Novel power system source network load storage coordinated optimization planning method

Abstract

The invention discloses a novel electric power system source network load storage coordinated optimization planning method, and relates to the technical field of electric power system optimization, and the method comprises the steps: S1, building a load prediction model based on various user side load dynamic adjustment data, and improving the prediction precision, S2, determining parameters and constraint conditions of each link of source network load storage of an electric power system, s3, carrying out source network load storage collaborative optimization scheduling by utilizing an optimization algorithm; S4, evaluating and verifying an optimization result; according to the novel power system source-network-load-storage coordinated optimization planning method, the precision of load prediction is greatly improved, so that the problem that supply and demand matching is difficult is effectively solved, rapid fluctuation or sudden change of user loads can be rapidly responded, the adaptability and flexibility of a scheduling strategy are remarkably improved, and the scheduling efficiency is improved. The problems of supply and demand matching and scheduling flexibility caused by low load prediction precision under the condition of dynamic adjustment of various user side loads in the prior art are solved.

Description

Technical Field

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

[0002] In modern power systems, the coordinated optimization planning method of sources, grids, loads and storage is an important technical means to achieve high efficiency and reliability of power system operation through the coordinated optimization of the power generation side (source), power grid (grid), user side load (load) and energy storage system (storage).

[0003] However, in practical applications, especially when the dynamic adjustment of user-side load is frequent, due to the low load forecasting accuracy, the existing coordinated optimization planning methods face the following major problems: First, the lack of load forecasting accuracy leads to difficulties in matching supply and demand. When the load demand changes rapidly, it is difficult for traditional forecasting methods to capture the load change trend in a timely and accurate manner, so that the power generation side cannot adjust the output according to actual demand, resulting in insufficient power supply capacity or waste of resources. Secondly, there is a lack of scheduling flexibility. Due to the lack of sufficient consideration of dynamic load adjustment, the existing planning methods are usually based on fixed load models and scheduling strategies, which are difficult to effectively adapt to the rapid fluctuations of user-side loads and reduce the system's ability to respond to sudden changes in load demand. In addition, the coordination between the various links of source, grid, load and storage is difficult to achieve. In the prior art, the optimization objectives and constraints of each link are often designed independently and lack global coordination. This fragmented planning method not only reduces the overall optimization efficiency, but also may cause safety risks in system operation. Summary of the invention

[0004] The purpose of the present invention is to provide a novel power system source-grid-load-storage coordinated optimization planning method to solve the problems of supply and demand matching and scheduling flexibility caused by low load forecasting accuracy under the condition of dynamic adjustment of multiple user-side loads in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a novel power system source-grid-load-storage coordinated optimization planning method, the method comprising:

[0006] S1. Based on the data of dynamic adjustment of load on the user side, a load forecasting model is established to improve the forecasting accuracy, including defining the influencing factors of dynamic adjustment of load on the user side, collecting data, calculating the prior probability and conditional probability of each factor based on historical data, updating the load forecast value, and iteratively improving the model to improve the forecasting accuracy. The specific formula is as follows;

[0007] Among them, P(A|B) represents the predicted value of future load based on known conditions, P(B|A) represents the degree of correlation between a certain load value and a certain condition in past data, P(A) represents the probability of a certain load value occurring without considering any conditions, P(B) represents the frequency of occurrence of a specific condition in historical data, A represents the predicted result of user-side load, and B represents the condition for dynamic adjustment of user-side load;

[0008] S2. Determine the parameters and constraints of each link of the power system including source, grid, load and storage;

[0009] S3. Use optimization algorithms to optimize the coordinated dispatch of sources, grids, loads and storage;

[0010] S4. Evaluate and verify the optimization results.

