Microgrid dynamic load balancing optimization method and system based on historical power consumption data

By building a load forecasting model and intelligent power dispatching based on historical electricity consumption data, the problems of inaccurate power load forecasting and unbalanced distribution in microgrids are solved, and efficient and stable operation of microgrids is achieved.

CN119051084BActive Publication Date: 2025-09-19NORTHEAST DIANLI UNIVERSITY
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202411165493.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2025-09-19
Estimated Expiration
2044-08-23

AI Technical Summary

Technical Problem

Inaccurate power load forecasting, unbalanced load distribution, and imperfect power dispatching strategies in microgrids lead to irrational resource allocation and affect the stable operation of microgrids.

Method used

By building a load forecasting model based on historical electricity consumption data, power load forecast data is generated, and based on this, load balancing optimization is performed, distribution plans are generated, and power intelligent scheduling is performed to optimize power resource allocation.

Benefits of technology

It improves the operating efficiency, stability and reliability of the microgrid and achieves optimal allocation and efficient utilization of power resources.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119051084B_ABST
    Figure CN119051084B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for dynamic load balancing optimization of microgrids based on historical electricity consumption data, relating to the field of power systems. The method comprises: sensing microgrid nodes to obtain historical electricity consumption data according to historical electricity consumption cycles, constructing a historical electricity change graph based on this data, building a load forecasting model based on the graph to generate power load forecast data, performing load balancing optimization based on this data to obtain a load distribution plan, and executing this plan to perform intelligent power optimization and dispatching of microgrid nodes. This method solves the technical problems of inaccurate microgrid power load forecasting, unbalanced load distribution, and imperfect abnormal situation response and power dispatching strategies in the prior art, thereby achieving the technical effect of improving the overall efficiency, stability, and reliability of microgrid operation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to fields related to power systems, and in particular to a method and system for optimizing dynamic load balancing in microgrids based on historical electricity consumption data. Background Art

[0002] In microgrid operation and management scenarios, the issue of stable power supply during peak load periods is particularly prominent, and the contradiction between power resource allocation and demand is relatively more pronounced. How to efficiently allocate power resources to better meet the power needs of different nodes has become a crucial link in solving the problem of stable microgrid operation. Traditional microgrid management is often relatively simple and localized, relying only on fixed monitoring equipment or limited scheduling strategies. It lacks control over the overall operation status of the microgrid and lacks in-depth analysis and utilization of historical power consumption data. It is difficult to fully and accurately assess the power supply and demand status of each node. There are also unreasonable situations in resource allocation, resulting in insufficient resources at some key nodes and idle resources at others. The formulation of power scheduling plans is relatively simple and fixed, and cannot effectively respond to the ever-changing actual operation conditions of the microgrid.

[0003] At present, relevant technologies in microgrids have technical problems such as inaccurate prediction of power load, unbalanced load distribution, and imperfect response to abnormal situations and power dispatching strategies. Summary of the Invention

[0004] The present application provides a microgrid dynamic load balancing optimization method and system based on historical electricity consumption data. The method uses microgrid node sensing according to historical electricity consumption cycles to obtain historical electricity consumption data, constructs a historical power change graph based on this, and constructs a load prediction model based on the graph to generate power load prediction data. Based on this, load balancing optimization is performed to obtain a load distribution plan. The plan is executed to perform intelligent power optimization and scheduling of microgrid nodes, thereby achieving optimal configuration and efficient utilization of power resources, and achieving the technical effect of improving the overall efficiency, stability and reliability of microgrid operation.

[0005] This application provides a microgrid dynamic load balancing optimization method and system based on historical electricity consumption data, including:

[0006] Data sensing is performed on multiple nodes of the microgrid according to historical power consumption cycles to obtain multiple historical power consumption data; a historical power change graph is constructed based on the multiple historical power consumption data; a load prediction model is constructed based on the historical power change graph to predict the microgrid and generate multiple power load prediction data; load balancing optimization is performed based on the multiple power load prediction data to generate a load distribution plan; and the load distribution plan is executed to perform intelligent power optimization scheduling on multiple nodes of the microgrid.

[0007] This application also provides a microgrid dynamic load balancing optimization system based on historical electricity consumption data, including:

[0008] A historical electricity consumption data acquisition module, the historical electricity consumption data acquisition module is used to perform data sensing on multiple nodes of the microgrid according to the historical electricity consumption cycle to obtain multiple historical electricity consumption data; a historical power change graph generation module, the historical power change graph generation module is used to construct a historical power change graph based on the multiple historical electricity consumption data; a microgrid load prediction module, the microgrid load prediction module is used to construct a load prediction model according to the historical power change graph to predict the microgrid and generate multiple power load prediction data; a load distribution plan generation module, the load distribution plan generation module is used to perform load balancing optimization based on the multiple power load prediction data and generate a load distribution plan; an electric power intelligent dispatching module, the electric power intelligent dispatching module is used to execute the load distribution plan to perform electric power intelligent optimization dispatch on multiple nodes of the microgrid.

[0009] The microgrid dynamic load balancing optimization method and system based on historical electricity consumption data proposed in this application first performs data sensing on multiple nodes of the microgrid according to the historical electricity consumption cycle to obtain multiple historical electricity consumption data, constructs a historical power change graph based on the multiple historical electricity consumption data, constructs a load prediction model based on the historical power change graph to predict the microgrid, generates multiple power load prediction data, performs load balancing optimization based on the multiple power load prediction data, generates a load distribution plan, executes the load distribution plan to perform intelligent power optimization and scheduling on multiple nodes of the microgrid, and achieves the technical effect of improving the overall efficiency, stability and reliability of the microgrid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0011] Figure 1 A flow chart of a microgrid dynamic load balancing optimization method based on historical electricity consumption data provided in an embodiment of the present application;

[0012] Figure 2 A schematic diagram of the structure of a microgrid dynamic load balancing optimization system based on historical electricity consumption data provided in an embodiment of the present application.

