A real-time coordinated dispatching method and system for main and distribution microgrids

Abstract

The present invention discloses a method and system for real-time coordinated scheduling of a main-distribution microgrid, and relates to the technical field of a main-distribution microgrid. The method comprises the following steps: making scheduling decisions based on dynamic programming under a preset energy window size; obtaining microgrid scheduling data and resource configuration data; performing data analysis on the microgrid scheduling data and resource configuration data based on a BP neural network algorithm, and constructing an energy window size influence model; performing arithmetic processing based on a preset energy window size, and respectively obtaining corresponding energy window size influence models; constructing an energy window size influence fluctuation curve, and performing data analysis on the energy window size influence fluctuation curve to obtain an optimized energy window size, and applying the energy window size to the real-time coordinated scheduling of the main-distribution microgrid. The present application solves the problem that it is difficult to balance flexible response and computational efficiency in real-time coordinated scheduling of the main-distribution microgrid.

Description

Technical Field

[0001] The present invention relates to the technical field of main distribution microgrids, and more specifically, to a main distribution microgrid real-time coordinated scheduling method and system. Background Art

[0002] The main distribution microgrid refers to the integrated system of the main grid, distribution grid and microgrid. The main grid usually refers to the large-scale power transmission and distribution network, the distribution grid refers to the medium-voltage or low-voltage grid part, and the microgrid refers to a small-scale grid system that can manage itself. This part shows that the method is designed to be used for the coordination and linkage between the main grid, distribution grid and microgrid, so as to realize the integrated management of multi-level grids.

[0003] A typical existing main-distribution-microgrid coordinated dispatching method adopts a hierarchical optimization structure, dividing the main grid, distribution grid and microgrid into different levels. The main grid layer is responsible for global power distribution and stability, the distribution grid layer mainly manages the load distribution of medium and low voltage power grids, and the microgrid layer uses distributed energy and energy storage systems to meet its own needs, and requests support from the distribution grid layer through information interfaces when necessary. Through load data collection and distributed optimization algorithms, each level coordinates while maintaining independence to form an overall optimization. However, this method has shortcomings in real-time performance, coordination complexity and computational complexity, especially in dealing with rapidly changing power demand and distributed energy output, it is difficult to achieve accurate and efficient real-time dispatch.

[0004] For example, the invention patent announcement with announcement number: CN105552888B announces a distribution network simulation method that considers the interaction between the main and distribution networks. The initial state is determined by the integrated flow calculation of the main and distribution networks, and reconstruction is performed on the distribution side according to the simulation settings. During the reconstruction process, whether the safety constraints of the main network are met is considered, and the main network operation state is adjusted according to the main network safety constraints and the distribution network reconstruction strategy is re-simulated until the simulation results that meet the main network operation constraints are obtained, thereby realizing the distribution network operation simulation that considers the interaction between the main and distribution networks. The method proposed in the present invention can consider the support capacity of the main network to the distribution network when simulating the distribution network operation state, and effectively improve the authenticity and accuracy of the research on the interaction between the main and distribution networks and the simulation of the active distribution network operation.

[0005] For example, the invention patent with announcement number: CN112910002B announces a method and device for smoothly switching the operating mode of the adaptive main power supply of a microgrid, including a microgrid coordination controller that determines whether the main power supply is a storage energy or a diesel generator according to the current load demand when switching from grid-connected to island-connected, and automatically starts the diesel generator if the diesel generator is the main power supply, and limits the output of new energy, reduces the load and controls the power of the interconnection line according to the system operation status. When the power of the interconnection line is within the set value range, the main power supply operation mode is switched and the grid connection point switch is disconnected to complete the smooth switching from grid-connected to island-connected; when switching from island-connected to grid-connected, the microgrid coordination controller adopts a control strategy combining automatic synchronization with multiple synchronizations to control the current main power supply to adjust the voltage on the microgrid side to be consistent with the system side, increase or decrease the frequency on the microgrid side according to the system side frequency and the phase angle difference between the microgrid side and the system side, and the grid connection point switch is synchronously closed after the synchronous closing conditions are met to complete the smooth switching from island-connected to grid-connected.

[0006] The above disclosed technical solutions have at least the following technical problems: The energy dispatch window size plays an important role in the real-time coordinated dispatch of the main distribution microgrid, which defines the scope of energy distribution and dispatch within the corresponding time period. Selecting the appropriate dispatch window size can significantly affect the operating efficiency and responsiveness of the microgrid;

[0007] A larger dispatch window allows the system to comprehensively consider load forecasts and renewable energy generation capacity over a longer period of time, thereby achieving more optimized resource allocation, which helps reduce the frequency of dispatch decisions and reduce computing costs. However, this setting may result in a slower response speed and fail to adapt to rapidly changing load demands and emergencies in a timely manner.

[0008] Smaller dispatch windows can improve the system's ability to respond to real-time changes, making dispatch decisions more flexible and able to quickly adjust to cope with load fluctuations and uncertainties in renewable energy. However, this also means that more frequent calculations and decisions are required, which may increase the system's computational burden and may ignore the impact of long-term trends.