[0011] Preferably, S2 includes:

[0012] The energy input and output relationship of the power system is clarified, and the energy parameters of each link are calculated based on the actual operation data. Constraints are set, including the power range of the power generation side, the charging and discharging limits of the energy storage system, and the peak and valley characteristics of the load demand, and a constraint model is constructed. The specific formula is:

[0013] E 源 +E 储 =E 荷 +E 网损 ;

[0014] Among them, E 源 Indicates the energy provided by the power generation side, E 储 Represents the energy released by the energy storage system, E 荷 Indicates the energy required by the load side, E 网损 Represents the energy loss of the power grid.

[0015] Preferably, S3 includes:

[0016] Define the objective function, minimize the operating cost as the optimization goal, set the constraints, the specific formula is:

[0017] minimize

[0018] The constraints are:

[0019] Where C represents the optimization target value of the entire power system operation, c i represents the unit cost of each link, x i represents the decision variables of each link, a ij The coefficient matrix representing the constraints, b jIt represents the right-hand side value of the constraint condition, i represents the specific decision variable of each link, j represents the number of the constraint condition, and n represents the total number of variables.

[0020] Preferably, S4 includes:

[0021] Compare the optimization results with the actual operation data, calculate the mean absolute error, analyze the source of the error, adjust the optimization model parameters for the part with large error, and re-evaluate. The specific formula for calculating the mean absolute error is:

[0022]

[0023] Where D represents the mean absolute error, m represents the total number of data points evaluated, l represents the number of each specific data point, and y l Indicates the real data recorded during actual runtime. Represents the result calculated by the optimization model.

[0024] Preferably, S1 also includes collecting various user-side load historical data and relevant influencing factor data, constructing a multivariate load forecasting model based on the historical data and relevant influencing factor data, and continuously correcting and optimizing the parameters of the load forecasting model using actual operation data.

[0025] Preferably, the influencing factors of the dynamic adjustment of the user-side load in S1 include time, weather conditions and user electricity consumption behavior patterns.

[0026] Preferably, the parameters of each link in S2 include the maximum power output range of the power generation equipment, the charging and discharging power limit of the energy storage system, the transmission capacity of the power grid, and the dynamic change range of the load demand.

[0027] Preferably, the data for dynamic adjustment of user-side load in S1 includes seasonal power consumption trends and holiday power consumption patterns.

[0028] Preferably, S2 further includes:

[0029] The power range on the power generation side includes the minimum start and stop power constraints of the power generation equipment and the maximum charge and discharge power constraints of the energy storage system.

[0030] Preferably, S2 also includes describing the peak and valley characteristics of load demand through a dynamic segmentation model to match the fluctuation pattern of user-side load.

[0031] It can be seen from the above technical solution that the present invention has the following beneficial effects:

[0032] This new power system source-grid-load-storage coordinated optimization planning method, based on the data of dynamic adjustment of multiple user-side loads, establishes a load forecasting model to improve the prediction accuracy, determines the parameters and constraints of each link of the source-grid-load-storage of the power system, uses the optimization algorithm to perform coordinated optimization scheduling of the source-grid-load-storage, evaluates and verifies the optimization results, and greatly improves the accuracy of load forecasting, thereby effectively solving the problem of difficult supply and demand matching, and can quickly respond to the rapid fluctuations or sudden changes of user loads, significantly improve the adaptability and flexibility of the scheduling strategy, reduce the risk of insufficient power supply capacity or waste of resources caused by prediction deviation, realize the coordinated optimization of each link of the source-grid-load-storage, improve the efficiency and effect of the overall planning, reduce the safety risks in operation, improve the utilization efficiency of energy storage equipment, and smooth the volatility of renewable energy output. The operating efficiency is improved, meeting the requirements of high efficiency and reliability, with wide applicability and practical value, and solving the problems of supply and demand matching and scheduling flexibility caused by low load forecasting accuracy in the existing technology under the condition of dynamic adjustment of multiple user-side loads. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] like Figure 1 As shown, the present invention provides a technical solution: a novel power system source-grid-load-storage coordinated optimization planning method, the method comprising:

