Method for integrating intelligent planning and load prediction of power grid

Abstract

The invention discloses a power grid intelligent planning and load prediction integration method. The method comprises the following steps: S1, acquiring load data of a power system; s2, acquiring historical load data, and predicting load data by adopting an LSTM neural network according to the historical load data and the acquired load data; s3, acquiring renewable resource generating capacity, historical meteorological data and meteorological data predicted by weather forecast, and predicting the renewable resource generating capacity by adopting a random forest according to the renewable resource generating capacity, the historical meteorological data and the meteorological data predicted by the weather forecast; s4, evaluating demand data of the power system for the energy storage system according to the predicted renewable energy power generation amount and load data; and S5, according to the demand data, generating a power grid energy storage system planning strategy including charging and discharging time and power distribution. By optimizing the operation strategy of the energy storage system, the power grid can more flexibly cope with load change and fluctuation of renewable energy sources, and the operation risk of the power grid is reduced.

Description

Technical Field

[0001] The present invention relates to the field of system planning, and in particular to a method for integrating intelligent power grid planning and load forecasting. Background Art

[0002] With the continuous expansion of power systems and the rapid integration of renewable energy, the complexity and uncertainty of power grids have increased significantly. Intelligent grid planning can achieve efficient grid operation and rational resource allocation through advanced technologies and optimization methods. However, while existing intelligent grid planning technologies have made some progress, they still have some shortcomings. For example, traditional load forecasting methods have limitations when dealing with complex nonlinear relationships and time-dependent problems. When integrated with grid planning, they often lack comprehensive consideration of renewable energy generation and energy storage system requirements. Summary of the Invention

[0003] In view of the deficiencies of the existing technology, the present invention provides an integrated method for power grid intelligent planning and load forecasting.

[0004] The integrated method of power grid intelligent planning and load forecasting includes the following steps:

[0005] S1: Collect load data of the power system;

[0006] S2: Obtain historical load data, and use an LSTM neural network to predict load data based on the historical load data and the collected load data;

[0007] S3: Obtaining renewable energy power generation, historical meteorological data, and meteorological data predicted by weather forecasts, and using random forest to predict renewable energy power generation based on the renewable energy power generation, historical meteorological data, and meteorological data predicted by weather forecasts;

[0008] S4: Evaluate the power system's demand data for the energy storage system based on the predicted renewable energy generation and load data; the demand data includes power demand, capacity demand, and energy demand;

[0009] S5: Generate a grid energy storage system planning strategy based on the demand data, including charging and discharging time and power allocation.

[0010] Preferably, the step S2 is specifically:

[0011] Acquire historical load data and preprocess the historical load data, including data cleaning, data normalization, and data segmentation;

[0012] Train the LSTM neural network model preset with the pre-processed historical load data;

[0013] The load data collected in step S1 is input into the trained LSTM neural network model to predict the load data within a preset time period, including the load curve, load peak value and load valley value.

[0014] Preferably, in step S3, specifically,

[0015] Clean renewable energy power generation, historical meteorological data, and forecasted meteorological data, including processing missing values, outliers, and noisy data;

[0016] Normalizing or standardizing the historical meteorological data and the predicted meteorological data;

[0017] Align renewable energy generation, historical meteorological data, and forecasted meteorological data in time series to form a unified dataset;

[0018] Extract key features from historical and forecasted meteorological data, including wind direction, wind speed, and solar radiation;

[0019] Train a random forest model based on key features of historical meteorological data and historical power generation;

[0020] The preprocessed forecasted meteorological data is input into the random forest model to predict the renewable energy power generation in the preset time period.

[0021] Preferably, the step S4 is specifically:

[0022] Calculating a net load curve of the power system based on the predicted renewable energy power and load data, wherein the net load curve is the difference between the load data and the renewable energy power generation;

[0023] identifying a power imbalance period of the power system according to the net load curve, wherein the power imbalance period includes a period when the power generation of the power system is insufficient or excessive;

[0024] Calculating the power demand and energy demand of the power system for the energy storage system according to the power imbalance period;

[0025] Obtaining a capacity requirement of the energy storage system based on the operating characteristics of the power system, the technical parameters of the energy storage system, and a preset response time of the energy storage system, wherein the capacity requirement includes the rated power and rated energy capacity of the energy storage system;

[0026] The power demand, energy demand and capacity demand of the energy storage system are integrated into the demand data of the power system for the energy storage system.

[0027] Preferably, the step S4 is specifically:

[0028] Obtaining a charge and discharge power range and an energy storage range of the energy storage system based on the power demand, capacity demand, and energy demand in the demand data;

[0029] Identify charging and discharging periods for energy storage systems based on the power system's net load curve and renewable energy generation forecast data;

[0030] Formulate charging and discharging time strategies based on the charging and discharging periods of the energy storage system;

[0031] Determine the power allocation strategy based on the power and capacity requirements of the energy storage system;

[0032] Obtaining a grid energy storage system planning strategy based on the charging and discharging time strategy and the power allocation strategy;

[0033] The grid energy storage system planning strategy is optimized by combining the grid topology data and the installation location of the energy storage system.