[0009] The existing energy scheduling window size setting often fails to balance flexible response and computational efficiency, resulting in failure to make timely adjustments under rapidly changing load demands, or neglecting short-term fluctuations under larger windows, thus reducing the operating efficiency of the microgrid. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0010] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for solving the problem that it is difficult to balance flexible response and computing efficiency in real-time coordinated scheduling of main and distribution microgrids by analyzing the size of the energy scheduling window.

[0011] To achieve the above object, the present invention provides the following technical solutions:

[0012] A method for real-time coordinated dispatching of a main distribution microgrid comprises the following steps: obtaining load data and historical load data of the main distribution microgrid, performing load forecasting based on variational multimodality, and performing dispatching decisions under a preset energy window size based on dynamic programming; obtaining microgrid dispatching data and resource configuration data corresponding to the H time and B time interval during real-time coordinated dispatching of the main distribution microgrid; performing data analysis on the microgrid dispatching data and resource configuration data based on a bp neural network algorithm, and constructing an energy window size influence model; performing arithmetic processing based on a preset energy window size to obtain a number of different energy window sizes, and respectively obtaining corresponding energy window size influence models; constructing an energy window size influence fluctuation curve according to the outputs of the several different energy window sizes and the corresponding energy window size influence models, and performing data analysis on the energy window size influence fluctuation curve to obtain an optimized energy window size, and applying it to the real-time coordinated dispatching of the main distribution microgrid.

[0013] In a preferred embodiment, the load data includes load demand, renewable energy generation data, and energy storage status; the load forecasting based on variational multimodality and the scheduling decision based on dynamic programming within a preset energy window size are specifically: using variational multimodal technology to process load data, modeling different load modes, and capturing corresponding load characteristics; extracting time series features from historical load data, wherein the time series features include seasonal characteristics, trend characteristics, and periodic characteristics; integrating and iterating the time series features and predicted load characteristics based on variational multimodal technology to obtain a load forecasting model for load forecasting; within the preset energy window size, based on the current load forecast value, the availability of power generation resources, and the energy storage status, with the goal of balancing load demand and power generation capacity and minimizing operating costs, gradually determine the optimal scheduling strategy for each time point.

[0014] In a preferred embodiment, the microgrid dispatching data includes a dispatching flexibility evaluation coefficient and a power quality evaluation coefficient. The specific analysis method of the dispatching flexibility evaluation coefficient is as follows: obtain a number of dispatching response rate data of the dispatching decision during the real-time coordinated dispatching of the main and distribution microgrids, and perform a conversion from time series to frequency domain to obtain dispatching response fluctuation frequency data; perform feature extraction based on the dispatching response fluctuation frequency data to obtain response characteristics, and the response characteristics include response overlap characteristics, response range characteristics and frequency characteristics; the response overlap characteristics include overlapping peak values ​​and overlapping frequency differences; the response range characteristics include the proportion of dispatching resources; the frequency characteristics include frequency fluctuation values; according to the response characteristics and the conventional dispatching range, the difference between the calculated response characteristics and the conventional dispatching range based on the preset dispatching flexibility evaluation coefficient calculation formula is used as the dispatching flexibility evaluation coefficient.

[0015] In a preferred embodiment, the resource configuration data includes a resource configuration efficiency impact coefficient, and the specific method for obtaining the power quality assessment coefficient is as follows: The method for obtaining the power quality assessment coefficient is as follows: obtain the load changes, power generation capacity, and usage of power electronic equipment of the main distribution microgrid, and construct a topological structure through simulation software; cluster the topological structure; extract features from the processed topological structure, and calculate the power intensity assessment coefficient based on the principle of energy constancy, energy loss coefficient and constraints; simulate the scheduling of the main distribution microgrid through the topological structure, and establish a power quality model to obtain the power loss assessment coefficient; calculate the power quality assessment coefficient in combination with the power intensity assessment coefficient.

[0016] In a preferred embodiment, the specific method for obtaining the resource configuration efficiency influence coefficient is as follows: obtain a resource configuration set, the resource configuration set including resource proportion, resource configuration distribution and configuration characteristics; calculate the efficiency of each resource configuration in the resource configuration set based on the resource configuration distribution and the corresponding coefficient of variation; for each resource configuration scheme, count the number of occurrences of the corresponding configuration characteristics to obtain the resource configuration scheme characteristic frequency; calculate the standard deviation and mean value of the resource configuration scheme characteristic frequency, and calculate the resource configuration efficiency influence coefficient in combination with the efficiency.

[0017] In a preferred embodiment, the energy window size influence fluctuation curve is constructed according to the output of several different energy window sizes and the corresponding energy window size influence model, specifically: the energy window size influence coefficient under different energy window sizes is obtained from the energy window size influence model; several energy window sizes and energy window size influence coefficients are connected to construct the energy window size influence fluctuation curve.

[0018] In a preferred embodiment, the energy window size influence fluctuation curve is subjected to data analysis to obtain an optimized energy window size, and the optimized energy window size is applied to the real-time coordinated scheduling of the main distribution microgrid, specifically: obtaining the maximum values ​​of several energy window size influence fluctuation curves; performing secondary derivative on each maximum value respectively; taking the energy window size corresponding to the maximum value whose secondary derivative value is closest to zero as the optimized energy window size, and applying it to the real-time coordinated scheduling of the main distribution microgrid.