[0013] Explanation of the accompanying symbols: historical electricity consumption data acquisition module 10, historical power change diagram generation module 20, microgrid load prediction module 30, load distribution plan generation module 40, power intelligent scheduling module 50. DETAILED DESCRIPTION

[0014] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0015] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0017] The present application embodiment provides a microgrid dynamic load balancing optimization method based on historical power consumption data, such as Figure 1 As shown, the method includes:

[0018] Step S100 , performing data sensing on multiple nodes of the microgrid according to a historical power consumption cycle to obtain multiple historical power consumption data.

[0019] Specifically, the historical electricity consumption cycle is accurately determined based on the historical electricity consumption ladder information. The historical electricity consumption ladder information reflects the patterns and laws of electricity consumption in different time periods. Through in-depth analysis, key cycle stages such as peak periods, valley periods, and stable periods can be identified. According to the historical power supply and demand information of the microgrid, multiple nodes are set. A clear understanding of the overall architecture of the microgrid and the power transmission path is required to ensure that the set nodes can comprehensively and effectively reflect the flow and use of electricity in the microgrid. According to the determined historical electricity consumption cycle and the supply and demand relationship of the set multiple nodes, Electricity consumption data is collected from microgrids. During the collection process, the electricity consumption and electricity load of multiple nodes in different time periods are accurately measured and recorded. The data obtained include multiple historical electricity consumption data and multiple historical electricity load data. The electricity consumption data directly reflects the total amount of electricity consumed by each node in a specific time period, while the electricity load data reflects the power load pressure borne by each node in the same time period. The carefully collected multiple historical electricity consumption data and multiple historical electricity load data are integrated and added to multiple historical electricity consumption data to provide rich, accurate and comprehensive basic data support for subsequent analysis and processing.

[0020] In one possible implementation, data sensing is performed on multiple nodes of the microgrid according to historical electricity consumption cycles to obtain multiple historical electricity consumption data. Step S100 further includes step S110, determining the historical electricity consumption cycle based on historical electricity consumption step information. Specifically, the historical electricity consumption cycle is determined based on the historical electricity consumption step information. The historical electricity consumption step information includes the classification of electricity prices in different time periods and the corresponding distribution of electricity consumption. Through in-depth analysis of this information, peak hours, low hours, and relatively stable periods of electricity consumption are discovered. For example, in some areas, daytime working hours may be peak electricity consumption, while nighttime may be low electricity consumption. The electricity consumption patterns on weekends and weekdays may also be different. By identifying these patterns and patterns, representative time periods can be identified, thereby determining the historical electricity consumption cycle.

[0021] Step S120: Set multiple nodes based on the microgrid's historical power supply and demand information. Specifically, setting multiple nodes based on the microgrid's historical power supply and demand information requires a comprehensive understanding of the microgrid's architecture, power transmission lines, and the power usage characteristics of each region. For example, the power demand and supply characteristics vary between industrial, commercial, and residential areas. Based on these different supply and demand characteristics, multiple nodes are set at key locations, such as power input points, major power equipment access points, and key power transmission branch points, to ensure comprehensive and accurate monitoring and analysis of power flow and usage.

[0022] Step S130, based on the historical power consumption cycle, combined with the supply and demand relationship of the multiple nodes, the power consumption of the microgrid is collected to obtain multiple historical power consumption data and multiple historical power load data. Specifically, based on the determined historical power consumption cycle, combined with the supply and demand relationship of the multiple nodes, the power consumption of the microgrid is collected. In each historical power consumption cycle, for the set multiple nodes, professional power monitoring equipment and sensors are used to measure and record the relevant power data in real time. Since the supply and demand relationship of different nodes in different cycles is different, some nodes may bear a larger power load during peak hours, while some nodes consume less power during off-peak hours. Through precise measurement and collection, multiple historical power consumption data are obtained, that is, the total amount of electric energy actually consumed by each node in different cycles, and multiple historical power load data, that is, the instantaneous power load size borne by each node in the corresponding time period.

[0023] Step S140: Add the multiple historical electricity consumption data and the multiple historical electricity load data to the multiple historical electricity consumption data. Specifically, the collected multiple historical electricity consumption data and the multiple historical electricity load data are added to the multiple historical electricity consumption data to build a comprehensive and systematic historical electricity consumption database. By combining the newly collected data with the existing historical data, a more complete presentation of the microgrid's electricity consumption history can be achieved, providing a rich and reliable data foundation for subsequent analysis, modeling, and optimization.

[0024] Step S200: constructing a historical power change graph based on the plurality of historical power consumption data.

[0025] Specifically, the acquired multiple historical electricity consumption data are sorted and classified. The data include information such as electricity consumption and electricity load in different time periods and at different nodes. During the sorting process, it is necessary to ensure the accuracy and completeness of the data, and to process the existing errors or missing values. The first coordinate axis and the second coordinate axis are determined based on multiple historical electricity consumption data. The first coordinate axis represents time, such as a specific year, month, date or hour, etc.; the second coordinate axis is used to represent the numerical value of electricity consumption, such as kilowatt-hour or megawatt-hour, etc. The third coordinate axis and the fourth coordinate axis are determined based on multiple historical electricity load data. The third coordinate axis also represents time, which is consistent with the time scale of the first coordinate axis; the fourth coordinate axis is used to represent the numerical value of the electricity load, such as kilowatts or megawatts. After the coordinate axes are determined, A historical electricity consumption coordinate system is constructed based on the first coordinate axis and the second coordinate axis. In this coordinate system, the electricity consumption data corresponding to different time points are marked and plotted to form a curve showing the change of electricity consumption over time, showing the change trend of historical electricity consumption. A historical electric load coordinate system is constructed based on the third coordinate axis and the fourth coordinate axis. The electric load data at different time points are marked in this coordinate system to form a curve showing the change of electric load over time, which clearly shows the change trend of historical electric load. Taking into account the change trend of historical electricity consumption and the change trend of historical electric load, a complete historical electricity change graph is constructed. The change graph can intuitively reflect the dynamic changes of power usage and load of the microgrid in the past period of time, providing important visualization basis for subsequent analysis and prediction.

