Micro-grid storage load coordination control optimization method and system

Abstract

The invention discloses a microgrid storage load coordination control optimization method and system, and relates to the technical field of power grid control, and the method comprises the steps: analyzing a target historical generating capacity record of a target microgrid, and obtaining a predicted generating capacity; analyzing the target historical electricity utilization record to obtain a target electricity utilization type set; performing correlation analysis on the power consumption time and the power consumption types to obtain a correlation result, and establishing a schedulable power consumption type set; calling a load prediction plan to obtain a predicted electrical load; and comparing the predicted generating capacity with the predicted electrical load to obtain a predicted comparison result, and generating a coordination control plan. According to the method, the technical problems of low energy utilization efficiency and operation stability of the micro-grid caused by inaccurate power generation capacity and load prediction and insufficient energy storage and load scheduling optimization in the prior art are solved, and coordinated optimization control of energy storage and load of the micro-grid is realized; and the energy utilization efficiency and the operation stability of the micro-grid are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid control, and in particular to a microgrid load-storage coordinated control optimization method and system. Background Art

[0002] With the continuous growth of energy demand and the increasing emphasis on sustainable development, microgrids, as a technology that can integrate distributed energy, achieve efficient energy utilization and reliable power supply, have received widespread attention. However, the operation of microgrids faces many challenges. On the one hand, the power generation in microgrids is affected by many factors, such as environmental conditions, the unstable characteristics of distributed energy, etc. Accurate prediction of power generation is crucial to optimizing the operation of microgrids. At the same time, the power demand of users of different types of electricity at different times varies greatly, and the power consumption behavior has a certain degree of uncertainty, which makes it difficult to accurately predict the power load. On the other hand, how to achieve coordinated control of energy storage and load in microgrids to balance power generation and power consumption, improve energy utilization efficiency, and ensure the stable operation of microgrids is also a problem that needs to be solved urgently. Traditional microgrid management methods often find it difficult to fully consider the influence of various complex factors and cannot achieve accurate prediction and efficient coordinated control.

[0003] The existing technology has technical problems such as inaccurate power generation and load forecasting, insufficient optimization of energy storage and load scheduling, which lead to low energy utilization efficiency and operation stability of microgrids. Summary of the invention

[0004] The present application provides a microgrid storage-load coordinated control optimization method and system, which is used to solve the technical problems in the prior art of inaccurate power generation and load prediction, insufficient optimization of energy storage and load scheduling, and low energy utilization efficiency and operation stability of the microgrid.

[0005] In view of the above problems, the present application provides a microgrid load-storage coordinated control optimization method and system.

[0006] In a first aspect of the present application, a microgrid load-storage coordinated control optimization method is provided, the method comprising:

[0007] Based on a predetermined unit period, the target historical power generation record of the target microgrid is analyzed to obtain the predicted power generation of the target microgrid at a preset future time; the target historical power consumption record is analyzed to obtain a target power consumption type set, and the target power consumption type set includes multiple power consumption types with historical power consumption time identifiers; a correlation analysis between power consumption time and power consumption type is performed based on the multiple power consumption types with historical power consumption time identifiers to obtain a correlation result; a dispatchable power consumption type set is formed based on the correlation result, and a peak avoidance incentive strategy is provided for the dispatchable power consumption type set; a load forecasting plan is retrieved, and a load forecasting analysis is performed on the target microgrid under the peak avoidance incentive strategy based on the load forecasting plan to obtain a predicted power load at the preset future time; the predicted power generation is compared with the predicted power load to obtain a prediction comparison result, and a coordinated control plan at the preset future time is generated based on the prediction comparison result.

[0008] A second aspect of the present application provides a microgrid load-storage coordinated control optimization system, the system comprising:

[0009] A predicted power generation acquisition module, which analyzes the target historical power generation records of the target microgrid based on a predetermined unit period to obtain the predicted power generation of the target microgrid at a preset future time; a target power type set acquisition module, which is used to analyze the target historical power consumption records to obtain a target power type set, and the target power type set includes multiple power types with historical power consumption time identifiers; a correlation result acquisition module, which is used to perform a correlation analysis between power consumption time and power consumption type based on the multiple power consumption types with historical power consumption time identifiers to obtain a correlation result; a peak avoidance incentive strategy providing module Block, the peak avoidance incentive strategy providing module is used to form a dispatchable electricity consumption type set according to the correlation result, and provide a peak avoidance incentive strategy for the dispatchable electricity consumption type set; a predicted electricity load acquisition module, the predicted electricity load acquisition module is used to call the load forecast plan, and according to the load forecast plan, perform load forecast analysis on the target microgrid under the peak avoidance incentive strategy to obtain the predicted electricity load at the preset future time; a coordinated control plan generation module, the coordinated control plan generation module is used to compare the predicted power generation with the predicted electricity load to obtain a predicted comparison result, and generate a coordinated control plan at the preset future time according to the predicted comparison result.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] Based on the predetermined unit period, the target historical power generation record of the target microgrid is analyzed to obtain the predicted power generation of the target microgrid at the preset future time; the target historical power consumption record is analyzed to obtain the target power consumption type set; the correlation analysis between power consumption time and power consumption type is performed according to multiple power consumption types with historical power consumption time identification, and the correlation result is obtained to form a dispatchable power consumption type set; the load forecast plan is retrieved, and the load forecast analysis of the target microgrid under the peak avoidance incentive strategy is performed according to the load forecast plan to obtain the predicted power load at the preset future time; the predicted power generation is compared with the predicted power load to obtain the prediction comparison result, and the coordinated control plan at the preset future time is generated. The technical effect of realizing the coordinated optimization control of microgrid energy storage and load and improving the energy utilization efficiency and operation stability of the microgrid is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic diagram of a flow chart of a microgrid load-storage coordinated control optimization method provided in an embodiment of the present application;

[0014] Figure 2 A schematic diagram of the structure of a microgrid load-storage coordinated control optimization system provided in an embodiment of the present application.

[0015] Explanation of the reference numerals: predicted power generation acquisition module 10, target power consumption type set acquisition module 20, correlation result acquisition module 30, peak avoidance incentive strategy providing module 40, predicted power load acquisition module 50, coordinated control plan generation module 60. DETAILED DESCRIPTION

[0016] The present application provides a microgrid storage-load coordinated control optimization method and system, which is used to solve the technical problems in the prior art such as inaccurate power generation and load prediction, insufficient optimization of energy storage and load scheduling, and low energy utilization efficiency and operation stability of the microgrid.