[0019] A system for a real-time coordinated dispatching method of a main distribution microgrid, characterized in that it includes a prediction module, a data acquisition module, a window size analysis module, a window size setting module and a screening module, and there are connections between the modules; the prediction module is used to obtain the load data and historical load data of the main distribution microgrid, perform load prediction based on variational multimodality, and perform dispatching decisions under a preset energy window size based on dynamic programming; the data acquisition module is used to obtain the microgrid dispatching data and resource configuration data corresponding to the H time and B time interval during the real-time coordinated dispatching of the main distribution microgrid; the window size analysis module is used to perform data analysis on the microgrid dispatching data and resource configuration data based on the bp neural network algorithm, and construct an energy window size influence model; the window size setting module is used to perform arithmetic processing based on the preset energy window size to obtain several different energy window sizes, and respectively obtain the corresponding energy window size influence models; the screening module is used to construct an energy window size influence fluctuation curve according to the output of several different energy window sizes and the corresponding energy window size influence model, and perform data analysis on the energy window size influence fluctuation curve to obtain the optimized energy window size, and apply it to the real-time coordinated dispatching of the main distribution microgrid.

[0020] The technical effects and advantages of the real-time coordinated dispatching method of the main distribution microgrid of the present invention are as follows:

[0021] 1. The present invention improves the system's ability to respond to load fluctuations and uncertainties in renewable energy generation through load forecasting based on variational multimodal technology, making the forecast more accurate and ensuring that electricity demand is effectively met in a changing environment. Secondly, through a dynamic programming algorithm, the optimal scheduling strategy at each time point is gradually determined to maximize the balance between load demand and power generation capacity, significantly reducing operating costs. In addition, the introduction of a scheduling flexibility evaluation coefficient provides a quantitative basis for the impact of the scheduling window size on system flexibility and computational efficiency, helping operators make timely decisions and optimize resource allocation in a rapidly changing power market. At the same time, the use of a power quality evaluation coefficient ensures that power quality is maintained while improving response speed, reducing the risk of failure and enhancing system stability. In short, this method achieves efficient, stable and economical operation of a microgrid by comprehensively considering multiple factors such as load demand, power generation capacity and power quality, and has broad application prospects.

[0022] 2. The present invention conducts in-depth analysis of microgrid dispatching data and resource allocation data through the BP neural network algorithm, and constructs an energy window size impact model. This model not only evaluates the comprehensive impact of dispatching flexibility, power quality and resource allocation efficiency, but also generates an energy window size impact coefficient, thereby providing a scientific basis for the assessment of traffic shaping hazards. Through the arithmetic processing of different energy window sizes and the analysis of model output, an energy window size impact fluctuation curve is finally constructed. The optimization analysis of this fluctuation curve can effectively identify the optimal energy window size, ensuring that the real-time coordinated dispatching of the microgrid is more efficient and flexible. This process improves the accuracy of dispatching decisions, reduces resource waste, optimizes overall energy management, and has stronger adaptability and responsiveness, thereby improving the operating efficiency and reliability of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention is a structural schematic diagram of a method for real-time coordinated dispatching of a main distribution microgrid.

[0024] Figure 2 This is a structural schematic diagram of a main-distribution microgrid real-time coordinated dispatching system of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Embodiment 1, Figure 1 The present invention provides a real-time coordinated dispatching method for a main distribution microgrid, comprising the following steps:

[0027] S1, obtain the load data and historical load data of the main distribution microgrid, perform load forecasting based on variational multimodality, and make scheduling decisions under the preset energy window size based on dynamic programming.

[0028] In this embodiment, the load data includes load demand, renewable energy generation data, and energy storage status;

[0029] The load forecasting based on variational multimodality and the scheduling decision based on dynamic programming under a preset energy window size are specifically: using variational multimodality technology to process load data, modeling different load modes, and capturing corresponding load characteristics;

[0030] Extracting time series features from historical load data, wherein the time series features include seasonal features, trend features, and periodic features;

[0031] Based on variational multimodal technology, the time series characteristics and predicted load characteristics are integrated and iterated to obtain the load forecasting model for load forecasting;

[0032] Within the preset energy window size, based on the current load forecast value, the availability of power generation resources and the energy storage status, the optimal scheduling strategy for each time point is gradually determined with the goal of balancing load demand and power generation capacity and minimizing operating costs.

[0033] It should be noted that variational multimodality for load forecasting is an advanced machine learning technology that aims to improve the accuracy of forecasting by integrating the characteristics of multiple load modes. This method first extracts multiple time series features from historical load data, such as seasonality, trend, and periodicity, and then uses variational inference technology to model these features; by jointly modeling different load modes, variational multimodality can capture the complex changes and diverse characteristics of load demand, thereby achieving accurate forecasting of future loads. This method is particularly suitable for dealing with the uncertainty brought about by fluctuations in renewable energy generation, and provides a reliable basis for the scheduling decision of microgrids, which will not be repeated here.

[0034] It should be noted that dynamic programming is an optimization algorithm that aims to efficiently solve the optimal solution by decomposing complex problems into simpler sub-problems. The method uses a recursive approach to first solve the smallest sub-problem and store its results to avoid repeated calculations. This bottom-up solution strategy is particularly suitable for scenarios with overlapping sub-problems and optimal substructure properties, such as in path optimization, resource allocation, and scheduling problems. Through dynamic programming, the global optimal solution can be quickly found in the multi-stage decision-making process, significantly improving computational efficiency, and is widely used in fields such as operations research, computer science, and economics.