[0036] S1. Based on the data of dynamic adjustment of load on the user side, a load forecasting model is established to improve the forecasting accuracy, including defining the influencing factors of dynamic adjustment of load on the user side, collecting data, calculating the prior probability and conditional probability of each factor based on historical data, updating the load forecast value, and iteratively improving the model to improve the forecasting accuracy. The specific formula is as follows;

[0037] Among them, P(A|B) represents the predicted value of future load based on known conditions, P(B|A) represents the degree of correlation between a certain load value and a certain condition in past data, P(A) represents the probability of a certain load value occurring without considering any conditions, P(B) represents the frequency of occurrence of a specific condition in historical data, A represents the predicted result of user-side load, and B represents the condition for dynamic adjustment of user-side load;

[0038] S2. Determine the parameters and constraints of each link of the power system including source, grid, load and storage;

[0039] S3. Use optimization algorithms to optimize the coordinated dispatch of sources, grids, loads and storage;

[0040] S4. Evaluate and verify the optimization results.

[0041] This method analyzes the dynamic adjustment data of user-side loads, combines the conditional probability and prior probability in historical data, and gradually iterates to improve the accuracy of the load forecasting model. Its core lies in building a mathematical model to achieve accurate prediction of dynamic load adjustment. At the same time, it combines the parameters and constraints of each link of source, grid, load and storage to build an overall optimization model, and uses the optimization algorithm to achieve collaborative scheduling. Through the evaluation and verification steps, the feasibility and accuracy of the optimization results are ensured. By introducing the calculation of conditional probability and prior probability, the model is dynamically adjusted to improve the accuracy of load forecasting, providing accurate basic data for subsequent optimization. Through the source, grid, load and storage collaborative optimization method, resources are reasonably allocated, the operating cost of the system is reduced, and the energy utilization efficiency is improved. By evaluating and verifying the optimization results, it is ensured that the model is not only feasible in theory, but also can achieve stable performance in actual operation. This method can adapt to the needs of different types of power systems and has strong flexibility and scalability.

[0042] S2 includes clarifying the energy input and output relationship of the power system, calculating the energy parameters of each link based on the actual operation data, setting constraints, including the power range of the power generation side, the charging and discharging restrictions of the energy storage system, and the peak and valley characteristics of the load demand, and building a constraint model. The specific formula is:

[0043] E 源 +E 储 =E 荷 +E 网损 ;

[0044] Among them, E 源 Indicates the energy provided by the power generation side, E 储 Represents the energy released by the energy storage system, E 荷 Represents the energy required by the load side, E 网损 Represents the energy loss of the power grid.

[0045] By clarifying the balance relationship between energy input and output in the power system, using formula E 源 +E 储 =E 荷 +E 网损, which can accurately describe the dynamic balance of each energy link in the system. According to the actual operation data, the energy parameters of each link are calculated, and a complete set of constraint models are constructed in combination with the power range limit on the power generation side, the charge and discharge constraints of the energy storage system, and the peak and valley characteristics of the load side. This model will be used as the basic framework of the optimization algorithm to ensure that the optimization scheduling is technically feasible, while improving the economy and stability of the system operation. By clarifying the input and output energy relationship, it is possible to accurately predict and regulate the flow of electricity, achieve a dynamic balance between power generation, energy storage and load, build constraints based on actual operation data, combine theoretical models with actual needs, and effectively reduce deviations. By adding the charge and discharge limits of the energy storage system and the peak and valley characteristics of the load, it is possible to better cope with the uncertainty of load fluctuations and grid energy loss. By reasonably setting the power range and energy storage limits on the power generation side, energy scheduling is optimized, energy utilization efficiency is improved, and operating costs are reduced.

[0046] S3 includes defining the objective function, minimizing the operating cost as the optimization goal, and setting constraints. The specific formula is:

[0047] minimize

[0048] The constraints are:

[0049] Where C represents the optimization target value of the entire power system operation, c i represents the unit cost of each link, x i represents the decision variables of each link, a ij The coefficient matrix representing the constraints, b j It represents the right-hand side value of the constraint condition, i represents the specific decision variable of each link, j represents the number of the constraint condition, and n represents the total number of variables.