[0026] In one possible implementation, a historical power change graph is constructed based on the multiple historical power consumption data, and step S200 further includes step S210, determining a first coordinate axis and a second coordinate axis based on the multiple historical power consumption data. Specifically, when determining the first coordinate axis and the second coordinate axis based on the multiple historical power consumption data, the characteristics of the historical power consumption data will be analyzed. The first coordinate axis will select a time dimension, such as a day, week, month, or year as a unit, so as to clearly show the change in power consumption over time. The second coordinate axis is used to represent the numerical value of power consumption, and its unit may be kilowatt-hour, megawatt-hour, etc. Through selection, it can lay the foundation for the subsequent construction of an intuitive display of power consumption changes.

[0027] Step S220: Determine a third coordinate axis and a fourth coordinate axis based on the plurality of historical electric load data. Specifically, based on the characteristics of the data, the third coordinate axis represents time, which is consistent with the time scale of the first coordinate axis to ensure that changes in power consumption and electric load can be compared within the same time frame. The fourth coordinate axis is used to represent the value of the electric load, which may be in units of kilowatts, megawatts, etc., to accurately reflect the magnitude of the electric load.

[0028] Step S230: Construct a historical electricity consumption coordinate system based on the first and second coordinate axes. Specifically, the historical electricity consumption coordinate system is constructed based on the determined first and second coordinate axes. The first coordinate axis is scaled according to the selected time interval, and the second coordinate axis is scaled appropriately according to the numerical range of electricity consumption, providing a clear framework for depicting electricity consumption data.

[0029] Step S240: Construct a historical electric load coordinate system based on the third and fourth coordinate axes. Specifically, the historical electric load coordinate system is constructed based on the third and fourth coordinate axes. Similar to the historical electric consumption coordinate system, the third coordinate axis is marked with a time scale, and the fourth coordinate axis is divided into scales based on the electric load value range, forming a space specifically for displaying electric load data.

[0030] Step S250: Synchronize the multiple historical electricity consumption data to the historical electricity consumption coordinate system to obtain the historical electricity consumption change trend. Specifically, synchronize the multiple historical electricity consumption data to the historical electricity consumption coordinate system, accurately mark the electricity consumption value corresponding to each time point in chronological order in the coordinate system, and connect these points with lines, thereby intuitively presenting a curve of historical electricity consumption changes over time, clearly showing the change trends such as increase, decrease, and fluctuation of historical electricity consumption.

[0031] Step S260: Synchronize the multiple historical load data to the historical load coordinate system to obtain the historical load change trend. Specifically, synchronize the multiple historical load data to the historical load coordinate system, annotate the load values ​​at the corresponding time points, and connect them to form a curve to obtain the historical load change trend, including characteristics such as peaks, valleys, and stable periods of the load.

[0032] Step S270 constructs the historical power change graph based on the historical power consumption change trend and the historical power load change trend. Specifically, the historical power change graph is constructed based on the obtained historical power consumption change trend and the historical power load change trend. In this graph, the historical power consumption and load change curves are plotted on the same graph, or compared and displayed in separate graphs. A comprehensive analysis of these two trends can provide a more comprehensive understanding of the microgrid's past power usage and load change patterns, providing a strong reference basis for subsequent power management and optimization.

[0033] Step S300: constructing a load forecasting model based on the historical power change graph to forecast the microgrid and generate a plurality of power load forecasting data.

[0034] Specifically, the acquired historical power change graph is subjected to in-depth analysis and feature extraction, including observation of the periodic variation patterns of historical power consumption and power load, seasonal fluctuations, differences between working days and non-working days, and the impact of emergencies or special circumstances on power changes. According to the extracted features and patterns, a suitable load forecasting model is selected. Common models include time series models, regression models, neural network models, etc. Taking the neural network model as an example, it handles complex nonlinear relationships and is suitable for power load, which is affected by multiple factors and has more complex variation patterns. After determining the model, multiple power consumption time scales are set according to the historical power consumption cycle. The time scale can be hours, days, weeks, months, or even seasons, so as to capture the change pattern of power load from different time dimensions. Based on the set power consumption time scale, multiple power change data are extracted in combination with the historical power change graph. The data will be used as the input of the model for training and optimizing the model. The multiple extracted power change data are divided into training data set, supervision data set, and test data set. The training data set is used for model learning and parameter adjustment, the supervision data set is used to monitor and guide the model during the training process, and the test data set is used to evaluate the performance and accuracy of the model. Based on the selected BP neural network model, the training data set and the supervision data set are combined for supervised iterative training. During the training process, the model continuously adjusts the internal weights and biases to minimize the error between the predicted value and the actual value. The training process is repeated until the test data set passes the test of the training results, the prediction error of the model on the test data reaches an acceptable range, or the performance indicators of the model (such as mean square error, mean absolute error, etc.) meet the preset requirements. The model training is completed and passed the test, and the current relevant data of the microgrid is input into the model to predict the power load. The model will generate power load forecast data for multiple future time periods based on the input data and the learned rules. The forecast data covers different time points and nodes, providing important decision-making basis for the operation planning, resource allocation and load balancing optimization of the microgrid.