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] Embodiment 1, as Figure 1 As shown, the present application provides a microgrid storage-load coordinated control optimization method, the method comprising:

[0019] Step S100: Analyze the target historical power generation record of the target microgrid based on a predetermined unit period to obtain the predicted power generation of the target microgrid at a preset future time.

[0020] Specifically, first, a predetermined unit cycle is determined, which can be a day, a week, or other appropriate time lengths. The target historical power generation records of the target microgrid are the basis for analysis, and these records contain the actual power generation of the microgrid over a period of time in the past. By analyzing these historical power generation records according to the predetermined unit cycle, the law and trend of power generation over time can be found. For example, it is found that the power generation is higher in some seasons or time periods, while the power generation is lower in other time periods. Various data analysis methods are used in the analysis process, such as statistical analysis and trend analysis. Based on the in-depth analysis of historical data, combined with factors affecting power generation, such as weather conditions and equipment operating status, the power generation of the target microgrid at the preset future time is predicted. This prediction result is crucial for the operation and management of the microgrid. It can help managers make power generation plans in advance and reasonably arrange the charging and discharging of energy storage equipment to ensure that the microgrid can provide users with electricity stably. By accurately predicting power generation, the microgrid can better achieve the optimal configuration of energy, improve energy utilization efficiency, reduce operating costs, and also help improve the reliability and stability of the microgrid.

[0021] Step S200: Analyze the target historical electricity usage records to obtain a target electricity usage type set, wherein the target electricity usage type set includes a plurality of electricity usage types with historical electricity usage time identifiers.

[0022] Specifically, first, the target historical power consumption records of the target microgrid are deeply analyzed. These historical power consumption records cover the power consumption of each user or device in the microgrid in the past period of time. By carefully combing these records, different power consumption behaviors are classified to determine various power consumption types. The target power consumption type set is an important result of the analysis, in which each power consumption type has a historical power consumption time mark. This means that the power consumption of each power consumption type at different time points can be clearly known. For example, some power consumption types may consume more electricity during the day, while others may have peak power consumption at night. By clarifying different power consumption types and their historical power consumption time marks, we can better understand the power consumption patterns and characteristics of the microgrid, which will help to formulate more accurate power consumption management strategies for different power consumption types in the future. For example, for types with concentrated power consumption in a specific time period, take peak power consumption measures or provide incentive policies to optimize the load distribution of the microgrid and improve energy utilization efficiency. At the same time, understanding the target power consumption type set also provides important basic information for predicting future power loads, enabling microgrid managers to carry out energy planning and scheduling more scientifically.

[0023] Step S300: performing a correlation analysis between electricity usage time and electricity usage type according to the plurality of electricity usage types with historical electricity usage time identifiers to obtain a correlation result.

[0024] Specifically, for the multiple electricity consumption types with historical electricity consumption time identification, the correlation analysis between electricity consumption time and electricity consumption type is started. First, the historical electricity consumption time pattern of each electricity consumption type is examined one by one. By comparing and analyzing the historical electricity consumption time of different electricity consumption types in detail, the relationship between electricity consumption time and electricity consumption type can be determined. This relationship can be positive, that is, a specific electricity consumption type has a higher electricity consumption in a specific time period; it can also be negative, or present other complex correlation patterns. In the analysis process, various data analysis techniques and statistical methods are used to accurately quantify the degree of correlation between electricity consumption time and electricity consumption type. The final correlation result is an important discovery, which provides a strong basis for subsequent microgrid management decisions. If it is found that some electricity consumption types are highly correlated with specific time periods, more targeted measures can be taken for these types when formulating electricity management strategies. For example, for electricity consumption types with high electricity consumption during peak hours, a peak avoidance incentive strategy can be provided to adjust the electricity consumption time, thereby optimizing the load distribution of the microgrid and improving the overall energy utilization efficiency and system stability.

[0025] Step S400: forming a dispatchable electricity consumption type set according to the correlation result, and providing a peak avoidance incentive strategy for the dispatchable electricity consumption type set.

[0026] Specifically, based on the correlation results, we start to build a set of dispatchable electricity types. Dispatched electricity types refer to loads that adjust their electricity consumption time and amount according to the needs of the power system. In contrast, non-dispatched loads often have strict requirements on electricity consumption time and cannot be adjusted at will. By carefully analyzing the correlation results, we can accurately identify which electricity types have the potential to be dispatched.

[0027] For example, if the electricity consumption of certain commercial electricity types in a specific time period has a certain correlation with the overall peak electricity consumption, and it is evaluated that the electricity consumption time can be adjusted to a certain extent, then these commercial electricity types can be included in the set of dispatchable electricity types.

[0028] At the same time, in order to encourage the dispatchable electricity consumption types to make reasonable adjustments to electricity consumption, a peak avoidance incentive strategy is provided for the dispatchable electricity consumption type set. This incentive strategy is multi-dimensional, such as providing environmental protection guard certification, or taking the form of reducing electricity prices to provide economic incentives. Through these incentives, the dispatchable electricity consumption types are effectively guided to reduce electricity consumption during peak hours and increase electricity consumption during off-peak hours, thereby optimizing the load distribution of the microgrid.

[0029] Such operations help improve the stability and reliability of microgrids and reduce energy costs, while also meeting the requirements of sustainable development and promoting efficient energy use and environmental protection.

[0030] Step S500: Retrieve a load forecasting plan, and perform load forecasting analysis on the target microgrid under the peak avoidance incentive strategy according to the load forecasting plan to obtain a predicted power load at the preset future time.

[0031] Specifically, first, the load forecasting plan is retrieved. This plan is a set of methods and processes for predicting the load of the microgrid, which takes into account various factors that affect the load, such as historical electricity consumption data, seasonal changes, weather conditions, etc. Then, according to this load forecasting plan, the target microgrid under the peak-avoidance incentive strategy is subjected to load forecasting analysis. Since the peak-avoidance incentive strategy has been provided for the dispatchable electricity type set, the electricity consumption of the microgrid will change at this time. When conducting load forecasting analysis, it is necessary to fully consider the impact of the peak-avoidance incentive strategy on different electricity types. For example, some dispatchable electricity types will adjust the electricity consumption time due to the incentive strategy, thereby affecting the overall electricity load curve. Through comprehensive analysis and calculation of various factors, the predicted electricity load at the preset future time is predicted. This predicted electricity load is crucial to the operation and management of the microgrid. It can help managers make scheduling arrangements for power generation and energy storage equipment in advance to ensure that the microgrid can stably meet the electricity needs of users in the future. At the same time, the peak-avoidance incentive strategy can also be adjusted and optimized according to the prediction results to further improve the energy utilization efficiency and operation stability of the microgrid.