[0050] The optimization process takes minimizing the operating cost as the core goal, and constructs the objective function The cost optimization of each link of the power system is realized. The set constraints ensure the feasibility of the optimization results in technology and operation. For example, through the coefficient matrix a ij and the right-hand value b j Describe the physical limitations, resource constraints, and operational requirements of each link to prevent unrealistic results from appearing during optimization. iIt is used to characterize the operating status and scheduling plan of each link. The calculation of the objective function combines unit cost and decision variables to achieve global cost optimization. By establishing clear objective functions and constraints, the decision-making of each link is optimized and configured to effectively reduce the overall operating cost. The setting of constraints in the model fits the actual operating needs, such as resource limitations and load requirements, to ensure that the optimization scheme can be directly applied to the actual system. The optimization model has a high degree of parameterization and can adjust variables and constraints according to the scale and needs of different power systems. It has good adaptability and scalability. By reasonably allocating resources and optimizing scheduling, it can minimize resource waste and improve the operating efficiency of the power system.

[0051] S4 includes comparing the optimization results with the actual operation data, calculating the mean absolute error, analyzing the source of the error, adjusting the optimization model parameters for the part with large errors, and re-evaluating. The specific formula for calculating the mean absolute error is:

[0052]

[0053] Where D represents the mean absolute error, m represents the total number of data points evaluated, l represents the number of each specific data point, and y l Indicates the real data recorded during actual runtime. Represents the result calculated by the optimization model.

[0054] This method compares the calculation results of the optimization model with the actual operation data point by point and calculates the mean absolute error D to evaluate the accuracy of the optimization model. The specific steps are as follows:

[0055] For each evaluation data point, the difference between the predicted value and the actual value is calculated, and the absolute error is calculated. The errors of all data points are averaged to obtain the average absolute error D. Data points with large errors are classified to identify factors that may cause errors, such as unreasonable model parameter settings and changes in the actual operating environment. Model parameters are adjusted according to the source of errors to make the optimization results closer to the actual operating data. After adjustment, the optimization model is re-run, the error is calculated again, and the optimization is iterated until the error reaches an acceptable range. Through this closed-loop feedback mechanism, the accuracy of the optimization model can be continuously improved, and the operating efficiency and reliability of the system can be improved. Through the comparison of actual data and error feedback, the model parameters are dynamically adjusted so that the model can better adapt to the actual operating conditions. Error source analysis can help the optimization model cope with dynamic changes in operating conditions, such as load fluctuations and changes in equipment status. Through multiple evaluations and adjustments, an improved closed loop of the optimization model is formed to ensure the long-term applicability and stability of the model. Based on the evaluation of actual operating data, the optimization results are not only theoretically reasonable, but also practically feasible.

[0056] S1 also includes collecting various user-side load historical data and related influencing factor data, building a multivariate load forecasting model based on the historical data and related influencing factor data, and using actual operation data to continuously correct and optimize the parameters of the load forecasting model.

[0057] This implementation method is based on the historical data of user-side load and the data of influencing factors, and by constructing a multivariable load forecasting model, it is possible to accurately predict the dynamic changes of user-side load. The specific steps include:

[0058] Collect historical data of user-side load, including daily load curves, seasonal changes, and abnormal power consumption records, and obtain data on related factors that affect the load, such as temperature, humidity, working day and non-working day characteristics, and other external conditions. Build a load forecasting model based on methods such as multivariate regression analysis or time series analysis. Introduce multiple variables (such as time, temperature, holidays, etc.) through the model to improve the model's responsiveness to dynamic changes in load. In the actual operation process, use real-time operation data to continuously correct model parameters. Through error analysis (such as calculating the error between the predicted value and the actual value), dynamically adjust the weight coefficient in the model or modify the variable relationship to improve the prediction accuracy. Combine historical data with real-time data, and continuously improve model performance through iterative optimization methods (such as least squares or machine learning algorithms) to ensure that the model has high accuracy and adaptability in various scenarios. Through the introduction and real-time correction of multivariate models, it is possible to capture the complex laws of load changes and significantly improve the accuracy of load forecasting. The parameters of the load forecasting model can be dynamically adjusted with real-time data, and can quickly adapt to changes in electricity demand, such as seasonal fluctuations or sudden load fluctuations. High-precision load forecasting provides reliable input data for the coordinated optimization of source, grid, load and storage, reducing the uncertainty in the scheduling process. By combining historical data and real-time data, the model can be continuously improved in the long-term operation to form data-driven dynamic optimization capabilities.