[0035] In one possible implementation, a load forecasting model is constructed based on the historical power change graph to forecast the microgrid and generate multiple power load forecast data. Step S300 further includes step S310, setting multiple power consumption time scales according to the historical power consumption cycle. Specifically, setting multiple power consumption time scales according to the historical power consumption cycle requires in-depth analysis of the periodicity and regularity of the historical power consumption data. For example, if the historical power consumption data shows a significant difference between weekdays and weekends, a time scale of days is set to distinguish between weekdays and weekends; or if there are seasonal power consumption peaks and troughs, a time scale of months or quarters is set. Through such detailed divisions, the characteristics of power changes can be captured more accurately.

[0036] Step S320 extracts multiple power variation data based on the multiple power consumption time scales and the historical power variation graph. Specifically, based on the multiple power consumption time scales set and combined with the historical power variation graph, multiple power variation data are extracted. This step filters and extracts corresponding data points or data segments from the historical power variation graph according to different time scales. The data reflects the changes in power consumption and power load at specific time scales, including various trends such as increases, decreases, and stability.

[0037] Step S330: Divide the plurality of power variation data into a training dataset, a supervision dataset, and a test dataset. Specifically, the extracted plurality of power variation data are divided into a training dataset, a supervision dataset, and a test dataset. This division is typically performed randomly or in a certain proportion. The training dataset is generally large and is used for the main learning process of the model, allowing the model to learn the patterns and regularities of power variation. The supervision dataset plays a monitoring and adjustment role during the training process, helping the model to better optimize parameters. The test dataset is independent of the training and supervision processes and is used to ultimately evaluate the performance and accuracy of the model.

[0038] Step S340 , based on the BP neural network, supervised iterative training is performed in combination with the training data set and the supervision data set until the test data set passes the test on the training result, and then the load forecasting model is output. Specifically, the training process is started based on the BP neural network. The BP neural network is a powerful machine learning model that can handle complex nonlinear relationships. During training, the training data set is input into the network. The network calculates the output based on the input data and the preset initial parameters, and compares it with the actual target value to calculate the error. Through the back propagation algorithm, the weights and biases in the network are adjusted according to the error to reduce the error. During the training process, the supervision data set is also combined. The supervision data set provides additional information and constraints to help the model learn and adjust more accurately to avoid overfitting or underfitting. The above training process is repeated continuously for multiple iterations to gradually optimize and converge the model. The training results are tested regularly using the test data set. If the test results do not meet the preset accuracy requirements, continue training and adjustment; until the test data set passes the test of the training results, that is, the prediction error of the model on the test data set reaches an acceptable range or meets specific performance indicators. At this time, the final load forecasting model is output. The load forecasting model can predict future power load conditions based on the new input data, providing an important reference basis for the management and optimization of microgrids.

[0039] Step S400: performing load balancing optimization based on the plurality of power load forecast data to generate a load distribution plan.

[0040] Specifically, a comprehensive compilation and analysis of the multiple power load forecast data obtained is performed. This includes detailed recording and classification of the predicted load values ​​for different time periods and different nodes, collecting the real-time power operation information set of the microgrid, and obtaining key elements such as power operation efficiency information, power transmission loss information, and power demand fluctuation information. The information can reflect the current operating status and potential problems of the microgrid. Based on the information obtained, a preset optimization target is constructed. The preset optimization target includes a preset power operation efficiency target, a preset power transmission loss target, and a preset power demand fluctuation target. For example, the preset power operation efficiency target may be to increase the overall operating efficiency to a certain percentage; the preset power transmission loss target may be to reduce the loss during the transmission process to below a certain specific value; the preset power demand fluctuation target may be to control the demand fluctuation within a certain range to ensure the stability of the power supply. According to the preset power operation efficiency target, the first load distribution data is generated in combination with multiple power load forecast data. When considering how to distribute the load, priority is given to ensuring that the power operation efficiency can be achieved. To reach the preset goal, by analyzing the load forecast data and equipment performance of each node and other factors, calculate the load distribution plan that can achieve the highest operating efficiency, generate the second load distribution data according to the preset power transmission loss target combined with multiple power load forecast data, focus on the power loss during transmission, minimize the transmission loss by optimizing line selection, adjusting load distribution, etc., generate the third load distribution data according to the preset power demand fluctuation target combined with multiple power load forecast data, aiming to balance the power demand in different time periods and avoid large fluctuations in demand, thereby ensuring the stable operation of the power system, comprehensively consider the first load distribution data, the second load distribution data and the third load distribution data, integrate and weigh them, analyze the distribution plan under each goal, find an optimal balance point, and construct the final load distribution plan, which can not only meet the requirements of operating efficiency, but also reduce transmission loss, while also stabilizing power demand fluctuations and realizing load balancing optimization of the microgrid.

[0041] In one possible implementation, load balancing optimization is performed based on the multiple power load forecast data to generate a load distribution plan. Step S400 further includes step S410, which collects a real-time power operation information set of the microgrid to obtain power operation efficiency information, power transmission loss information, and power demand fluctuation information. Specifically, by deploying sensors and monitoring equipment at key locations in the microgrid, the power operation information of the microgrid is collected in real time. The equipment will continuously collect data such as current, voltage, and power, and aggregate it into a real-time power operation information set. The information is then deeply analyzed and calculated to obtain power operation efficiency information, such as the actual operating efficiency of the equipment and energy conversion efficiency; power transmission loss information, including energy loss caused by line resistance and transformer loss; and power demand fluctuation information, i.e., the magnitude and frequency of changes in power demand within different time periods.