[0032] Step S600: Compare the predicted power generation with the predicted power load to obtain a prediction comparison result, and generate a coordinated control plan for the preset future time according to the prediction comparison result.

[0033] Specifically, first, the predicted power generation obtained through the previous steps is compared with the predicted power load. The predicted power generation represents the power generation capacity of the microgrid at the preset future time, while the predicted power load reflects the user's power demand. By comparing these two key indicators, we can clearly understand the supply and demand situation that the microgrid may face in the future. If the predicted power generation is greater than the predicted power load, it means that the microgrid will have excess power supply in the future, and it can be considered to store the excess power or export it to other power grids. On the contrary, if the predicted power generation is less than the predicted power load, it means that the microgrid faces the risk of insufficient power supply, and measures need to be taken in advance to balance supply and demand. Based on the predicted comparison results obtained by comparison, a coordinated control plan for the preset future time is generated. This plan comprehensively considers the power generation, energy storage and power consumption of the microgrid, aiming to achieve supply and demand balance and optimize energy utilization. For example, if insufficient power supply is predicted, measures such as increasing the operating time of power generation equipment, calling energy storage equipment to discharge, or adjusting the power consumption time of dispatchable power types are formulated. If excess power is predicted, strategies such as reducing the operation of power generation equipment, charging energy storage equipment, or encouraging more dispatchable power types to use electricity at this time can be considered. By generating coordinated control plans, microgrid managers can prepare in advance and respond to various possible situations in a timely manner to ensure the stable operation and efficient energy utilization of the microgrid.

[0034] In a possible implementation, step S100 further includes:

[0035] Step S110: dividing the target historical power generation record according to the predetermined unit period to obtain a division result, wherein the division result includes a plurality of historical periods with historical power generation identifiers.

[0036] Step S120: establishing a target historical power generation sequence according to the correspondence between the historical periods and the historical power generation in the plurality of historical periods with historical power generation identifiers.

[0037] Step S130: performing a trend analysis on the target historical power generation time series in combination with the preset future time to obtain the predicted power generation.

[0038] Specifically, first, the predetermined unit period is a time interval determined according to the characteristics of the microgrid and the analysis requirements. This period can be one day, one week, one month, etc., depending on the operating law of the microgrid and the availability of data. When dividing the target historical power generation record of the target microgrid, the historical power generation data is divided into multiple fragments based on the predetermined unit period. Each fragment corresponds to a historical period and is marked with a historical power generation. Such a division makes the huge historical power generation record more orderly and easier to analyze. For example, if the predetermined unit period is one week, the historical power generation record will be divided into multiple consecutive period intervals. Each period has its specific power generation data, which can reflect the power generation of the microgrid during that week. In this way, the changing trends and characteristics of power generation in different historical periods can be more clearly observed. This division result provides a clear data structure for the subsequent analysis steps. Each historical period can be analyzed separately, or the differences between different periods can be compared, so as to better understand the power generation behavior and law of the microgrid and lay the foundation for accurately predicting future power generation.

[0039] When there are multiple historical cycles with historical power generation identification, the target historical power generation time series is established according to the correspondence between these historical cycles and historical power generation. First, the position of each historical cycle on the time axis is clarified, and then the corresponding historical power generation is associated with it. In this way, a sequence arranged in chronological order is formed, which clearly shows the power generation of the microgrid at different time points. For example, if the historical cycle is divided into months, then starting from the earliest month, the power generation of each month is recorded in sequence, so as to establish a continuous target historical power generation time series. This time series can intuitively show the trend of the power generation of the microgrid over time, whether it is increasing, decreasing or fluctuating. By establishing the target historical power generation time series, it is not only easier to observe the evolution of historical power generation, but also to dig deeper into the laws and potential trends, so as to more accurately predict the power generation of the microgrid at the preset future time, and provide decision support for the efficient operation and management of the microgrid.

[0040] Observe the data change trend in the target historical power generation time series, and draw a line chart of the historical power generation over time to intuitively see whether the trend is rising, falling or fluctuating. If the overall trend is relatively stable, use a simple linear trend extrapolation method. If the trend is more complex, the weighted moving average method can be used to assign different weights to each historical data point. The recent data points have a larger weight and the distant data points have a smaller weight. When calculating the weighted average, the power generation of each data point is multiplied by its corresponding weight, and then the sum is divided by the total weight. By constantly adjusting the weight setting, find the weighting scheme that best reflects the current trend. For historical power generation time series with seasonal changes, use the seasonal index method to determine the seasonal cycle in the historical data, calculate the average power generation in each seasonal cycle, and obtain the seasonal index. Adjust the trend analysis according to the location of the current season and the seasonal index. When conducting trend analysis, consider the influence of external factors, such as weather changes, equipment maintenance plans, etc. If it is known that certain external factors will have a greater impact on power generation, the trend can be corrected based on these factors. Once the trend model is determined, the power generation at the preset future time is predicted by extrapolating along the trend line according to the interval between the preset future time and the current time. For example, if the trend is linearly rising, and it is known that the power generation per unit time increases by a certain amount, then the predicted power generation can be obtained by multiplying the difference between the preset future time and the current time by the increase in power generation per unit time, and adding the current power generation. By comprehensively considering the trend characteristics of the historical power generation time series, external factors, and reasonable algorithm selection, the predicted power generation at the preset future time can be obtained more accurately, providing an important basis for the storage-load coordinated control of the microgrid.

[0041] In a possible implementation, step S130 further includes:

[0042] Step S131: drawing a scatter plot of target historical power generation according to a first corresponding relationship, wherein the first corresponding relationship refers to a corresponding relationship between any set of cycles in the target historical power generation sequence and the power generation.

[0043] Step S132: performing polynomial regression fitting on the target historical power generation scatter plot to obtain a target fitting polynomial.

[0044] Step S133: inputting the preset future time into the target fitting polynomial to obtain the predicted power generation.

[0045] Specifically, first, the first corresponding relationship is clarified, that is, the corresponding relationship between any set of cycles and power generation in the target historical power generation time series. This corresponding relationship is the basis for subsequent analysis, which reflects the power generation of the microgrid in different time periods. According to this corresponding relationship, the target historical power generation scatter plot is drawn. In the two-dimensional coordinate system, the cycle is used as the horizontal coordinate and the power generation is used as the vertical coordinate. For each set of cycle and power generation data pairs in the target historical power generation time series, the corresponding position is found in the coordinate system and a scatter point is marked. By drawing a scatter plot, the change of historical power generation with the cycle can be intuitively seen. The distribution of scatter points can show some potential trends and characteristics. For example, if the scatter points show an upward trend, it means that the power generation is gradually increasing over time; if the scatter points are relatively scattered, it means that the change of power generation is relatively unstable. This scatter plot provides a visual basis for further analysis. Analysts can preliminarily judge the law of historical power generation data by observing the scatter plot, provide intuitive reference for subsequent steps such as polynomial regression fitting, and help to more accurately predict future power generation, so as to better optimize the storage-load coordinated control of the microgrid.