[0059] The factors affecting the dynamic adjustment of user-side load in S1 include time, weather conditions and user electricity consumption behavior patterns. In this embodiment, by defining and identifying three key factors affecting the dynamic adjustment of user-side load (time, weather conditions, and user electricity consumption behavior patterns), the load change trend can be predicted more accurately and dynamically adjusted: the time factor includes the daily load curve (such as peak and valley periods), seasonal changes (differences in electricity consumption between summer and winter), and specific time nodes (such as holidays, weekdays and weekends). The introduction of the time factor can capture the periodic characteristics of the load. Weather conditions include meteorological factors such as temperature, humidity, rainfall, and wind speed. For example, the air conditioning load may increase significantly in hot weather; while in cold weather, the heating load becomes the main variable. The user's electricity consumption behavior pattern includes the user's electricity consumption habits (such as daily peak electricity consumption time), industry characteristics (such as differences in electricity consumption between commercial and residential users), and the behavior pattern of responding to demand management. Such data can be obtained through historical electricity consumption records and user-side equipment monitoring systems. By combining these three factors, the dynamic adjustment of the load can comprehensively consider the regularity of time, weather volatility, and the diversity of user behavior, and significantly improve the accuracy of load prediction and adjustment. By comprehensively analyzing multiple key factors such as time, weather, and user behavior, the characteristics of dynamic changes in load can be accurately captured and prediction errors can be reduced. The impact of time periodicity and weather volatility on load is comprehensively considered, and the diversity of user behavior patterns is incorporated to make the load adjustment strategy more adaptable to actual needs. High-precision load forecasting provides a reliable data foundation for the coordinated optimization of source, grid, load and storage, which can effectively reduce waste in energy scheduling and improve the operating efficiency of the overall system. The introduction of user electricity consumption behavior patterns supports the application of demand response mechanisms, which can better cope with peak loads and sudden changes and improve the flexibility and stability of the system.

[0060] The parameters of each link in S2 include the maximum power output range of the power generation equipment, the charging and discharging power limit of the energy storage system, the transmission capacity of the power grid, and the dynamic range of load demand. In this embodiment, by accurately defining and limiting the parameters of each link of the source, grid, load and storage, it is ensured that the optimization model can meet the operating requirements and constraints of the system. Specifically as follows: Maximum power output range of power generation equipment: The output power of each power generation equipment is limited by the equipment performance and operating conditions. For example, thermal power generators, wind turbines and photovoltaic power generation systems all have their maximum power output limits (such as rated power), and it is necessary to ensure that the power generation power does not exceed this range in the optimization scheduling. Charging and discharging power limit of energy storage system: Energy storage systems (such as battery energy storage systems) have power limits during the charging and discharging process, which depends on the design parameters of the equipment and the current state of charge (SOC). For example, the charging power cannot exceed the maximum charging power of the equipment, and the discharging power must be within the safe working range of the equipment. Transmission capacity of the power grid: The transmission capacity of the power grid (i.e., the maximum transmission power of the line) is a key constraint parameter for the operation of the power grid. Overload of the transmission capacity may cause power grid failure. In the optimization, the transmission capacity limit of each line needs to be included in the constraint conditions to ensure safe operation. Dynamic range of load demand: Load demand changes dynamically with time and external conditions (such as weather, user behavior, etc.). The optimization model needs to set the upper and lower limits of load demand based on historical data and forecast data, and ensure that the scheduling plan meets the real-time load demand. Through accurate modeling of the above parameters, a constraint model can be constructed to provide accurate input data and technical support for the coordinated optimization of source, grid, load and storage. Incorporating the parameter restrictions of power generation equipment, energy storage system, power grid and load into the optimization model can effectively avoid the irrationality of the operation plan and ensure the safe and stable operation of the system. Comprehensively considering the constraints of each link of source, grid, load and storage can achieve efficient resource utilization and global optimization of the scheduling plan. By introducing the dynamic range of load changes, the optimization model can maintain real-time response capabilities to load demand in multiple scenarios and improve the flexibility and adaptability of the system. The setting of parameter constraints reduces the risk of equipment overload and system failure, while extending the service life of the equipment and improving the reliability of the system.