[0035] It should be noted that the load pattern refers to the changing characteristics and trends of electricity demand within a corresponding time range. It reflects the user's electricity consumption habits in different time periods (such as days, weeks, seasons, etc.), including load demand during peak and trough periods. Load patterns are usually affected by many factors, such as weather conditions, holidays, social activities, and economic activities. By analyzing the load pattern, the electricity consumption characteristics of the corresponding period can be identified, providing an important basis for the planning, scheduling and management of the power system, especially when optimizing the integration and scheduling strategies of renewable energy.

[0036] It should be noted that in the dispatching process of the microgrid, the goal is to balance the load demand and power generation capacity and minimize the operating cost, and to achieve efficient management by gradually determining the optimal dispatching strategy at each time point. First, the system will evaluate the current supply and demand situation based on the real-time load demand and available power generation capacity (including renewable energy and conventional power generation) data. Then, the optimization algorithm is used to gradually adjust the output level of each power generation unit and the charging and discharging strategy of the energy storage device to ensure that the overall operating cost is reduced while meeting the load demand. Through this dynamic adjustment, it is possible to effectively respond to instantaneous load changes and ensure the economy and stability of the system.

[0037] Determining the optimal dispatch strategy at each point in time usually involves a specific calculation and evaluation process. For example, at a certain point in time, assume that the real-time load demand is 100 kW, while the renewable energy generation capacity is 60 kW and the current power of the energy storage device is 40 kWh. First, the system will evaluate whether the power generation capacity is sufficient to meet the load demand and find that the gap is 40 kW. Next, the dispatch algorithm will consider mobilizing the energy storage device and decide whether to discharge to make up for this gap. If the energy storage discharge efficiency is 90%, 36 kW of support can be provided. Through optimization calculations, the system may determine that the power generation unit should output 60 kW at this point in time, and the energy storage device should discharge 36 kW at the same time to achieve a total power generation of 96 kW, ultimately reaching 100 kW close to the load demand, while minimizing operating costs. This step-by-step optimization process ensures the rational allocation and economy of resources.

[0038] S2, obtaining the microgrid dispatching data and resource configuration data corresponding to the H time and B time interval during the real-time coordinated dispatching of the main and distribution microgrids.

[0039] The microgrid dispatching data includes a dispatching flexibility evaluation coefficient and a power quality evaluation coefficient, and the resource configuration data includes a resource configuration efficiency impact coefficient.

[0040] The analysis method of the dispatch flexibility evaluation coefficient combines the impact assessment of the dispatch window size on the flexibility of the microgrid. First, by setting different dispatch window sizes, the scenarios of load fluctuations and changes in renewable energy generation are simulated, and the response time and decision accuracy of the system in these scenarios are recorded. Then, the average response time and decision accuracy under each window size are calculated, and the flexibility index is determined through statistical analysis. In addition, a weighted scoring system is used to comprehensively consider the response speed, accuracy and calculation efficiency to form a dispatch flexibility evaluation coefficient. This coefficient helps to evaluate the real-time coordinated dispatch effect of the main and distribution microgrids under different energy window sizes, thereby solving the balance problem between flexible response and calculation efficiency.

[0041] Analyzing the dispatch flexibility evaluation coefficient is helpful for evaluating the real-time coordinated dispatch effect of the main distribution microgrid under different energy window sizes, thereby solving the problem that the real-time coordinated dispatch of the main distribution microgrid is difficult to balance the flexible response and computational efficiency. It has the following advantages:

[0042] Accurate responsiveness: This coefficient can quantify the response speed of the dispatch window to load fluctuations and changes in renewable energy generation, helping operators to choose the appropriate window size, thereby improving the flexibility and responsiveness of the system.

[0043] Optimizing dispatch decisions: By evaluating the flexibility of different window settings, operators can identify the dispatch strategy that best suits the current electricity demand and supply situation, ensuring timely decisions in a rapidly changing environment.

[0044] Comprehensive evaluation indicators: The scheduling flexibility evaluation coefficient incorporates multiple dimensions such as response time, decision accuracy, and computational efficiency into the evaluation, providing a comprehensive performance evaluation framework that helps to better understand the advantages and disadvantages of scheduling strategies.

[0045] Balancing flexibility and efficiency: By analyzing the flexible evaluation coefficient, operators can effectively find the optimal balance between flexible response and computing efficiency, avoiding the delayed response caused by over-reliance on large windows or the computing burden caused by small windows.

[0046] Dynamic adjustment strategy: Based on the evaluation results, the microgrid can continuously optimize the dispatching strategy, improve the system's adaptability under different load and power generation conditions, and ensure the stability and economy of the overall operation.

[0047] The specific analysis method of the scheduling flexibility evaluation coefficient is as follows:

[0048] Acquire several dispatch response rate data of dispatch decision during real-time coordinated dispatch of main and distribution microgrids, and convert them into time series and then frequency domain to obtain dispatch response fluctuation frequency data;

[0049] Feature extraction is performed based on the scheduling response fluctuation frequency data to obtain response features, which include response overlap features, response range features, and frequency features;

[0050] The response overlap characteristics include the overlap peak and the overlap frequency difference;

[0051] Response scope characteristics include the proportion of dispatch resources;

[0052] The frequency characteristics include frequency fluctuation value;

[0053] According to the response characteristics and the conventional scheduling range, the difference between the response characteristics and the conventional scheduling range is calculated based on a preset scheduling flexibility evaluation coefficient calculation formula as the scheduling flexibility evaluation coefficient.