[0046] For the target historical power generation scatter plot, the polynomial regression fitting is to find a mathematical expression that can best describe the relationship between the historical cycle and power generation. First of all, polynomial regression is a powerful data analysis tool, which assumes that the relationship between power generation and cycle can be expressed by a polynomial function. For example, a quadratic polynomial function is in the form of y=a 0 +a 1 x+a 2 x 2 , where x represents the period, y represents the power generation, and a 0 、a 1 and a 2is the coefficient to be determined. In the process of fitting, the least squares method is used to minimize the error between the fitted polynomial function and each data point on the scatter plot. It is to calculate the difference between the actual power generation of each data point and the power generation predicted by the polynomial function, and then sum the squares of these differences, and adjust the value of the coefficient to minimize the total error. After a series of calculations and optimizations, the target fitting polynomial is finally obtained. This polynomial function can well capture the trends and laws in historical data. It can not only describe the changes in power generation in the past, but also be used to predict future power generation. For example, if a quadratic polynomial fitting result is obtained, then the preset future time can be substituted into the periodic variable according to this function to calculate the corresponding predicted power generation. By performing polynomial regression fitting on the target historical power generation scatter plot, an important analysis tool and prediction basis are provided for the storage-load coordinated control of the microgrid, which helps managers to better plan and manage the operation of the microgrid.

[0047] After the previous steps, the target fitting polynomial has been obtained. This polynomial is a mathematical abstraction of the relationship between the cycle and the power generation in the historical power generation data. The preset future time is input into this target fitting polynomial. The preset future time is a specific time point or time period determined according to actual needs, which represents the moment of predicting power generation. After substituting the preset future time into the target fitting polynomial, the polynomial will be calculated according to its function form and coefficients. Since the polynomial has captured the trends and laws in the historical data as much as possible during the fitting process, the calculation result at this time can be considered as a reasonable prediction of the power generation at the preset future time. This predicted power generation can help managers plan power generation resources, adjust energy storage strategies, and optimize load distribution in advance to ensure that when the preset future time arrives, the microgrid can operate stably and efficiently to meet the power demand of users. At the same time, by continuously updating historical data and re-fitting predictions, the accuracy of the prediction can be improved, providing strong support for the sustainable development of the microgrid.

[0048] In a possible implementation, step S133 further includes:

[0049] Step S1331: Read predetermined environmental factors, and collect multi-dimensional features of the environment of the target microgrid based on the predetermined environmental factors to obtain target environmental data.

[0050] Step S1332: Perform normalized weighted processing on the target environment data using the coefficient of variation principle to obtain a target environment index.

[0051] Step S1333: Using the target environmental index as a power generation adjustment coefficient, the predicted power generation is adjusted.

[0052] Specifically, first, the predetermined environmental factors are read. These predetermined environmental factors are pre-determined factors that may affect the power generation of the target microgrid. For example, temperature, humidity, light intensity, wind speed, etc. are all common environmental factors. The selection of these factors is based on the analysis of the power generation principle of the microgrid and past experience, and it is known that they are related to the power generation process of the microgrid. Next, based on these predetermined environmental factors, multi-dimensional feature collection of the environment of the target microgrid is performed, which means obtaining data related to these environmental factors through various means, and using sensors installed in the area where the microgrid is located to monitor environmental parameters in real time, such as temperature sensors measuring temperature and humidity sensors measuring humidity. At the same time, relevant environmental data is obtained from an external meteorological database to ensure the accuracy and comprehensiveness of the data. During the collection process, specific data collection will be performed for different environmental factors. For example, for light intensity, a photosensitive sensor is used to measure at different time periods, and the light intensity values ​​at different times are recorded. For wind speed, an anemometer is used to measure the wind speed around the microgrid. Through such multi-dimensional feature collection, detailed data on various aspects of the environment in which the target microgrid is located can be obtained. The target environmental data finally obtained is a set of values ​​of various environmental factors. These data will provide a specific basis for analyzing the impact of the environment on power generation in subsequent steps, making the prediction of microgrid power generation more accurate and reliable, and helping to better coordinate the storage and load control of the microgrid.

[0053] First, the coefficient of variation is a statistic used to measure the degree of data dispersion. For each environmental factor in the target environmental data, its coefficient of variation is calculated. The coefficient of variation is calculated by dividing the standard deviation of the environmental factor by its mean. By calculating the coefficient of variation, the fluctuation and relative stability of each environmental factor can be understood. The normalization weighting process is performed using the principle of the coefficient of variation. Due to the different dimensions and numerical ranges of different environmental factors, direct comparison and comprehensive analysis will be difficult. Through normalization, the data values ​​of each environmental factor can be converted to the same scale to make them comparable. For each environmental factor, the original data value is subtracted from the minimum value of the factor, and then divided by the difference between the maximum and minimum values ​​of the factor to obtain the normalized value. Then, the weight of each environmental factor is determined according to the coefficient of variation. Generally speaking, environmental factors with larger coefficients of variation indicate larger fluctuations and may have a more significant impact on the power generation of the microgrid, so they are given higher weights; while environmental factors with smaller coefficients of variation are given lower weights. In this way, the role of important environmental factors can be highlighted while avoiding excessive interference of minor factors on the results. Finally, the normalized environmental factor values ​​are multiplied by the corresponding weights and summed to obtain the target environmental index. This index comprehensively reflects the overall condition of the environment in which the target microgrid is located, as well as the comprehensive impact of various environmental factors on power generation. The target environmental index can be used as an important reference indicator to adjust and optimize the prediction of microgrid power generation, so that the prediction results more accurately reflect the actual situation and provide a more scientific basis for the storage-load coordinated control of the microgrid.