[0061] The data for dynamic adjustment of user-side load in S1 include seasonal power consumption trends and holiday power consumption patterns. This implementation method incorporates seasonal power consumption trends and holiday power consumption patterns into the load forecasting and optimization model by analyzing the data for dynamic adjustment of user-side load. The specific steps are as follows: Seasonal power consumption trend analysis: The power demand of users usually has significant seasonal changes. For example, the load demand increases significantly due to the use of air conditioning in summer, while the operation of heating equipment in winter may bring additional load demand. By analyzing the seasonal power consumption trends in historical data, the power consumption change rules of each season can be extracted and used to predict and adjust the load. Holiday power consumption pattern analysis: The power consumption pattern of holiday users is usually different from that of daily working days. For example, the power consumption of residential users may increase during holidays, while the power consumption of industrial users may decrease. By extracting historical holiday power consumption data, the corresponding pattern characteristics are established, and the load is dynamically adjusted and predicted according to the holiday arrangements. Data integration and model construction: The seasonal power consumption trend and holiday power consumption pattern are used as the key input of dynamic adjustment data and integrated into the load forecasting model to form a multi-dimensional load adjustment mechanism. The model can be dynamically updated based on time series analysis or machine learning algorithms to ensure prediction accuracy. Real-time update and optimization: In actual operation, the load forecasting model is corrected according to the real-time monitoring data, and the prediction results of the load trend are dynamically adjusted so that the model can adapt to real-time changes. This implementation method can significantly improve the ability and accuracy of dynamic load adjustment by introducing seasonal power consumption trends and holiday power consumption patterns, which is mainly reflected in the following aspects: Incorporating seasonal changes and holiday patterns into the load forecasting model can effectively capture the periodicity and specific time period characteristics of user power consumption and improve the prediction accuracy. After the load characteristics of different seasons and holidays are accurately predicted, the optimization model can more flexibly dispatch source, network, load and storage resources to adapt to load demands in different time periods. Comprehensive consideration of seasonal and holiday data avoids prediction deviations caused by ignoring load characteristics at specific time points and optimizes the accuracy of the scheduling plan. The inclusion of holiday patterns enables the system to better cope with sudden load changes (such as power consumption peaks caused by extended holidays or large-scale holiday activities).