[0054] The specific calculation formula of the dispatch response fluctuation frequency data is:

[0055]

[0056] The specific calculation formula of the overlapping peak value and the overlapping frequency difference is:

[0057]

[0058] The calculation formula for the scheduling resource ratio is:

[0059]

[0060] The calculation formula of the frequency fluctuation value is:

[0061]

[0062] The specific calculation formula of the scheduling flexibility evaluation coefficient is as follows:

[0063]

[0064] In the formula, To dispatch response to fluctuating frequency data, is the dispatch response rate data at time t, is the conversion function from time domain to frequency domain, is the scale of the transfer function from time domain to frequency domain, is the translation parameter of the conversion function from time domain to frequency domain; is the overlap peak and the overlap frequency difference, is the center frequency, and are the minimum and maximum values ​​of the response range, is the scheduling resource ratio, is the frequency fluctuation value, is the maximum resource within the scheduling range, is the minimum resource within the scheduling range, It is the scheduling flexibility assessment factor.

[0065] The power quality assessment coefficient is a quantitative indicator used to measure the power quality and system stability of a microgrid under corresponding dispatching conditions. This coefficient takes into account multiple power quality factors such as voltage fluctuation, frequency stability, harmonic content, and power supply reliability. By analyzing the impact of different dispatch window sizes on power quality, the power quality assessment coefficient can reflect the effectiveness of the dispatch strategy. Smaller dispatch windows may lead to frequent adjustments, thereby improving the stability of the power grid in the short term, but may also increase the risk of failure. A reasonable window size helps maintain power quality and reduce the occurrence of failures. By using the power quality assessment coefficient, microgrid operators can optimize dispatching decisions, ensure a balance between flexible response and power quality, and improve the reliability and safety of the overall system.

[0066] Analyzing the power quality assessment coefficient is helpful for evaluating the real-time coordinated dispatching effect of the main distribution microgrid under different energy window sizes, thereby solving the problem that the real-time coordinated dispatching of the main distribution microgrid is difficult to balance flexible response and computational efficiency. It has the following advantages:

[0067] Comprehensive reflection of power quality: This coefficient comprehensively considers multiple power quality factors such as load changes, frequency fluctuations and harmonic content, providing a comprehensive assessment of the overall performance of the power grid and helping to identify potential power quality problems.

[0068] Optimize dispatch decisions: By quantifying the impact of different dispatch windows on power quality, operators can better balance flexible response and computational efficiency in dispatch strategies to ensure grid stability under rapidly changing demand.

[0069] Reduced risk of failure: Power quality assessment factors help identify and prevent system instability caused by frequent dispatch, thereby reducing the occurrence of failures and improving the reliability of the power grid.

[0070] Guiding window selection: Based on the analysis results of the power quality assessment coefficient, operators can choose a more reasonable dispatch window size, thereby improving response speed while maintaining the power quality and stability of the power grid.

[0071] Dynamic adjustment and feedback: The power quality assessment coefficient allows dynamic adjustment of the dispatch strategy. Through real-time monitoring and feedback mechanism, operators can continuously optimize dispatch decisions and improve the adaptability and operating efficiency of the microgrid.

[0072] The specific method for obtaining the power quality assessment coefficient is as follows:

[0073] The method for obtaining the power quality assessment coefficient is as follows:

[0074] Obtain the load changes, power generation capacity, and usage of power electronic equipment of the main and distribution microgrids, and build the topology structure through simulation software;

[0075] Cluster the topological structure;

[0076] The processed topological structure is feature extracted, and the power intensity assessment coefficient is calculated based on the principle of energy constancy, energy loss coefficient and constraint conditions;

[0077] The dispatching situation of the main distribution microgrid is simulated through the topological structure, and the power quality model is established to obtain the power loss assessment coefficient; the power quality assessment coefficient is calculated by combining the power intensity assessment coefficient;

[0078] The specific calculation formula of the power intensity assessment coefficient is as follows:

[0079]

[0080] The specific calculation formula of the power loss assessment coefficient is as follows:

[0081]

[0082] The specific calculation formula of the power quality assessment coefficient is as follows:

[0083]

[0084] In the formula, is the power intensity assessment coefficient, is the power loss assessment coefficient, is the power quality assessment coefficient, is the error term, The capacitor of the main distribution microgrid, The power dispatching scope of the main distribution microgrid, is the line bandwidth, is the energy loss coefficient, is the Laplace operator in heat conduction, is the power loss rate, is the average electrical energy, This is the adjustment item for power loss.

[0085] The resource allocation efficiency impact coefficient is a key indicator used to measure the optimization allocation effect of resources in microgrids under different dispatch window sizes. This coefficient reflects the impact of the choice of dispatch window on resource allocation efficiency: a larger dispatch window can comprehensively consider the load and power generation capacity over a longer period of time, thereby achieving more reasonable resource allocation, reducing the use of redundant resources, and improving overall operating efficiency; while a smaller dispatch window may not be able to fully utilize long-term trend data, resulting in unreasonable resource allocation and affecting the overall performance of the system. By evaluating the resource allocation efficiency impact coefficient, microgrid operators can better understand the impact of different energy window settings on real-time coordinated dispatch effects, thereby finding the best balance between flexible response and computing efficiency and optimizing resource allocation strategies.