[0054] When the target environmental index is used as the power generation adjustment coefficient, it means that the potential impact of environmental factors on power generation is quantified into a specific value. If the target environmental index is high, it means that the current environmental conditions have a positive impact on power generation and will increase power generation. For example, if environmental factors such as strong light intensity and suitable wind speed are conducive to the better performance of the power generation equipment of the microgrid, then the target environmental index will be correspondingly high. On the contrary, if the target environmental index is low, it means that the environmental conditions have an adverse effect on power generation and the predicted power generation needs to be reduced. For example, severe weather conditions and excessive humidity will reduce the efficiency of power generation equipment, resulting in a reduction in power generation. In the process of adjusting the predicted power generation, the predicted power generation is multiplied by the target environmental index to obtain the adjusted power generation. This can make the prediction results more in line with the actual situation and fully consider the impact of the dynamic changes of environmental factors on the power generation of the microgrid. In this way, the microgrid manager can more accurately carry out storage-load coordination control based on the adjusted power generation, make adjustments to the power generation plan, dispatch energy storage equipment, and manage the load in advance, so as to ensure that the microgrid can operate stably under different environmental conditions and meet the power needs of users. At the same time, continuously monitoring and updating environmental data, as well as recalculating the target environmental index to adjust power generation, can improve the adaptability and flexibility of microgrid operation and management.

[0055] In a possible implementation, step S500 further includes:

[0056] Step S510: Divide the target microgrid into a target power consumption block set, wherein the target power consumption block set includes a plurality of power consumption blocks.

[0057] Step S520: extracting any power usage block from the plurality of power usage blocks, and obtaining a plurality of adjacent power usage blocks of the arbitrary power usage block.

[0058] Step S530: obtaining a first predicted power load of a first adjacent power block, where the first adjacent power block refers to any one power block obtained from the plurality of adjacent power blocks based on a random sampling principle.

[0059] Step S540: performing aggregate analysis on the first predicted power load to obtain the predicted power load.

[0060] Specifically, the target microgrid is divided into a target power consumption block set, and a complex network is divided into multiple relatively independent but interrelated sub-areas. Each power consumption block is regarded as a node in the graph neural network, and the electrical connections and mutual influences between them constitute the edges of the graph.

[0061] Further extract any power consumption block and determine its multiple adjacent power consumption blocks. From the perspective of graph neural network, this is equivalent to determining the neighborhood nodes of a specific node. These adjacent power consumption blocks and the extracted power consumption blocks often have power interactions and influences in actual microgrids.

[0062] Based on the principle of random sampling, a first adjacent power consumption block is selected from multiple adjacent power consumption blocks, and its first predicted power load is obtained, which is to randomly select a neighborhood node in the graph neural network and obtain its specific attribute value. Through random sampling, the amount of calculation can be reduced to a certain extent, while better reflecting the overall trend.

[0063] From the perspective of graph neural network, each power block can be regarded as a node in the graph, and the electrical connections and mutual influences between them constitute the edges of the graph. First, for the first predicted power load, it represents the predicted power demand value of the first adjacent power block obtained from random sampling. When performing aggregation analysis, the graph neural network integrates the information of the surrounding nodes through the message passing mechanism. For the node where the first predicted power load is located, it will receive information from its adjacent nodes (i.e., the nodes corresponding to other adjacent power blocks), which includes the predicted power load of the adjacent power blocks, the historical power consumption pattern, and other characteristics that affect power consumption. Through the aggregation function of the graph neural network, the first predicted power load is fused with the information from the adjacent nodes. This aggregation function is designed according to different strategies, such as weighted averaging, where the weights can be determined based on factors such as the connection strength and historical correlation between nodes. After aggregation analysis, the predicted power load obtained is no longer just the predicted value of a single power block, but a combination of the influence and information of multiple adjacent power blocks. Such a predicted power load more accurately reflects the overall power demand trend and distribution in the microgrid. It can help microgrid managers better forecast and coordinate loads, rationally arrange power generation resources, energy storage equipment, and adjust dispatchable power types to ensure stable operation and efficient energy utilization of microgrids. The aggregation analysis method of graph neural networks can fully consider the relationship between power blocks in microgrids, providing more powerful tools and more accurate results for predicting power loads.

[0064] In a possible implementation, step S540 further includes:

[0065] Step S541: Acquire a first user set of the first adjacent power consumption block, where the first user set includes a plurality of users.

[0066] Step S542: Read a predetermined sampling strategy, and sample the multiple users based on the predetermined sampling strategy to obtain a first target user set.

[0067] Step S543: obtaining a first target predicted electricity consumption of a first target user, where the first target user refers to any electricity user in the first target user set.

[0068] Step S544: obtaining a predetermined influence coefficient of the peak avoidance incentive strategy, and adjusting the first target predicted power consumption by the predetermined influence coefficient to obtain a first target power load.

[0069] Step S545: performing aggregate analysis on the first target power load to obtain the first predicted power load.

[0070] Specifically, after the first adjacent power consumption block is determined, the first user set of the block needs to be obtained. Each power consumption block is composed of multiple users, and the power consumption behaviors of these users jointly determine the power load of the power consumption block. The multiple users in the first user set include different types of power users such as residential users, commercial users, and industrial users. Each user has its own unique power demand and pattern. For example, the power consumption of residential users is mainly concentrated on the use of household appliances, such as lighting, television, air conditioning, etc., and their power consumption time is usually related to daily life; commercial users include shops, restaurants, offices, etc., and their power demand will be affected by business hours and business activities; the power consumption of industrial users is usually closely related to the production process, and their power consumption is large and the power consumption pattern is more complex. By obtaining the first user set of the first adjacent power consumption block, a specific object is provided for subsequent analysis. Understanding the power consumption of these users helps to more accurately predict the power load of the power consumption block, thereby providing a basis for the overall storage and load coordination control of the microgrid. For example, targeted power management strategies can be formulated based on the power consumption characteristics of different user types to achieve efficient operation of the microgrid and rational allocation of energy.

[0071] A greedy sampling strategy is used to sample multiple users in the first adjacent power consumption block to obtain the first target user set. Greedy sampling is a targeted sampling method. First, it is clear that its purpose is to select a representative part of users from many users so that subsequent analysis and prediction can be carried out efficiently without processing all user data. For multiple users in the first adjacent power consumption block, the greedy sampling strategy will select according to specific criteria. For example, factors such as the user's power consumption, the regularity of power consumption time, and the type of power consumption equipment can be considered. If a user's power consumption is large and his power consumption pattern is relatively stable, then this user is more likely to be selected. In the specific implementation process, users are evaluated one by one in a certain order, and it is decided whether to include them in the first target user set based on the current evaluation results. For example, the evaluation starts with the user with the largest power consumption. If the user meets certain preset conditions, it will be added to the first target user set. Then continue to evaluate the next user until the predetermined number of samples is reached or other stop conditions are met. The first target user set obtained by greedy sampling not only includes representative users, but also reduces the complexity of data processing. These selected users can reflect the overall electricity consumption characteristics of the users in the entire first adjacent electricity consumption block, providing a reliable basis for the subsequent acquisition of the predicted electricity consumption of the first target user and the aggregate analysis of the electricity load.