[0042] Step S420 constructs a preset optimization target based on the power operation efficiency information, the power transmission loss information, and the power demand fluctuation information. The preset optimization target includes a preset power operation efficiency target, a preset power transmission loss target, and a preset power demand fluctuation target. Specifically, the preset optimization target is constructed based on the acquired power operation efficiency information, power transmission loss information, and power demand fluctuation information. For power operation efficiency, a desired operating efficiency value is set as the preset power operation efficiency target based on the actual situation and expected performance of the microgrid. For power transmission loss, an acceptable maximum loss value is determined to form a preset power transmission loss target. For power demand fluctuation, an allowable fluctuation range is set as the preset power demand fluctuation target.

[0043] Step S430: Generate first load distribution data based on the preset power operation efficiency target and the plurality of power load forecast data. Specifically, the first load distribution data is generated based on the preset power operation efficiency target and the plurality of power load forecast data. The first load distribution data is generated by comprehensively considering the performance characteristics of each device and node, the current load forecast, and the preset efficiency target. An optimal load distribution scheme for each node and device is calculated using an optimization algorithm and model while meeting the preset power operation efficiency target, thereby obtaining the first load distribution data.

[0044] Step S440: Generate second load distribution data based on the preset power transmission loss target and the plurality of power load forecast data. Specifically, the second load distribution data is generated based on the preset power transmission loss target and the plurality of power load forecast data, focusing on power transmission losses. By adjusting line loads, optimizing the layout of power sources and loads, and other methods, with the goal of minimizing transmission losses, a corresponding load distribution scheme, i.e., the second load distribution data, is calculated based on the load forecast data.

[0045] Step S450 generates third load distribution data based on the preset power demand fluctuation target and the plurality of power load forecast data. Specifically, the third load distribution data is generated based on the preset power demand fluctuation target and the plurality of power load forecast data. The third load distribution data is analyzed by analyzing power demand forecasts for different time periods, as well as the supply capacity and response characteristics of the microgrid. Loads are rationally allocated to ensure that demand fluctuations are controlled within a preset range while meeting demand, thereby obtaining the third load distribution data.

[0046] Step S460: Constructing the load distribution plan based on the first load distribution data, the second load distribution data, and the third load distribution data. Specifically, the first load distribution data, the second load distribution data, and the third load distribution data are comprehensively considered, compared, integrated, and weighed, and a weighted average or multi-objective optimization algorithm or different priorities are assigned based on actual conditions are used to find an optimal solution that simultaneously takes into account power operation efficiency, transmission loss, and demand fluctuations, thereby constructing a final load distribution plan.

[0047] Step S500: Execute the load distribution scheme to perform intelligent power optimization scheduling on multiple nodes of the microgrid.

[0048] Specifically, when the load distribution plan is determined, the control system of the microgrid starts the execution process. The system will obtain the specific power distribution instructions for each node in the load distribution plan, and perform real-time power distribution and scheduling operations on multiple nodes of the microgrid according to the instructions. The system will closely monitor the power load data of each node and determine whether the multiple power load data of multiple nodes are greater than or equal to the preset threshold. If the multiple power load data are less than the preset threshold, it indicates that the power load of the node is within a relatively safe and normal range. The multiple nodes of the microgrid are monitored in real time, and the operation data of the nodes are continuously collected through sensors and monitoring equipment, including voltage, current, power factor, etc. Based on the real-time operation monitoring data, operation feedback information is generated. The feedback information includes the real-time operation status of the node, the change trend of the power parameters, etc. The load balancing evaluation of the microgrid is performed based on the operation feedback information. The evaluation process will comprehensively consider the power distribution situation of each node, the load change situation and the deviation from the preset target, etc., thereby generating a load balancing score. An emergency response scheduling strategy is generated based on the load balancing score. For example, if the score is low, it may be necessary to adjust the power distribution of certain nodes, increase or decrease the supply, to achieve more optimized load balancing. If multiple power load data are greater than or equal to the preset threshold, it means that the node may be facing overload or abnormal conditions. The abnormality of multiple nodes in the microgrid is located. Through data analysis and comparison, the multiple nodes with abnormalities are accurately determined, and weights are assigned to the multiple abnormal nodes. Multiple weight factors are generated according to factors such as the importance and load level of the node. These weight factors are arranged in descending order to generate a weight sequence. Deviation correction is performed based on the weight sequence based on multiple abnormal nodes. According to the weight, important and large-deviation nodes are prioritized. Multiple deviation correction results are generated by adjusting the power supply, switching lines, etc. Multiple abnormal nodes are corrected based on the multiple deviation correction results to ensure that the power load of the node returns to the normal range. Based on the multiple corrected nodes, the microgrid is intelligently optimized and dispatched to ensure the stable and efficient operation of the microgrid.

[0049] In one possible implementation, executing the load distribution plan to intelligently optimize power dispatch for multiple nodes of a microgrid, step S500 further includes step S510, executing the load distribution plan and determining whether multiple load data of the multiple nodes are greater than or equal to a preset threshold. Specifically, the system begins executing the pre-established load distribution plan. During execution, the system simultaneously monitors and collects real-time load data of the multiple nodes of the microgrid, compares the collected load data of the multiple nodes with the preset threshold, and determines whether the data is greater than or equal to the preset threshold.

[0050] In step S520, if the multiple electrical load data are less than the preset threshold, real-time operation monitoring of the multiple nodes of the microgrid is performed to generate operational feedback information. Specifically, if the multiple electrical load data are less than the preset threshold, it means that the electrical load of the current node is within a relatively safe and acceptable range. At this time, the system will initiate a real-time operation monitoring mechanism for the multiple nodes of the microgrid. Through sensors and monitoring equipment installed at each node, a series of key operating parameters and data, including voltage, current, and power, are continuously collected. After the data is aggregated and organized, operational feedback information that can reflect the real-time operating status of the node is generated.