[0086] Analyzing the resource allocation efficiency impact coefficient is helpful for evaluating the real-time coordinated dispatching effect of the main distribution microgrid under different energy window sizes, thereby solving the problem that the real-time coordinated dispatching of the main distribution microgrid is difficult to balance flexible response and computational efficiency. It has the following advantages:

[0087] Optimize resource allocation: By quantifying resource allocation efficiency, operators can better understand the impact of different scheduling windows on resource utilization, thereby achieving more efficient resource allocation and reducing unnecessary redundancy.

[0088] Improve the rationality of scheduling decisions: This coefficient helps operators consider long-term load trends and power generation capacity when formulating scheduling strategies, ensuring that resource allocation not only takes into account short-term needs but also adapts to long-term changes, thereby enhancing the scientific nature of scheduling decisions.

[0089] Enhanced flexibility and stability: Through reasonable resource allocation, microgrids can respond to load fluctuations more flexibly while maintaining system stability and reducing the risk of failures caused by improper resource allocation.

[0090] Support dynamic adjustment: The resource allocation efficiency impact coefficient allows operators to adjust the scheduling window according to real-time data and evaluation results, optimize resource allocation strategies, and achieve a dynamic balance between flexible response and computing efficiency.

[0091] Improve overall operational efficiency: By effectively evaluating and optimizing resource allocation, microgrids can achieve higher operational efficiency, reduce operating costs, and enhance overall economic performance and sustainability.

[0092] The specific method for obtaining the resource allocation efficiency influence coefficient is as follows:

[0093] Obtaining a resource configuration set, wherein the resource configuration set includes resource proportion, resource configuration distribution and configuration characteristics;

[0094] Calculate the efficiency of each resource configuration in the resource configuration set based on the resource configuration distribution and the corresponding coefficient of variation;

[0095] For each resource configuration scheme, count the number of occurrences of the corresponding configuration feature to obtain the resource configuration scheme feature frequency;

[0096] The standard deviation and mean of the characteristic frequencies of resource allocation schemes are calculated, and the resource allocation efficiency impact coefficient is calculated in combination with the efficiency.

[0097] The specific calculation formula of the characteristic frequency of the resource allocation scheme is as follows:

[0098]

[0099] The specific calculation formula for the standard deviation of the characteristic frequency of the resource allocation scheme is as follows:

[0100]

[0101] The specific calculation formula of the resource allocation efficiency influence coefficient is as follows:

[0102]

[0103] In the formula, is the characteristic frequency of feature a in the resource allocation scheme, H is the number of resources in the resource allocation scheme, is the average frequency of resource allocation scheme characteristics, is the total number of configured features, is the standard deviation of the resource allocation scheme characteristic frequency, It is the coefficient affecting resource allocation efficiency.

[0104] This embodiment improves the system's ability to respond to load fluctuations and uncertainty in renewable energy generation through load forecasting based on variational multimodal technology, making the forecast more accurate and ensuring that power demand is effectively met in a changing environment. Secondly, through the dynamic programming algorithm, the optimal scheduling strategy at each time point is gradually determined to maximize the balance between load demand and power generation capacity, significantly reducing operating costs. In addition, the introduction of the scheduling flexibility evaluation coefficient provides a quantitative basis for the impact of the scheduling window size on system flexibility and computing efficiency, helping operators make timely decisions and optimize resource allocation in the rapidly changing power market. At the same time, the use of the power quality evaluation coefficient ensures that the power quality is maintained while improving the response speed, reducing the risk of failure and enhancing the stability of the system. In short, this method realizes the efficient, stable and economical operation of the microgrid by comprehensively considering multiple factors such as load demand, power generation capacity and power quality, and has broad application prospects.

[0105] Embodiment 2, S3, performs data analysis on microgrid dispatching data and resource configuration data based on BP neural network algorithm, and constructs an energy window size influence model.

[0106] The BP neural network algorithm is used to analyze the microgrid dispatching data and resource configuration data, and to construct an energy window size influence model, specifically:

[0107] The obtained dispatch flexibility assessment coefficient, power quality assessment coefficient and resource allocation efficiency impact coefficient are used to establish a traffic shaping hidden danger assessment model to generate an energy window size impact coefficient;

[0108] The specific calculation formula of the energy window size influence coefficient is as follows:

[0109]

[0110] In the formula, is the energy window size influence coefficient, is the preset proportional coefficient of the scheduling flexibility assessment coefficient, is the preset proportionality factor of the power quality assessment coefficient, is the preset proportional coefficient of the resource allocation efficiency impact coefficient, is the scheduling flexibility assessment coefficient, is the preset proportionality factor of the power quality assessment coefficient, It is the preset proportional coefficient of the resource allocation efficiency impact coefficient.

[0111] S4, performing arithmetic processing based on the preset energy window size to obtain several different energy window sizes, and respectively obtaining corresponding energy window size influence models.