[0072] In the microgrid load-storage coordinated control optimization method, for any power user in the first target user set, that is, the first target user, obtaining its first target predicted power consumption is a key step. After determining the first target user, it is necessary to predict its power consumption based on multiple factors. The historical power consumption data of the user can be considered to analyze its power consumption patterns in different time periods in the past. For example, if the user's power consumption has increased significantly in the summer period of the past few months, it is due to the increased frequency of use of equipment such as air conditioners. Then, in the current similar season and time period, it can be predicted that its power consumption will increase to a certain extent based on historical trends. At the same time, consider the impact of external factors on the user's power consumption. For example, weather changes can affect the use of certain electrical appliances. Hot weather will increase the power consumption of air conditioners, while cold weather will cause the power consumption of electric heating equipment to increase. By comprehensively analyzing these factors, time series analysis is used to predict the trend and seasonal changes of historical power consumption data to obtain the first target predicted power consumption of the first target user.

[0073] Obtain the predetermined influence coefficient of the peak-avoidance incentive strategy, which represents the degree of influence of the peak-avoidance incentive strategy on the user's electricity consumption behavior. The peak-avoidance incentive strategy is usually intended to guide users to reduce electricity consumption during peak hours and increase electricity consumption during off-peak hours to achieve load balance and efficient operation of the microgrid. The predetermined influence coefficient is obtained by analyzing and statistics of historical data. For the first target user's first target predicted electricity consumption, the predetermined influence coefficient is adjusted. If the predetermined influence coefficient is positive, it means that the peak-avoidance incentive strategy prompts the user to increase electricity consumption. Then, a certain amount is correspondingly increased on the basis of the first target predicted electricity consumption to obtain the first target electricity load. For example, if a user's first target predicted electricity consumption is 100 kWh and the predetermined influence coefficient is 0.2, then the adjusted first target electricity load is 100×(1+0.2)=120 kWh. If the predetermined influence coefficient is negative, it means that the peak-avoidance incentive strategy causes the user to reduce electricity consumption. At this time, a certain amount is correspondingly reduced on the basis of the first target predicted electricity consumption. In this way, the actual impact of the peak-avoidance incentive strategy on the user's electricity consumption behavior is taken into account, so that the first target electricity load obtained is more in line with the actual situation. This helps to more accurately predict the power load of the first adjacent power consumption block and provide a more reliable basis for the coordinated control of the storage and load of the microgrid. According to the adjusted power load, managers can reasonably arrange power generation resources, adjust energy storage strategies, and further optimize peak-avoidance incentive strategies to achieve stable operation and efficient energy utilization of the microgrid.

[0074] The first target power load is the power load value obtained for a single user in the first target user set after considering the predetermined influence coefficient of the peak avoidance incentive strategy. However, the power load of a single user cannot fully represent the power consumption of the entire first adjacent power block. When performing aggregation analysis, it is necessary to comprehensively consider the power loads of multiple users in the first target user set, and adopt multi-weighted averaging. The weighted average method can assign different weights to the power load of each user according to different factors. For example, the weight can be determined based on factors such as the user's power consumption, the regularity of power consumption time, and the degree of influence on the stability of the microgrid. Users with larger power consumption are given higher weights because they contribute more to the overall power load; users with more regular power consumption time are also given relatively higher weights because their power consumption behavior is easier to predict. Through aggregation analysis, the first predicted power load obtained can more comprehensively reflect the power demand of the first adjacent power block. This predicted value will provide an important basis for the subsequent power load forecasting and storage-load coordinated control of the entire microgrid. According to the first predicted power load, power generation resources can be reasonably arranged, energy storage strategies can be adjusted, and more precise power management measures can be formulated to ensure the stable operation of the microgrid and efficient energy utilization.

[0075] In a possible implementation, step S542 further includes:

[0076] Step S5421: Obtain the first user among the multiple users.

[0077] Step S5422: Collect the first historical electricity usage record of the first user.

[0078] Step S5423: Analyze the first historical electricity consumption record to obtain the first average unit electricity consumption of the first user.

[0079] Step S5424: When the first average unit power consumption is greater than the power consumption threshold, sampling the first user based on the predetermined sampling strategy.

[0080] Specifically, when it is necessary to sample multiple users in the first adjacent power consumption block, the first user among these users must be obtained first. This first user is selected in a certain specific order, such as the order of user numbers or the first user selected randomly. The importance of this step lies in providing a specific starting point for subsequent analysis. By acquiring and analyzing each user one by one, the power consumption of users in the entire power consumption block can be fully understood. The determination of the first user also establishes a benchmark for subsequent operations, so that subsequent processing of other users can be compared and referenced with it. After obtaining the first user, further collect its historical power consumption records, analyze power consumption characteristics, etc., to provide basic data for subsequent sampling decisions and power load forecasting. At the same time, this step also lays the foundation for the orderly conduct of the entire sampling process, ensuring that each user has the opportunity to be included in the analysis scope, thereby improving the accuracy and representativeness of the sampling.

[0081] After the first user is determined, the first historical electricity consumption record of the user is collected. These historical electricity consumption records can be obtained from multiple channels. For example, the electricity consumption data recorded by devices such as smart meters can accurately reflect the electricity consumption of the user in different time periods. The historical electricity consumption records include the electricity consumption at different times of the day, the trend of total electricity consumption changes in a week or month, and the difference in electricity consumption in different seasons. These detailed historical electricity consumption records provide a basis for in-depth understanding of the electricity consumption habits and patterns of the first user. By analyzing these data, the user's peak and low periods of electricity consumption can be found, and the user's use of different electrical equipment at different times can be understood. The collected first historical electricity consumption record not only helps to evaluate the user's contribution to the electricity load of the entire electricity consumption block, but also provides basic data for subsequent analysis steps, such as calculating the average unit electricity consumption and judging whether it meets the sampling standards. At the same time, these historical electricity consumption records also provide a reference for microgrid managers to formulate more accurate electricity management strategies, such as taking appropriate measures to guide users to use electricity reasonably during peak electricity consumption periods to achieve stable operation and efficient energy utilization of the microgrid.