[0034] Preferably, in step S4, the grid energy storage system planning strategy is optimized by combining the grid topology data and the installation location of the energy storage system, specifically,

[0035] Determine the installation location of the energy storage system based on the grid topology, and analyze the grid power flow distribution based on the installation location of the energy storage system. The grid topology includes node information, line information, transformer information, and grid partition information.

[0036] Analyze the energy storage system's power support capabilities for key nodes and lines in the grid, based on its installation location. These are areas of the grid with concentrated loads, concentrated renewable energy access, or weak stability.

[0037] Optimize charging and discharging time strategies based on the energy storage system's installation location and grid power distribution;

[0038] Optimize power allocation strategy based on the installation location of the energy storage system and the grid zoning characteristics.

[0039] The present invention discloses an integrated method for intelligent power grid planning and load forecasting, which has the following beneficial effects: by collecting load data of the power system and combining it with historical load data, an LSTM neural network is used to perform load forecasting. This method can effectively handle nonlinear relationships and long-term dependencies in the load data, thereby significantly improving the accuracy of load forecasting. This provides a more accurate basis for intelligent power grid planning and helps optimize power grid resource allocation and operation strategies. By predicting renewable energy generation, this helps better integrate renewable energy, reduce wind and solar power curtailment, and improve the power grid's ability to accept renewable energy. Based on the predicted renewable energy generation and load data, a comprehensive assessment is made of the power system's demand for energy storage systems, including power demand, capacity demand, and energy demand. This allows for more accurate determination of the configuration requirements of the energy storage system and ensures the effective application of the energy storage system in the power grid. By generating a planning strategy for the power grid energy storage system based on the assessed demand data, the operating state of the energy storage system can be dynamically adjusted to better match the load demand and renewable energy generation, thereby improving the utilization efficiency of the energy storage system and extending its service life. By optimizing the operation strategy of the energy storage system, the power grid can more flexibly respond to load changes and the volatility of renewable energy, enhance the operational stability of the power grid, and reduce the operational risks of the power grid. This not only improves the power supply reliability of the power grid, but also provides strong support for the intelligent and low-carbon development of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 This is a flow chart of the integrated method for intelligent power grid planning and load forecasting provided by the present invention. DETAILED DESCRIPTION

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0043] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0044] refer to Figure 1The integrated method of intelligent power grid planning and load forecasting provided by the present invention comprises the following steps:

[0045] S1: Collect load data of the power system;

[0046] S2: Obtain historical load data and use the LSTM neural network to predict load data based on the historical load data and the collected load data;

[0047] S3: Obtain renewable energy power generation, historical meteorological data, and meteorological data predicted by weather forecasts, and use random forests to predict renewable energy power generation based on the renewable energy power generation, historical meteorological data, and meteorological data predicted by weather forecasts;

[0048] S4: Evaluate the power system's demand data for the energy storage system based on the predicted renewable energy generation and load data; the demand data includes power demand, capacity demand, and energy demand;

[0049] S5: Generate a grid energy storage system planning strategy based on demand data, including charging and discharging time and power allocation.

[0050] The integrated method of intelligent planning and load forecasting of power grids provided by the present invention can collect load data of the power system and combine it with historical load data to perform load forecasting using an LSTM neural network. It can effectively process nonlinear relationships and long-term dependencies in load data, thereby significantly improving the accuracy of load forecasting. This provides a more accurate basis for intelligent planning of power grids, helps to optimize power grid resource allocation and operation strategies, and by predicting renewable energy generation, it helps to better integrate renewable energy, reduce wind and solar power abandonment, and improve the power grid's ability to accept renewable energy. Based on the predicted renewable energy generation and load data, the power system's demand for energy storage systems is comprehensively evaluated, including power demand, capacity demand, and energy demand. This can more accurately determine the configuration requirements of the energy storage system and ensure the effective application of the energy storage system in the power grid. By generating a planning strategy for the power grid energy storage system based on the evaluated demand data, the operating state of the energy storage system can be dynamically adjusted to better match the load demand and renewable energy generation, improve the utilization efficiency of the energy storage system, and extend its service life. By optimizing the operation strategy of the energy storage system, the power grid can more flexibly respond to load changes and the volatility of renewable energy, enhance the operational stability of the power grid, and reduce the operational risks of the power grid. This not only improves the power supply reliability of the power grid, but also provides strong support for the intelligent and low-carbon development of the power grid.