[0112] S5, according to several different energy window sizes and the corresponding energy window size impact model output, construct an energy window size impact fluctuation curve, and perform data analysis on the energy window size impact fluctuation curve to obtain the optimized energy window size, and apply it to the real-time coordinated scheduling of the main distribution microgrid.

[0113] The energy window size influence fluctuation curve is constructed according to a number of different energy window sizes and the corresponding energy window size influence model output, specifically:

[0114] Obtaining energy window size influence coefficients under different energy window sizes from an energy window size influence model;

[0115] Connecting a number of energy window sizes and energy window size influence coefficients to construct an energy window size influence fluctuation curve;

[0116] The energy window size is analyzed for fluctuation curves, and the optimized energy window size is obtained, and applied to the real-time coordinated dispatch of the main distribution microgrid, specifically:

[0117] Obtain the maximum value of the fluctuation curve affected by the size of several energy windows;

[0118] Take the second derivative of each maximum value respectively;

[0119] The energy window size corresponding to the maximum value of the quadratic derivative closest to zero is taken as the optimized energy window size, and is applied to the real-time coordinated scheduling of the main distribution microgrid.

[0120] This embodiment uses the BP neural network algorithm to conduct an in-depth analysis of the microgrid dispatching data and resource allocation data, and constructs an energy window size impact model. This model not only evaluates the comprehensive impact of dispatching flexibility, power quality and resource allocation efficiency, but also generates an energy window size impact coefficient, thereby providing a scientific basis for the assessment of traffic shaping hazards. Through the arithmetic processing of different energy window sizes and the analysis of the model output, the energy window size impact fluctuation curve is finally constructed. The optimization analysis of this fluctuation curve can effectively identify the optimal energy window size, ensuring that the real-time coordinated dispatching of the microgrid is more efficient and flexible. This process improves the accuracy of dispatching decisions, reduces resource waste, optimizes overall energy management, and has stronger adaptability and responsiveness, thereby improving the operating efficiency and reliability of the microgrid.

[0121] Embodiment 3, Figure 2 A system for a real-time coordinated dispatching method of a main distribution microgrid, comprising a prediction module, a data acquisition module, a window size analysis module, a window size setting module and a screening module;

[0122] The prediction module is used to obtain the load data and historical load data of the main distribution microgrid, perform load prediction based on variational multimodality, and make scheduling decisions under the preset energy window size based on dynamic programming;

[0123] A data acquisition module is used to obtain the microgrid dispatching data and resource configuration data corresponding to the H time and B time interval during the real-time coordinated dispatching of the main and distribution microgrids;

[0124] The window size analysis module is used to analyze the microgrid dispatch data and resource configuration data based on the BP neural network algorithm and build an energy window size impact model;

[0125] A window size setting module is used to perform arithmetic processing based on a preset energy window size to obtain a number of different energy window sizes, and obtain the corresponding energy window size influence models respectively;

[0126] The screening module is used to construct an energy window size impact fluctuation curve according to several different energy window sizes and the corresponding energy window size impact model output, and perform data analysis on the energy window size impact fluctuation curve to obtain the optimized energy window size, and apply it to the real-time coordinated scheduling of the main distribution microgrid.

[0127] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0128] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0129] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the corresponding application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each corresponding application, but such implementation should not be considered to be beyond the scope of this application.