[0082] After obtaining the first historical electricity consumption record of the first user, it is necessary to conduct in-depth analysis of these records to obtain the first average unit electricity consumption. First, check the various data in the historical electricity consumption record, including the electricity consumption values ​​of different dates and different time periods. These records reflect the electricity consumption behavior pattern of the first user in the past period of time. In order to calculate the average unit electricity consumption, the historical electricity consumption record is divided according to a specific time unit, such as days, and the total electricity consumption per day in a period of time is counted. Then, this sum is divided by the corresponding number of days to obtain the average daily electricity consumption of the first user during this period, or a similar calculation is performed in months to obtain the average monthly electricity consumption. In addition, analysis is performed according to different types of electricity consumption. For example, different types of electricity consumption records such as household electricity, commercial electricity or industrial electricity are distinguished, and the average unit electricity consumption of each type is calculated separately. In this way, the electricity consumption of the first user in different electricity consumption scenarios can be understood in more detail. The first average unit electricity consumption obtained is an important indicator that helps us intuitively understand the electricity consumption intensity and regularity of the first user. If the average unit electricity consumption is high, it means that the user has a large demand for electricity or there are some high-energy-consuming equipment usage. On the contrary, if the average unit power consumption is low, it means that the power consumption is relatively economical, or the power-consuming equipment is relatively energy-efficient. This indicator is crucial for subsequent operations. When sampling users, the comparison between the average unit power consumption and the power consumption threshold can be used to decide whether to sample the user. At the same time, for microgrid managers, understanding the average unit power consumption of different users helps to formulate more accurate power management strategies, such as optimizing energy distribution, implementing demand response measures, etc., to improve the operating efficiency and stability of the microgrid.

[0083] If the first average unit power consumption of the first user is greater than the power consumption threshold value, this indicates that the user has a higher intensity or greater influence in terms of power consumption. The power consumption threshold value is a standard value set according to the overall situation of the microgrid and specific analysis requirements. When the average unit power consumption of the user exceeds this threshold value, it means that the user has a more significant impact on the power load of the entire power consumption block. In this case, the first user is sampled based on a predetermined sampling strategy. The predetermined sampling strategy is greedy sampling, and its purpose is to select a representative part of users from many users for a deeper analysis. Sampling the first user who meets the conditions can ensure that those users with large power consumption and outstanding contributions to the power load are included in the analysis scope. By sampling and analyzing these high-power users, more valuable information is provided for the storage-load coordinated control of the microgrid. The peak power consumption hours and main power consumption equipment types of these users can be analyzed to formulate targeted power management strategies, such as taking demand response measures during peak hours to guide these users to reasonably adjust the power consumption time or reduce power consumption, thereby achieving load balance and stable operation of the microgrid. At the same time, sampling analysis can also provide a basis for microgrid planning and resource allocation, ensuring that there is sufficient power generation and storage capacity to meet the needs of high-power users.

[0084] Embodiment 2, based on the same inventive concept as a microgrid load-storage coordinated control optimization method in the above embodiment, Figure 2 As shown, the present application provides a microgrid load-storage coordinated control optimization system, and the system and method embodiments in the present application embodiments are based on the same inventive concept. The system includes:

[0085] The predicted power generation acquisition module 10 analyzes the target historical power generation records of the target microgrid based on a predetermined unit period to obtain the predicted power generation of the target microgrid at a preset future time.

[0086] The target electricity usage type set acquisition module 20 is used to analyze the target historical electricity usage records to obtain a target electricity usage type set, wherein the target electricity usage type set includes a plurality of electricity usage types with historical electricity usage time identifiers.

[0087] The correlation result acquisition module 30 is used to perform a correlation analysis between the electricity usage time and the electricity usage type according to the multiple electricity usage types with historical electricity usage time identifiers to obtain a correlation result.

[0088] The peak avoidance incentive strategy providing module 40 is used to form a dispatchable electricity consumption type set according to the correlation result and provide a peak avoidance incentive strategy for the dispatchable electricity consumption type set.

[0089] The predicted power load acquisition module 50 is used to retrieve the load prediction plan, and perform load prediction analysis on the target microgrid under the peak avoidance incentive strategy according to the load prediction plan to obtain the predicted power load at the preset future time.

[0090] The coordinated control plan generation module 60 is used to compare the predicted power generation with the predicted power load to obtain a predicted comparison result, and generate a coordinated control plan for the preset future time according to the predicted comparison result.

[0091] Furthermore, the predicted power generation acquisition module 10 also includes:

[0092] A division result acquisition unit is used to divide the target historical power generation record according to the predetermined unit period to obtain a division result, wherein the division result includes a plurality of historical periods with historical power generation identifiers.

[0093] A target historical power generation sequence establishing unit is used to establish a target historical power generation sequence according to the correspondence between the historical cycles and the historical power generation in the multiple historical cycles with historical power generation identifiers.

[0094] A trend analysis unit is used to perform a trend analysis on the target historical power generation time sequence in combination with the preset future time to obtain the predicted power generation.

[0095] Furthermore, the trend analysis unit also includes:

[0096] A power generation scatter diagram drawing unit is used to draw a target historical power generation scatter diagram according to a first corresponding relationship, wherein the first corresponding relationship refers to a corresponding relationship between any group of cycles in the target historical power generation sequence and the power generation.

[0097] A fitting polynomial acquisition unit is used to perform polynomial regression fitting on the target historical power generation scatter plot to obtain a target fitting polynomial.

[0098] A predicted power generation acquisition unit is used to input the preset future time into the target fitting polynomial to obtain the predicted power generation.

[0099] Furthermore, the predicted power generation acquisition unit further includes:

[0100] A target environment data acquisition unit is used to read predetermined environmental factors and collect multi-dimensional features of the environment of the target microgrid based on the predetermined environmental factors to obtain target environment data.

[0101] The target environment index acquisition unit is used to perform normalized weighted processing on the target environment data using the coefficient of variation principle to obtain a target environment index.

[0102] A predicted power generation adjustment unit is used to adjust the predicted power generation using the target environmental index as a power generation adjustment coefficient.

[0103] Furthermore, the predicted power load acquisition module 50 further includes:

[0104] A target power block set acquisition unit is used to divide the target microgrid into target power block sets, wherein the target power block set includes a plurality of power blocks.

[0105] The adjacent power usage block acquisition unit is used to extract any power usage block from the multiple power usage blocks and obtain multiple adjacent power usage blocks of the any power usage block.

[0106] The first predicted power load acquisition unit is used to acquire a first predicted power load of a first adjacent power block, wherein the first adjacent power block refers to any one power block obtained from the plurality of adjacent power blocks based on a random sampling principle.

[0107] An aggregate analysis unit, wherein the aggregate analysis unit is used to perform an aggregate analysis on the first predicted power load to obtain the predicted power load.

[0108] Furthermore, the first predicted power load acquisition unit further includes:

[0109] A first user set acquisition unit is used to acquire a first user set of the first adjacent power consumption block, wherein the first user set includes a plurality of users.