[0051] Step S530: Based on the operational feedback information, the microgrid is evaluated for load balancing, generating a load balancing score. Specifically, the load balancing evaluation of the microgrid is performed based on the acquired operational feedback information. The evaluation process comprehensively considers multiple aspects, such as the power distribution status of each node, the load variation trend, and the degree of deviation from the ideal balanced state. Through calculation and analysis, a quantitative assessment result of the microgrid's load balancing status is generated, i.e., a load balancing score is generated.

[0052] Step S540: Generate an emergency response scheduling strategy based on the load balancing score, and perform intelligent power optimization scheduling on the multiple nodes. Specifically, the emergency response scheduling strategy is formulated based on the generated load balancing score. If the score is high, it indicates that the load balancing state of the microgrid is good and no major adjustments may be required. If the score is low, it means that there is an imbalance and a corresponding scheduling strategy needs to be generated. The strategy may include reallocating power resources, adjusting the power supply priority of certain nodes, or activating backup power sources. Based on the strategy, intelligent power optimization scheduling is performed on the multiple nodes to continuously optimize the operating state of the microgrid and ensure its stable and efficient power supply.

[0053] In one possible implementation, an emergency response scheduling strategy is generated based on the load balancing score, and intelligent power optimization scheduling is performed on the multiple nodes. Step S540 further includes step S541: if the multiple electrical load data are greater than or equal to the preset threshold, abnormality location is performed on the multiple nodes of the microgrid to identify multiple abnormal nodes. Specifically, when the multiple electrical load data are greater than or equal to the preset threshold, it indicates that the operation of the microgrid may be abnormal. The system will perform abnormality location on the multiple nodes of the microgrid, perform detailed analysis of the electrical load data of each node, and compare the difference with the normal range or expected value, thereby accurately identifying multiple abnormal nodes.

[0054] Step S542 assigns weights to the multiple abnormal nodes to generate multiple weight factors, which are then sorted in descending order to generate a weight sequence. Specifically, after the abnormal nodes are identified, weights are assigned to these abnormal nodes. The weight assignment may be based on factors such as the importance of the node, the critical load it carries, and its impact on the stability of the entire microgrid. Multiple weight factors are generated to reflect the relative importance of each abnormal node. The generated multiple weight factors are then sorted in descending order to form a weight sequence. The weight sequence clearly demonstrates the importance and priority of the abnormal nodes.

[0055] Step S543 performs deviation correction on the multiple abnormal nodes according to the weight sequence, generating multiple deviation correction results. Specifically, deviation correction is performed on the multiple abnormal nodes in the order of the weight sequence, with abnormal nodes with higher weights being prioritized for processing and correction. Correction methods include adjusting power supply distribution, changing line connections, activating backup power sources, or limiting some non-critical loads, so that the power load of the abnormal nodes returns to a normal range or approaches a preset ideal state. After the correction operations, multiple deviation correction results are generated.

[0056] Step S544: Correct the multiple abnormal nodes based on the multiple deviation correction results, and perform intelligent power optimization scheduling on the microgrid based on the multiple corrected nodes. Specifically, correct the multiple abnormal nodes based on the multiple deviation correction results, confirm that the power load of the abnormal nodes has been effectively adjusted and improved to meet the normal operation requirements of the microgrid, and perform overall intelligent power optimization scheduling on the microgrid based on the corrected nodes. The power distribution and operating status of the microgrid are reassessed to ensure that the entire microgrid can continue to operate stably and efficiently after the abnormal situation is handled, meet the power needs of users, and improve the reliability and cost-effectiveness of the microgrid.

[0057] The embodiment of the present application uses the method of sensing the microgrid nodes according to the historical power consumption cycle to obtain historical power consumption data, constructs a historical power change graph based on this, constructs a load prediction model according to the graph to generate power load prediction data, performs load balancing optimization based on this to obtain a load distribution plan, executes the plan to perform intelligent power optimization and scheduling of the microgrid nodes, realizes the optimal configuration and efficient utilization of power resources, and achieves the technical effect of improving the overall efficiency, stability and reliability of the microgrid operation.

[0058] In the above, refer to Figure 1 The microgrid dynamic load balancing optimization method based on historical power consumption data according to an embodiment of the present invention is described in detail. Figure 2 A microgrid dynamic load balancing optimization system based on historical power consumption data according to an embodiment of the present invention is described.

[0059] The microgrid dynamic load balancing optimization system based on historical electricity usage data, according to an embodiment of the present invention, is designed to address the existing technical issues of inaccurate microgrid load forecasting, unbalanced load distribution, and imperfect abnormal situation response and power dispatch strategies, thereby improving the overall efficiency, stability, and reliability of microgrid operations. The microgrid dynamic load balancing optimization system based on historical electricity usage data includes: a historical electricity usage data acquisition module 10, a historical electricity change graph generation module 20, a microgrid load forecasting module 30, a load distribution plan generation module 40, and a power intelligent dispatch module 50.

[0060] The historical electricity consumption data acquisition module 10 is used to perform data sensing on multiple nodes of the microgrid according to the historical electricity consumption cycle to obtain multiple historical electricity consumption data;

[0061] The historical power change graph generating module 20 is used to construct a historical power change graph based on the plurality of historical power consumption data;

[0062] The microgrid load prediction module 30 is used to construct a load prediction model based on the historical power change graph to predict the microgrid and generate multiple power load prediction data;

[0063] The load distribution plan generating module 40 is used to perform load balancing optimization based on the plurality of power load forecast data and generate a load distribution plan;

[0064] The power intelligent dispatching module 50 is used to execute the load distribution plan to perform power intelligent optimization dispatching on multiple nodes of the microgrid.