[0130] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0131] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0132] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A real-time coordinated dispatching method for a main distribution microgrid, characterized in that: The steps include: Obtain the load data and historical load data of the main distribution microgrid, perform load forecasting based on variational multi-modality, and make scheduling decisions under the preset energy window size based on dynamic programming; Obtain the microgrid dispatching data and resource configuration data corresponding to the H time and B time interval during the real-time coordinated dispatching of the main and distribution microgrids; The load data includes load demand, renewable energy generation data, and energy storage status; The load forecasting based on variational multimodality and the scheduling decision based on dynamic programming under a preset energy window size are specifically: using variational multimodality technology to process load data, modeling different load modes, and capturing corresponding load characteristics; Extracting time series features from historical load data, wherein the time series features include seasonal features, trend features, and periodic features; Based on variational multimodal technology, the time series characteristics and predicted load characteristics are integrated and iterated to obtain the load forecasting model for load forecasting; Within the preset energy window size, based on the current load forecast value, the availability of power generation resources and the energy storage status, the optimal scheduling strategy at each time point is gradually determined with the goal of balancing load demand and power generation capacity and minimizing operating costs; The microgrid dispatching data includes a dispatching flexibility evaluation coefficient and a power quality evaluation coefficient. The specific analysis method of the dispatching flexibility evaluation coefficient is as follows: Acquire several dispatch response rate data of dispatch decision during real-time coordinated dispatch of main and distribution microgrids, and convert them into time series and then frequency domain to obtain dispatch response fluctuation frequency data; Feature extraction is performed based on the scheduling response fluctuation frequency data to obtain response features, which include response overlap features, response range features, and frequency features; The response overlap characteristics include the overlap peak and the overlap frequency difference; Response scope characteristics include the proportion of dispatch resources; The frequency characteristics include frequency fluctuation value; According to the response characteristics and the conventional dispatch range, the difference between the response characteristics and the conventional dispatch range calculated based on the preset dispatch flexibility evaluation coefficient calculation formula is used as the dispatch flexibility evaluation coefficient; The resource configuration data includes a resource configuration efficiency impact coefficient, and the specific method for obtaining the power quality assessment coefficient is as follows: Obtain the load changes, power generation capacity, and usage of power electronic equipment of the main and distribution microgrids, and build the topology structure through simulation software; Cluster the topological structure; The processed topological structure is feature extracted, and the power intensity assessment coefficient is calculated based on the principle of energy constancy, energy loss coefficient and constraint conditions; The dispatching situation of the main distribution microgrid is simulated through the topological structure, and the power quality model is established to obtain the power loss assessment coefficient; the power quality assessment coefficient is calculated by combining the power intensity assessment coefficient; The specific method for obtaining the resource allocation efficiency influence coefficient is as follows: Obtaining a resource configuration set, wherein the resource configuration set includes resource proportion, resource configuration distribution and configuration characteristics; Calculate the efficiency of each resource configuration in the resource configuration set based on the resource configuration distribution and the corresponding coefficient of variation; For each resource configuration scheme, count the number of occurrences of the corresponding configuration feature to obtain the resource configuration scheme feature frequency; Calculate the standard deviation and mean of the characteristic frequency of resource allocation schemes, and calculate the resource allocation efficiency impact coefficient in combination with the efficiency; The specific calculation formula of the dispatch response fluctuation frequency data is: The specific calculation formula of the overlapping peak value and the overlapping frequency difference is: The calculation formula for the scheduling resource ratio is: The calculation formula of the frequency fluctuation value is: The specific calculation formula of the scheduling flexibility evaluation coefficient is as follows: In the formula, To dispatch response to fluctuating frequency data, is the dispatch response rate data at time t, is the conversion function from time domain to frequency domain, is the scale of the transfer function from time domain to frequency domain, is the translation parameter of the conversion function from time domain to frequency domain; is the overlap peak and the overlap frequency difference, is the center frequency, and are the minimum and maximum values ​​of the response range, is the scheduling resource ratio, is the frequency fluctuation value, is the maximum resource within the scheduling range, is the minimum resource within the scheduling range, The flexibility assessment factor for scheduling; The specific calculation formula of the power intensity assessment coefficient is as follows: The specific calculation formula of the power loss assessment coefficient is as follows: The specific calculation formula of the power quality assessment coefficient is as follows: In the formula, is the power intensity assessment coefficient, is the power loss assessment coefficient, is the power quality assessment coefficient, is the error term, The capacitor of the main distribution microgrid, The power dispatching scope of the main distribution microgrid, is the line bandwidth, is the energy loss coefficient, is the Laplace operator in heat conduction, is the power loss rate, is the average electrical energy, is the adjustment item for power loss; Based on the BP neural network algorithm, the microgrid dispatch data and resource configuration data are analyzed to build an energy window size impact model; Based on the preset energy window size, arithmetic processing is performed to obtain several different energy window sizes, and the corresponding energy window size influence models are obtained respectively; According to several different energy window sizes and the corresponding energy window size affecting the output of the model, an energy window size affecting fluctuation curve is constructed, and data analysis is performed on the energy window size affecting fluctuation curve to obtain the optimized energy window size, which is then applied to the real-time coordinated scheduling of the main distribution microgrid.

2. The real-time coordinated dispatching method for main and distribution microgrids according to claim 1 is characterized in that: The energy window size influence fluctuation curve is constructed according to a number of different energy window sizes and the corresponding energy window size influence model output, specifically: Obtaining energy window size influence coefficients under different energy window sizes from an energy window size influence model; Several energy window sizes and energy window size influence coefficients are connected to construct an energy window size influence fluctuation curve.

3. The real-time coordinated dispatching method for main and distribution microgrids according to claim 2 is characterized in that: The energy window size is analyzed for fluctuation curves, and the optimized energy window size is obtained, and applied to the real-time coordinated dispatch of the main distribution microgrid, specifically: Obtain the maximum value of the fluctuation curve affected by the size of several energy windows; Take the second derivative of each maximum value respectively; The energy window size corresponding to the maximum value of the quadratic derivative closest to zero is taken as the optimized energy window size, and is applied to the real-time coordinated scheduling of the main distribution microgrid.

4. A system using the real-time coordinated dispatching method for a main distribution microgrid as claimed in any one of claims 1 to 3, characterized in that: It includes a prediction module, a data acquisition module, a window size analysis module, a window size setting module and a screening module; The prediction module is used to obtain the load data and historical load data of the main distribution microgrid, perform load prediction based on variational multimodality, and make scheduling decisions under the preset energy window size based on dynamic programming; A data acquisition module is used to obtain the microgrid dispatching data and resource configuration data corresponding to the H time and B time interval during the real-time coordinated dispatching of the main and distribution microgrids; The window size analysis module is used to analyze the microgrid dispatch data and resource configuration data based on the BP neural network algorithm and build an energy window size impact model; A window size setting module is used to perform arithmetic processing based on a preset energy window size to obtain a number of different energy window sizes, and obtain the corresponding energy window size influence models respectively; The screening module is used to construct an energy window size impact fluctuation curve according to several different energy window sizes and the corresponding energy window size impact model output, and perform data analysis on the energy window size impact fluctuation curve to obtain the optimized energy window size, and apply it to the real-time coordinated scheduling of the main distribution microgrid.

Images