[0110] The first target user set acquisition unit is used to read a predetermined sampling strategy, and sample the multiple users based on the predetermined sampling strategy to obtain a first target user set.

[0111] The first target predicted electricity consumption acquisition unit is used to acquire a first target predicted electricity consumption of a first target user, where the first target user refers to any electricity user in the first target user set.

[0112] The first target power load acquisition unit is used to acquire a predetermined influence coefficient of the peak avoidance incentive strategy, and adjust the first target predicted power consumption with the predetermined influence coefficient to obtain a first target power load.

[0113] The first target power load aggregation analysis unit is used to perform aggregation analysis on the first target power load to obtain the first predicted power load.

[0114] Furthermore, the first target user set acquisition unit further includes:

[0115] A first user acquisition unit is configured to acquire a first user from among the plurality of users.

[0116] The first historical electricity usage record collecting unit is used to collect the first historical electricity usage record of the first user.

[0117] The first average unit power consumption acquisition unit is used to analyze the first historical power consumption record to obtain the first average unit power consumption of the first user.

[0118] The first user sampling unit is used to sample the first user based on the predetermined sampling strategy when the first average unit power consumption is greater than the power consumption threshold.

[0119] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0120] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0121] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A microgrid load-storage coordinated control optimization method, characterized in that: include: Analyze the target historical power generation record of the target microgrid based on a predetermined unit period to obtain the predicted power generation of the target microgrid at a preset future time; Analyze the target historical electricity consumption records to obtain a target electricity consumption type set, wherein the target electricity consumption type set includes multiple electricity consumption types with historical electricity consumption time identifiers; Performing a correlation analysis between the electricity usage time and the electricity usage type according to the multiple electricity usage types with historical electricity usage time identifiers to obtain a correlation result; According to the correlation results, a dispatchable electricity consumption type set is formed, and a peak avoidance incentive strategy is provided for the dispatchable electricity consumption type set; Retrieving a load forecasting plan, and performing load forecasting analysis on the target microgrid under the peak avoidance incentive strategy according to the load forecasting plan to obtain a predicted power load at the preset future time; The predicted power generation is compared with the predicted power load to obtain a prediction comparison result, and a coordinated control plan for the preset future time is generated according to the prediction comparison result.

2. According to claim 1, a microgrid load-storage coordinated control optimization method is characterized in that: include: Dividing the target historical power generation record by the predetermined unit period to obtain a division result, wherein the division result includes a plurality of historical periods with historical power generation identifiers; Establishing a target historical power generation time sequence according to the correspondence between the historical periods and the historical power generation in the multiple historical periods with historical power generation identifiers; The target historical power generation time series is subjected to trend analysis in combination with the preset future time to obtain the predicted power generation.

3. According to claim 2, a microgrid load-storage coordinated control optimization method is characterized in that: include: Draw a scatter plot of target historical power generation according to a first corresponding relationship, wherein the first corresponding relationship refers to a corresponding relationship between any set of cycles in the target historical power generation time series and the power generation; Performing polynomial regression fitting on the target historical power generation scatter plot to obtain a target fitting polynomial; The preset future time is input into the target fitting polynomial to obtain the predicted power generation.

4. According to claim 1, a microgrid load-storage coordinated control optimization method is characterized in that: Also includes: Reading predetermined environmental factors, and collecting multi-dimensional features of the environment of the target microgrid based on the predetermined environmental factors to obtain target environmental data; The target environment data is normalized and weighted using the coefficient of variation principle to obtain a target environment index; The predicted power generation is adjusted using the target environmental index as a power generation adjustment coefficient.

5. According to claim 1, a microgrid load-storage coordinated control optimization method is characterized in that: include: Dividing the target microgrid to obtain a target power consumption block set, wherein the target power consumption block set includes a plurality of power consumption blocks; Extracting any power consumption block from the multiple power consumption blocks, and obtaining multiple adjacent power consumption blocks of the any power consumption block; Obtaining a first predicted power load of a first adjacent power block, where the first adjacent power block refers to any one power block obtained from the plurality of adjacent power blocks based on a random sampling principle; Aggregate analysis is performed on the first predicted power load to obtain the predicted power load.

6. A microgrid load-storage coordinated control optimization method according to claim 5, characterized in that: include: Acquire a first user set of the first adjacent power consumption block, where the first user set includes a plurality of users; Reading a predetermined sampling strategy, and sampling the multiple users based on the predetermined sampling strategy to obtain a first target user set; Obtaining a first target predicted power consumption of a first target user, where the first target user refers to any power user in the first target user set; Obtaining a predetermined influence coefficient of the peak avoidance incentive strategy, and adjusting the first target predicted power consumption by the predetermined influence coefficient to obtain a first target power load; Aggregate analysis is performed on the first target power load to obtain the first predicted power load.

7. A microgrid load-storage coordinated control optimization method according to claim 6, characterized in that: include: Obtaining a first user among the multiple users; Collecting a first historical electricity consumption record of the first user; Analyzing the first historical electricity consumption record to obtain a first average unit electricity consumption of the first user; When the first average unit power consumption is greater than the power consumption threshold value, the first user is sampled based on the predetermined sampling strategy.

8. A microgrid load-storage coordinated control optimization system, characterized in that: The system comprises: A predicted power generation acquisition module, wherein the predicted power generation acquisition module analyzes the target historical power generation record of the target microgrid based on a predetermined unit period to obtain the predicted power generation of the target microgrid at a preset future time; A target electricity usage type set acquisition module, the target electricity usage type set acquisition module is used to analyze the target historical electricity usage records to obtain a target electricity usage type set, the target electricity usage type set includes a plurality of electricity usage types with historical electricity usage time identifiers; A correlation result acquisition module, the correlation result acquisition module is used to perform a correlation analysis between the electricity usage time and the electricity usage type according to the multiple electricity usage types with historical electricity usage time identifiers to obtain a correlation result; A peak avoidance incentive strategy providing module, the peak avoidance incentive strategy providing module is used to form a dispatchable power consumption type set according to the correlation result, and provide a peak avoidance incentive strategy for the dispatchable power consumption type set; A predicted power load acquisition module, which is used to retrieve a load forecast plan, and perform load forecast analysis on the target microgrid under the peak avoidance incentive strategy according to the load forecast plan to obtain a predicted power load at the preset future time; A coordinated control plan generation module is used to compare the predicted power generation with the predicted power load to obtain a predicted comparison result, and generate a coordinated control plan for the preset future time according to the predicted comparison result.

Images