[0062] S2 also includes the power range on the power side, including the minimum start-stop power constraint of the power generation equipment and the maximum charge-discharge power constraint of the energy storage system. This embodiment provides an accurate power range limit for the optimization scheduling by clarifying the minimum start-stop power constraint of the power generation equipment and the maximum charge-discharge power constraint of the energy storage system, thereby ensuring the stability and economy of the system operation. The specific steps are as follows: Minimum start-stop power constraint of the power generation equipment: The power generation equipment usually has a minimum start-stop power (i.e., the minimum output power) during operation, and the equipment cannot maintain stable operation below this value. The optimization model needs to use this minimum start-stop power as a constraint to ensure that the startup and operation of the equipment during the scheduling process are within a safe range. For example, some thermal power units may need to maintain a certain load rate (such as 20% rated power) to avoid damage to the equipment caused by frequent start-stop. Maximum charge-discharge power constraint of the energy storage system: The charge-discharge power of the energy storage system is limited by the equipment design parameters and the current state of charge (SOC). For example, the maximum charging power and maximum discharging power of the battery energy storage system cannot exceed the equipment rating, and the power limit needs to be dynamically adjusted according to the current state of charge. This constraint needs to be included in the optimization range in the model to prevent the performance of the energy storage equipment from being degraded or damaged due to over-power operation. Parameter modeling: In the optimization model, the minimum start-stop power of the power generation equipment and the maximum charge-discharge power of the energy storage system are defined as constraints, and are jointly optimized in combination with other system parameters to ensure that the power scheduling scheme meets the safety and economic requirements of equipment operation. This implementation method can significantly improve the operating efficiency and safety of the system by introducing the minimum start-stop power constraints of the power generation equipment and the maximum charge-discharge power constraints of the energy storage system. Specifically, the minimum start-stop power constraints of the power generation equipment prevent the equipment from starting and stopping frequently, effectively reducing the damage to the equipment life during operation, while ensuring stable power supply. The maximum charge-discharge power constraints of the energy storage system avoid overload operation of the equipment, extend the life of the energy storage equipment, and reduce operation and maintenance costs. Incorporating power constraints into the optimization model ensures that the scheduling results are technically feasible and avoids violations of equipment operation restrictions. By reasonably setting the power range, the optimization model can more efficiently allocate power generation and energy storage resources to achieve optimal resource utilization.

[0063] S2 also includes describing the peak and valley characteristics of load demand through a dynamic segmentation model to match the fluctuation law of user-side load. This embodiment describes the peak and valley characteristics of user-side load demand by establishing a dynamic segmentation model, which can accurately match the fluctuation law of load and provide a reliable basis for the coordinated optimization of source-grid-load-storage. The specific steps include: collecting historical data of user-side load, including peak power consumption periods, valley periods, and fluctuation laws of load changes. Data sources may include smart meters, power consumption monitoring systems, etc. According to the load fluctuation law, the load demand is segmented according to time and power value. For example: peak period (a period when the load demand is higher than a certain threshold). Flat period (a period when the load demand fluctuates within a medium range). Valley period (a period when the load demand is lower than a certain threshold). The dynamic segmentation model can adjust the duration and range of the segment according to real-time load data to more accurately reflect the current load demand characteristics. Through the dynamic segmentation model, it matches the actual power consumption behavior and load fluctuation trend of the user, and identifies the load characteristics of different time periods. According to the model output, the optimization scheduling strategy of source, grid, load and storage is adjusted, for example, the energy storage system is preferentially scheduled to release energy during peak hours and the energy storage system is charged during valley hours. According to the actual operation data, the segmentation parameters (such as peak-valley thresholds) and segmentation standards of the segmentation model are continuously optimized to ensure that the model can dynamically adapt to load changes. This embodiment describes the peak-valley characteristics of load demand through a dynamic segmentation model, which can effectively improve the optimization capability of the system, specifically: the dynamic segmentation model can refine the load demand characteristics, accurately capture the load fluctuation laws of peak, flat and valley, and reduce prediction errors. The model can dynamically adjust the segmentation range, flexibly respond to changes in electricity consumption behaviors and load characteristics of different users, and ensure that the optimization scheduling results are closer to actual needs. At different stages of peak, flat and valley, the dynamic segmentation model provides a decision-making basis for the coordinated optimization of source, grid, load and storage in different time periods, and realizes efficient allocation of resources. Through the dynamic segmentation model, the optimization scheme can adapt to real-time load fluctuations, improve the system's ability to cope with peak and valley power consumption, and reduce the pressure on power grid operation.