[0051] In a preferred embodiment, step S2 is specifically:

[0052] Obtain historical load data and preprocess the historical load data, including data cleaning, data normalization and data segmentation; among them, the load curve of the power system within a preset time period, load characteristic data and load-related external factor data are used, and the external factor data includes holiday information.

[0053] The LSTM neural network model preset with the preprocessed historical load data is trained; the load data collected in step S1 is input into the trained LSTM neural network model to predict the load data within the preset time period, including the load curve, load peak, and load valley. Specifically, an LSTM neural network model is constructed, which includes an input layer, an LSTM layer, a fully connected layer, and an output layer. The input layer is used to receive historical load data and related external factor data; the LSTM layer is used to capture the time series characteristics and long-term dependencies of the load data; the fully connected layer is used to map the output of the LSTM layer to a load forecast value; and the output layer is used to generate the forecast results of the load data.

[0054] In a preferred embodiment, step S3,

[0055] Clean renewable energy power generation, historical meteorological data, and forecasted meteorological data, including processing missing values, outliers, and noisy data;

[0056] Normalize or standardize historical meteorological data and forecast meteorological data;

[0057] Align renewable energy generation, historical meteorological data, and forecasted meteorological data in time series to form a unified dataset;

[0058] Extract key features from historical and forecasted meteorological data, including wind direction, wind speed, and solar radiation;

[0059] Train a random forest model based on key features of historical meteorological data and historical power generation;

[0060] The preprocessed forecasted meteorological data is input into the random forest model to predict the renewable energy power generation in the preset time period.

[0061] In a preferred embodiment, step S4 is specifically,

[0062] Based on the predicted renewable energy power and load data, the net load curve of the power system is calculated. The net load curve is the difference between the load data and the renewable energy power generation;

[0063] According to the net load curve, the power imbalance period of the power system is identified, and the power imbalance period includes the period when the power system has insufficient or excessive power generation;

[0064] Based on the power imbalance period, the power demand and energy demand of the power system for the energy storage system are calculated; the power demand is the maximum power that the energy storage system needs to provide or absorb during the power imbalance period, and the energy demand is the total energy that the energy storage system needs to store or release during the power imbalance period.

[0065] Based on the operating characteristics of the power system, the technical parameters of the energy storage system, and the preset response time of the energy storage system, the capacity requirement of the energy storage system is obtained. The capacity requirement includes the rated power and rated energy capacity of the energy storage system. Among them, the technical parameters refer to the rated power, rated energy capacity, charge and discharge rate, and response time.

[0066] The power demand, energy demand and capacity demand of the energy storage system are integrated into the demand data of the power system for the energy storage system.

[0067] In a preferred embodiment, step S4 is specifically,

[0068] According to the power demand, capacity demand and energy demand in the demand data, the charging and discharging power range and energy storage range of the energy storage system are obtained;

[0069] Based on the power system's net load curve and renewable energy generation forecast data, the energy storage system's charging and discharging periods are identified. The charging period is when the power system has excess power generation, during which the energy storage system absorbs excess energy. The discharging period is when the power system has insufficient power generation, during which the energy storage system releases stored energy.

[0070] Develop a charging and discharging time strategy based on the charging and discharging periods of the energy storage system; specifically, charging during peak renewable energy generation periods or low load periods; discharging during peak load periods or low renewable energy generation periods;

[0071] Determine the power allocation strategy based on the power and capacity requirements of the energy storage system. Specifically, during the discharge period, the output power of the energy storage system is dynamically allocated based on load demand and grid stability requirements; during the charging period, the input power of the energy storage system is dynamically allocated based on renewable energy generation and grid operating status.

[0072] The planning strategy of the energy storage system of the power grid is obtained according to the charging and discharging time strategy and the power allocation strategy; the planning strategy of the energy storage system of the power grid is optimized by combining the power grid topology data and the installation location of the energy storage system.

[0073] In a preferred embodiment, the influence of the installation location of the energy storage system on the power flow distribution of the power grid is determined based on the power grid topology, where the power grid topology includes node information, line information, transformer information, and power grid partition information;

[0074] Analyze the energy storage system's power support capabilities for key nodes and lines in the grid based on their installation location. These are areas of the grid where loads are concentrated, renewable energy access is concentrated, or stability is weak.

[0075] Optimize charging and discharging timing strategies based on the energy storage system's installation location and grid power distribution. Specifically, in areas with grid power congestion or poor voltage stability, prioritize discharging the energy storage system during peak load hours to alleviate line overload and voltage sag. In areas with concentrated renewable energy access, prioritize charging the energy storage system during peak power generation hours to absorb excess renewable energy generation.

