation scheme.

7. The adaptive optimization operation method for grid-connected microgrids as described in claim 6, characterized in that, The process of executing output commands based on a power allocation scheme and quantifying the contributions of each entity based on the operational results to achieve a balanced distribution of benefits includes: Output commands are executed based on the power allocation scheme, and the contribution of each entity, including distributed power sources, energy storage, users, and the grid, is quantified by the Shapley value to ensure that the contribution and benefits of each entity are matched. Calculate the total daily revenue and total daily cost, and then derive the net profit; The incremental contribution of a particular entity to the net revenue of the microgrid is calculated using the Shapley value. This ensures that the final benefits for all stakeholders are higher than those of the traditional model, and achieves a balanced distribution of benefits.

8. An adaptive optimization operation system for grid-connected microgrids, employing the method described in any one of claims 1 to 7, characterized in that, include: The multi-source data fusion prediction module is used to perform multi-source data fusion prediction based on historical and real-time meteorological data, historical equipment output data, user type and historical load data, and industrial production plans. The operating scenario definition and deviation identification module is used to dynamically define the current operating scenario of the system and identify power deviations by utilizing the results of multi-source data fusion prediction. The optimal power allocation scheme solution module is used to solve for the optimal power allocation scheme based on the defined scenario and deviation through a multi-objective decision model. The instruction execution and benefit sharing module is used to execute output instructions based on the power allocation scheme, and to quantify the contributions of each entity based on the operation results, and to distribute benefits in a balanced manner. The typical scenario index verification and power fine-tuning module is used to calculate and verify the target index values ​​for typical scenarios by collecting equipment output data, and to fine-tune the output power until the target index values ​​for typical scenarios are met.

9. An electronic device, characterized in that, include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores computer-executable instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images