[0064] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A novel power system source-grid-load-storage coordinated optimization planning method, characterized in that: The method comprises: S1. Based on the data of dynamic adjustment of load on the user side, a load forecasting model is established to improve the forecasting accuracy, including defining the influencing factors of dynamic adjustment of load on the user side, collecting data, calculating the prior probability and conditional probability of each factor based on historical data, updating the load forecast value, and iteratively improving the model to improve the forecasting accuracy. The specific formula is as follows; Among them, P(A|B) represents the predicted value of future load based on known conditions, P(B|A) represents the degree of correlation between a certain load value and a certain condition in past data, P(A) represents the probability of a certain load value occurring without considering any conditions, P(B) represents the frequency of occurrence of a specific condition in historical data, A represents the predicted result of user-side load, and B represents the condition for dynamic adjustment of user-side load; S2. Determine the parameters and constraints of each link of the power system including source, grid, load and storage; S3. Use optimization algorithms to optimize the coordinated dispatch of sources, grids, loads and storage; S4. Evaluate and verify the optimization results.

2. According to claim 1, a novel power system source-grid-load-storage coordinated optimization planning method is characterized by: The S2 includes: The energy input and output relationship of the power system is clarified, and the energy parameters of each link are calculated based on the actual operation data. Constraints are set, including the power range of the power generation side, the charging and discharging limits of the energy storage system, and the peak and valley characteristics of the load demand, and a constraint model is constructed. The specific formula is: AND 源 +E 储 =And 荷 +E 网损 ; Among them, E 源 Indicates the energy provided by the power generation side, E 储 Represents the energy released by the energy storage system, E 荷 Indicates the energy required by the load side, E 网损 Represents the energy loss of the power grid.

3. According to claim 1, a novel power system source-grid-load-storage coordinated optimization planning method is characterized by: The S3 includes: Define the objective function, minimize the operating cost as the optimization goal, set the constraints, and the specific formula is: Minimize The constraints are: Where C represents the optimization target value of the entire power system operation, c i represents the unit cost of each link, x i represents the decision variables of each link, a ij The coefficient matrix representing the constraints, b j It represents the right-hand side value of the constraint condition, i represents the specific decision variable of each link, j represents the number of the constraint condition, and n represents the total number of variables.

4. According to claim 1, a novel power system source-grid-load-storage coordinated optimization planning method is characterized by: The S4 includes: Compare the optimization results with the actual operation data, calculate the mean absolute error, analyze the source of the error, adjust the optimization model parameters for the part with large error, and re-evaluate. The specific formula for calculating the mean absolute error is: Where D represents the mean absolute error, m represents the total number of data points evaluated, l represents the number of each specific data point, and y l Indicates the real data recorded during actual runtime. Represents the result calculated by the optimization model.

5. According to claim 1, a novel power system source-grid-load-storage coordinated optimization planning method is characterized by: The S1 also includes collecting various user-side load historical data and related influencing factor data, building a multivariate load forecasting model based on the historical data and related influencing factor data, and continuously correcting and optimizing the parameters of the load forecasting model using actual operation data.

6. According to claim 1, a novel power system source-grid-load-storage coordinated optimization planning method is characterized by: The influencing factors of the dynamic adjustment of the user-side load in S1 include time, weather conditions and user electricity consumption behavior patterns.

7. According to claim 1, a novel power system source-grid-load-storage coordinated optimization planning method is characterized by: The parameters of each link in S2 include the maximum power output range of the power generation equipment, the charging and discharging power limit of the energy storage system, the transmission capacity of the power grid and the dynamic change range of the load demand.

8. According to claim 1, a novel power system source-grid-load-storage coordinated optimization planning method is characterized by: The data for dynamic adjustment of user-side load in S1 include seasonal power consumption trends and holiday power consumption patterns.

9. A novel power system source-grid-load-storage coordinated optimization planning method according to claim 2, characterized in that: The S2 further includes: The power range on the power generation side includes the minimum start and stop power constraints of the power generation equipment and the maximum charge and discharge power constraints of the energy storage system.

10. A novel power system source-grid-load-storage coordinated optimization planning method according to claim 2, characterized in that: The S2 also includes describing the peak and valley characteristics of load demand through a dynamic segmentation model to match the fluctuation pattern of user-side load.

Images