[0076] The power allocation strategy is optimized based on the installation location of the energy storage system and the characteristics of the grid zoning. Specifically, in load center areas, the discharge power of the energy storage system is dynamically adjusted according to load demand to ensure the stability of load power supply; in areas where renewable energy is concentrated, the charging power of the energy storage system is dynamically adjusted according to power generation forecasts to ensure the maximum utilization of renewable energy. Among them, the grid zoning characteristics refer to voltage levels, load characteristics, and topological structures.

[0077] Grid flow calculation and stability analysis methods are used to verify the impact of the optimized charging and discharging times and power allocation strategies on grid operation, ensuring that the operation of the energy storage system is coordinated with the overall grid dispatch. Based on the verification results, the final optimization strategy is obtained, which includes: the charging and discharging schedules of the energy storage system at different installation locations, and the power allocation plan for the energy storage system in different areas.

Claims

1. An integrated method for intelligent grid planning and load forecasting, characterized in that: The steps include: S1: Collect load data of the power system; S2: Obtain historical load data, and use an LSTM neural network to predict load data based on the historical load data and the collected load data; S3: Obtaining renewable energy power generation, historical meteorological data, and meteorological data predicted by weather forecasts, and using random forest to predict renewable energy power generation based on the renewable energy power generation, historical meteorological data, and meteorological data predicted by weather forecasts; S4: Evaluate the power system's demand data for the energy storage system based on the predicted renewable energy generation and load data; the demand data includes power demand, capacity demand, and energy demand; S5: Generate a grid energy storage system planning strategy based on the demand data, including charging and discharging time and power allocation.

2. The integrated method for intelligent power grid planning and load forecasting according to claim 1, characterized in that: The step S2 is specifically as follows: Acquire historical load data and preprocess the historical load data, including data cleaning, data normalization, and data segmentation; Train the LSTM neural network model preset with the pre-processed historical load data; The load data collected in step S1 is input into the trained LSTM neural network model to predict the load data within a preset time period, including the load curve, load peak value and load valley value.

3. The integrated method of power grid intelligent planning and load forecasting according to claim 1, characterized in that: In the step S3, specifically, Clean renewable energy power generation, historical meteorological data, and forecasted meteorological data, including processing missing values, outliers, and noisy data; Normalizing or standardizing the historical meteorological data and the predicted meteorological data; Align renewable energy generation, historical meteorological data, and forecasted meteorological data in time series to form a unified dataset; Extract key features from historical and forecasted meteorological data, including wind direction, wind speed, and solar radiation; Train a random forest model based on key features of historical meteorological data and historical power generation; The preprocessed forecasted meteorological data is input into the random forest model to predict the renewable energy power generation in the preset time period.

4. The integrated method for intelligent power grid planning and load forecasting according to claim 1, characterized in that: The step S4 is specifically as follows: Calculating a net load curve of the power system based on the predicted renewable energy power and load data, wherein the net load curve is the difference between the load data and the renewable energy power generation; identifying a power imbalance period of the power system according to the net load curve, wherein the power imbalance period includes a period when the power generation of the power system is insufficient or excessive; Calculating the power demand and energy demand of the power system for the energy storage system according to the power imbalance period; Obtaining a capacity requirement of the energy storage system based on the operating characteristics of the power system, the technical parameters of the energy storage system, and a preset response time of the energy storage system, wherein the capacity requirement includes the rated power and rated energy capacity of the energy storage system; The power demand, energy demand and capacity demand of the energy storage system are integrated into the demand data of the power system for the energy storage system.

5. The integrated method of power grid intelligent planning and load forecasting according to claim 1, characterized in that: The step S4 is specifically as follows: Obtaining a charge and discharge power range and an energy storage range of the energy storage system based on the power demand, capacity demand, and energy demand in the demand data; Identify charging and discharging periods for energy storage systems based on the power system's net load curve and renewable energy generation forecast data; Formulate charging and discharging time strategies based on the charging and discharging periods of the energy storage system; Determine the power allocation strategy based on the power and capacity requirements of the energy storage system; Obtaining a grid energy storage system planning strategy based on the charging and discharging time strategy and the power allocation strategy; The grid energy storage system planning strategy is optimized by combining the grid topology data and the installation location of the energy storage system.

6. The integrated method of power grid intelligent planning and load forecasting according to claim 5, characterized in that: In step S4, the grid energy storage system planning strategy is optimized based on the grid topology data and the installation location of the energy storage system. Specifically, Determine the installation location of the energy storage system based on the grid topology, and analyze the grid power flow distribution based on the installation location of the energy storage system. The grid topology includes node information, line information, transformer information, and grid partition information. Analyze the energy storage system's power support capabilities for key nodes and lines in the grid, based on its installation location. These are areas of the grid with concentrated loads, concentrated renewable energy access, or weak stability. Optimize charging and discharging time by combining the installation location of the energy storage system and the distribution of power grid currents; Optimize power distribution based on the installation location of the energy storage system and the grid zoning characteristics.

Images