[0065] The specific configuration of the historical electricity consumption data acquisition module 10 will be described in detail below. As described above, data sensing is performed on multiple nodes of the microgrid according to the historical electricity consumption cycle to obtain multiple historical electricity consumption data. The historical electricity consumption data acquisition module 10 can further include: a cycle determination unit for determining the historical electricity consumption cycle based on the historical electricity consumption ladder information; a node setting unit for setting multiple nodes according to the historical power supply and demand information of the microgrid; a power collection unit for collecting power consumption of the microgrid based on the historical electricity consumption cycle and the supply and demand relationship of the multiple nodes to obtain multiple historical electricity consumption data and multiple historical electricity load data; and a data addition unit for adding the multiple historical electricity consumption data and the multiple historical electricity load data to the multiple historical electricity consumption data.

[0066] Below, the specific configuration of the historical power change graph generation module 20 will be described in detail. As described above, based on the multiple historical power consumption data, a historical power change graph is constructed. The historical power change graph generation module 20 may further include: a first coordinate axis and a second coordinate axis determination unit for determining the first coordinate axis and the second coordinate axis based on the multiple historical power consumption data; a third coordinate axis and a fourth coordinate axis determination unit for determining the third coordinate axis and the fourth coordinate axis based on the multiple historical electric load data; a historical power consumption coordinate system construction unit for constructing a historical power consumption coordinate system based on the first coordinate axis and the second coordinate axis; a historical electric load coordinate system construction unit for constructing a historical electric load coordinate system based on the third coordinate axis and the fourth coordinate axis; a historical power consumption change trend acquisition unit for synchronizing the multiple historical power consumption data to the historical power consumption coordinate system to obtain a historical power consumption change trend; a historical electric load change trend acquisition unit for synchronizing the multiple historical electric load data to the historical electric load coordinate system to obtain a historical electric load change trend; a historical power change graph construction unit for constructing the historical power change graph according to the historical power consumption change trend and the historical electric load change trend.

[0067] The specific configuration of the microgrid load prediction module 30 will be described in detail below. As described above, a load prediction model is constructed based on the historical power change graph to predict the microgrid and generate multiple power load prediction data. The microgrid load prediction module 30 may further include: a time scale setting unit for setting multiple power consumption time scales according to the historical power consumption cycle; a power change data extraction unit for extracting multiple power change data based on the multiple power consumption time scales combined with the historical power change graph; a data partitioning unit for dividing the multiple power change data into a training data set, a supervision data set, and a test data set; a load prediction model output unit for performing supervised iterative training based on a BP neural network in combination with the training data set and the supervision data set until the test data set passes the test of the training results, and then outputting the load prediction model.

[0068] The specific configuration of the load distribution scheme generation module 40 will be described in detail below. As described above, load balancing optimization is performed based on the multiple power load forecast data to generate a load distribution scheme. The load distribution scheme generation module 40 may further include: a power operation efficiency information acquisition unit for collecting a real-time power operation information set of the microgrid to obtain power operation efficiency information, power transmission loss information, and power demand fluctuation information; a preset target unit for constructing a preset optimization target based on the power operation efficiency information, the power transmission loss information, and the power demand fluctuation information, wherein the preset optimization target includes a preset power operation efficiency target, a preset power transmission loss target, and a preset power demand fluctuation target; a first load distribution data generation unit for generating first load distribution data according to the preset power operation efficiency target combined with the multiple power load forecast data; a second load distribution data generation unit for generating second load distribution data according to the preset power transmission loss target combined with the multiple power load forecast data; a third load distribution data generation unit for generating third load distribution data according to the preset power demand fluctuation target combined with the multiple power load forecast data; and a load distribution scheme construction unit for constructing the load distribution scheme based on the first load distribution data, the second load distribution data, and the third load distribution data.

[0069] The specific configuration of the power intelligent dispatching module 50 will be described in detail below. As described above, the load distribution scheme is executed to perform power intelligent optimization and dispatch on multiple nodes of the microgrid. The power intelligent dispatching module 50 may further include: an electric load data judgment unit for executing the load distribution scheme and judging whether the multiple electric load data of the multiple nodes are greater than or equal to a preset threshold; an operation feedback information generation unit for performing real-time operation monitoring of the multiple nodes of the microgrid and generating operation feedback information if the multiple electric load data are less than the preset threshold; a load balancing score generation unit for performing load balancing evaluation of the microgrid based on the operation feedback information and generating a load balancing score; and an emergency response scheduling strategy generation unit for generating an emergency response scheduling strategy based on the load balancing score to perform power intelligent optimization and dispatch on the multiple nodes.

[0070] Among them, an emergency response scheduling strategy is generated according to the load balancing score, and power intelligent optimization scheduling is performed on the multiple nodes. The emergency response scheduling strategy generation unit may further include: an abnormal node determination subunit is used to locate the abnormalities of the multiple nodes of the microgrid if the multiple electric load data are greater than or equal to the preset threshold, and determine multiple abnormal nodes; a weight sequence generation subunit is used to assign weights to the multiple abnormal nodes, generate multiple weight factors, arrange the multiple weight factors in descending order, and generate a weight sequence; a deviation correction result generation subunit is used to perform deviation correction based on the multiple abnormal nodes according to the weight sequence, and generate multiple deviation correction results; an intelligent optimization scheduling subunit is used to correct the multiple abnormal nodes according to the multiple deviation correction results, and perform power intelligent optimization scheduling on the microgrid based on the multiple corrected nodes.

[0071] The microgrid dynamic load balancing optimization system based on historical electricity consumption data provided by an embodiment of the present invention can execute the microgrid dynamic load balancing optimization method based on historical electricity consumption data provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0072] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0073] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. 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.

Claims

1. A dynamic load balancing optimization method for microgrids based on historical electricity consumption data, characterized in that: The method comprises: Perform data sensing on multiple nodes of the microgrid according to the historical power consumption cycle to obtain multiple historical power consumption data; constructing a historical power change graph based on the plurality of historical power consumption data; Building a load forecasting model based on the historical power change graph to forecast the microgrid and generate multiple power load forecast data; Perform load balancing optimization based on the plurality of power load forecast data to generate a load distribution plan; Executing the load distribution scheme to perform intelligent power optimization scheduling on multiple nodes of the microgrid; The method of performing data sensing on multiple nodes of the microgrid according to the historical power consumption cycle to obtain multiple historical power consumption data includes: Determining the historical electricity usage cycle based on historical electricity usage step information; According to the historical power supply and demand information of the microgrid, multiple nodes are set; According to the historical electricity consumption cycle, combined with the supply and demand relationship of the multiple nodes, electricity consumption of the microgrid is collected to obtain multiple historical electricity consumption data and multiple historical electricity load data; adding the plurality of historical electricity consumption data and the plurality of historical electricity load data to the plurality of historical electricity consumption data; The method of performing load balancing optimization based on the plurality of power load forecast data and generating a load distribution plan includes: Collect real-time power operation information of the microgrid to obtain power operation efficiency information, power transmission loss information, and power demand fluctuation information; Constructing a preset optimization target based on the power operation efficiency information, the power transmission loss information, and the power demand fluctuation information, wherein the preset optimization target includes a preset power operation efficiency target, a preset power transmission loss target, and a preset power demand fluctuation target; generating first load distribution data according to the preset power operation efficiency target and the plurality of power load forecast data; generating second load distribution data according to the preset power transmission loss target and the plurality of power load forecast data; generating third load distribution data according to the preset power demand fluctuation target and the plurality of power load forecast data; constructing the load distribution plan based on the first load distribution data, the second load distribution data, and the third load distribution data; The method of executing the load distribution scheme to perform intelligent power optimization scheduling on multiple nodes of the microgrid includes: Executing the load distribution plan to determine whether multiple electrical load data of the multiple nodes are greater than or equal to a preset threshold; If the plurality of electric load data are less than the preset threshold, real-time operation monitoring is performed on the plurality of nodes of the microgrid to generate operation feedback information; Performing load balancing evaluation on the microgrid based on the operation feedback information to generate a load balancing score; Generate an emergency response scheduling strategy based on the load balancing score, and perform intelligent power optimization scheduling on the multiple nodes; If the plurality of electric load data are greater than or equal to the preset threshold, abnormality positioning is performed on the plurality of nodes of the microgrid to determine a plurality of abnormal nodes; Performing weight assignment on the multiple abnormal nodes to generate multiple weight factors, and arranging the multiple weight factors in descending order to generate a weight sequence; Performing deviation correction based on the multiple abnormal nodes according to the weight sequence to generate multiple deviation correction results; The multiple abnormal nodes are corrected according to the multiple deviation correction results, and the microgrid is intelligently optimized and dispatched based on the multiple corrected nodes.

2. The microgrid dynamic load balancing optimization method based on historical power consumption data according to claim 1, characterized in that: The method of constructing a historical power change graph based on the plurality of historical power consumption data includes: Determine a first coordinate axis and a second coordinate axis based on the plurality of historical electricity consumption data; determining a third coordinate axis and a fourth coordinate axis based on the plurality of historical electric load data; Constructing a historical electricity consumption coordinate system based on the first coordinate axis and the second coordinate axis; Constructing a historical electric load coordinate system based on the third coordinate axis and the fourth coordinate axis; Synchronizing the plurality of historical electricity consumption data to the historical electricity consumption coordinate system to obtain a historical electricity consumption change trend; Synchronizing the plurality of historical electric load data to the historical electric load coordinate system to obtain a historical electric load change trend; The historical power change graph is constructed according to the historical power consumption change trend and the historical power load change trend.

3. The microgrid dynamic load balancing optimization method based on historical power consumption data according to claim 1, characterized in that: The load forecasting model and method include: Setting multiple electricity consumption time scales according to the historical electricity consumption cycle; extracting a plurality of power change data based on the plurality of power consumption time scales in combination with the historical power change graph; Dividing the plurality of power change data into a training data set, a supervision data set, and a test data set; Based on the BP neural network, supervised iterative training is performed in combination with the training data set and the supervision data set until the test data set passes the test on the training result, and then the load forecasting model is output.

4. A microgrid dynamic load balancing optimization system based on historical electricity consumption data, characterized in that: The system is used to implement the microgrid dynamic load balancing optimization method based on historical electricity consumption data according to any one of claims 1 to 3, and the system includes: A historical electricity consumption data acquisition module is used to perform data sensing on multiple nodes of the microgrid according to historical electricity consumption cycles to obtain multiple historical electricity consumption data; A historical power change graph generating module, the historical power change graph generating module is used to construct a historical power change graph based on the plurality of historical power consumption data; A microgrid load prediction module is used to construct a load prediction model based on the historical power change graph to predict the microgrid and generate multiple power load prediction data; A load distribution plan generation module, configured to perform load balancing optimization based on the plurality of power load forecast data and generate a load distribution plan; The power intelligent dispatching module is used to execute the load distribution plan to perform power intelligent optimization and dispatch on multiple nodes of the microgrid.

Citation Information

Patent Citations

  • Short-term load prediction method based on multiple views

    CN117313039A