Capacity configuration optimization method and system for multi-energy power system

Abstract

The invention relates to the technical field of power systems, in particular to a capacity configuration optimization method and system for a multi-energy power system. The method comprises the following steps: acquiring historical load data, meteorological data and renewable energy output data of M regions in power grid operation, and constructing a load and meteorological association model and a meteorological and renewable energy output association model according to the acquired data; generating a future load scene according to results of the load and weather association model and the weather and renewable energy output association model; constructing a risk assessment model to carry out risk exposure assessment on the future load scene to obtain a risk score of the future load scene, and carrying out risk grade division; obtaining an initial redundancy capacity configuration scheme, constructing a multi-objective optimization model, establishing an optimization objective function in combination with the global redundancy capacity and the regional redundancy capacity, and iteratively updating the optimization objective function to obtain an optimal redundancy capacity configuration scheme; and according to the optimal redundancy capacity configuration scheme, dynamically configuring the redundancy capacity according to expert experience.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a capacity configuration optimization method and system for a multi-energy power system. Background Art

[0002] With the advancement of global energy transformation, the power system is gradually developing towards multi-energy grid connection, especially the widespread application of renewable energy (such as wind and solar energy). Although renewable energy helps to reduce dependence on traditional energy and achieve carbon emissions reduction, the instability and unpredictability of its power generation puts higher demands on the reliability and economy of the power system. The power grid needs to continuously adjust its redundant capacity configuration to cope with the impact of fluctuations in renewable energy generation. In addition, the spatiotemporal uncertainty of load also increases the complexity of the power system, especially in extreme weather conditions or sudden load conditions, the power grid must be able to respond flexibly.

[0003] Traditional redundant capacity configuration methods are usually based on historical data or assumptions, but due to the low accuracy of load forecasting, especially in sudden loads and extreme scenarios, there is often an excessive or insufficient redundant capacity, which in turn causes resource waste and grid security risks. Therefore, how to accurately assess the impact of load fluctuations and meteorological conditions on renewable energy output, and dynamically optimize redundant capacity in combination with multiple goals such as system security and economy, is a technical problem that needs to be solved urgently.

[0004] Therefore, a capacity configuration optimization method and system for a multi-energy power system are proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a capacity configuration optimization method and system for a multi-energy power system, which is suitable for dynamic configuration of redundant capacity of a multi-energy power system with sudden loads and extreme scenarios. The method includes: first, obtaining historical load data, meteorological data and renewable energy output data of M regions in power grid operation, and constructing a load-meteorological correlation model and a meteorological-renewable energy output correlation model based on the historical load data, meteorological data and renewable energy output data; secondly, generating future load scenarios based on the load-meteorological correlation model and the meteorological-renewable energy output correlation model; then, constructing a risk assessment model to conduct risk exposure assessment on future load scenarios, obtain risk scores for future load scenarios, and divide risk scores into risk levels according to preset classification rules; then, obtaining an initial redundant capacity configuration plan, constructing a multi-objective optimization model to establish an optimization objective function in combination with global redundant capacity and regional redundant capacity, and obtaining the optimal redundant capacity configuration plan by iteratively updating the optimization objective function; finally, according to the multi-objective optimization results, dynamically configuring redundant capacity in combination with risk levels.

[0006] To achieve the above object, the present invention provides the following technical solutions: A capacity configuration optimization method for a multi-energy power system, comprising: Acquire historical load data, meteorological data and renewable energy output data of M regions in power grid operation, and construct a load-meteorological correlation model and a meteorological-renewable energy output correlation model based on the historical load data, the meteorological data and the renewable energy output data; Generate a future load scenario according to the load-meteorological correlation model and the meteorological-renewable energy output correlation model; Constructing a risk assessment model to conduct risk exposure assessment on the future load scenario, obtaining a risk score for the future load scenario, and classifying the risk score into risk levels according to a preset classification rule; The initial redundant capacity configuration scheme is obtained, and a multi-objective optimization model is constructed to minimize risk exposure and redundant capacity configuration cost as the optimization goal. The optimization objective function is established by combining the global redundant capacity cost and the regional redundant capacity cost. The optimal redundant capacity configuration scheme is obtained by iteratively updating the optimization objective function; the optimization objective function is: ; in, To optimize the objective function; Dynamic weighting for risk scoring; is the risk scoring function; is the dynamic weight of global redundant capacity cost; is the global redundant capacity cost function; is the dynamic weight of power system reliability; Score system reliability; For the The regional redundant capacity cost function of each region; is the total number of regions; For the The regional redundant capacity cost weight of each region; According to the optimal redundant capacity configuration scheme, redundant capacity is dynamically configured based on expert experience.

[0007] Preferably, the historical load data includes: load demand, load curve and peak-to-valley value changes; The meteorological data include: temperature, humidity, wind speed, wind direction and sunshine duration; The renewable energy output data include: wind turbine power output and photovoltaic power station power generation.

[0008] Preferably, the load and weather correlation model includes: a load and weather correlation analysis layer, a load and weather time series feature extraction layer and a load and weather change prediction layer; The load and meteorological correlation analysis layer evaluates the correlation between different meteorological conditions in the meteorological data and the historical load data through a correlation analysis method, and selects highly correlated meteorological data; The load and meteorological time series feature extraction layer extracts load and meteorological time series features based on the highly correlated meteorological data; The load change with meteorology prediction layer predicts the time trend of load change through a time series model according to the load and meteorological time series characteristics, and obtains the prediction result of load change with meteorology.

[0009] Preferably, the meteorological and renewable energy output correlation model comprises: a meteorological and renewable energy correlation feature extraction layer, a wind speed and wind power generation output relationship layer, and a light intensity and photovoltaic power generation output relationship layer; The meteorological and renewable energy related feature extraction layer extracts features related to renewable energy output in the meteorological data based on the meteorological data, including: meteorological and wind power generation related features and meteorological and photovoltaic power generation related features; The wind speed and wind power generation output relationship layer is based on the meteorological and wind power generation correlation characteristics, combined with the wind turbine power curve, to fit the mapping relationship between wind speed and wind power generation output; The light intensity and photovoltaic power generation output relationship layer is based on the meteorological and photovoltaic power generation correlation characteristics, combined with the photoelectric conversion of photovoltaic components to fit the mapping relationship between light intensity and photovoltaic power generation output.

[0010] Preferably, the step of generating future load scenarios based on the results obtained from the load-meteorological correlation model and the meteorological-renewable energy output correlation model includes: combining the load-meteorological correlation model and the meteorological-renewable energy output correlation model to generate future load scenarios through a scenario generation algorithm; the future load scenarios include: future conventional load scenarios, future sudden load scenarios and future extreme load scenarios.

[0011] Preferably, a risk assessment model is constructed to perform risk exposure assessment on the future load scenario to obtain a risk score for the future load scenario, and the risk score is classified into risk levels according to a preset risk score threshold. The specific steps include: Based on the future load scenario, calculate the system load excess probability and renewable energy output insufficient probability under the conventional load scenario, the sudden load scenario and the extreme load scenario; Constructing the risk assessment model, combining the system load excess probability and the renewable energy output under-probability, quantifying the risk exposure under the future load scenario, and obtaining the risk score; The risk score is divided into risk levels according to a preset risk score threshold; the risk levels include: low risk, medium risk and high risk.

[0012] Preferably, the calculation method of the regional redundant capacity cost function is: based on the historical load data, the meteorological data and the renewable energy output data of the M regions, the redundant capacity configuration amount and unit cost of each region are calculated to obtain the regional redundant capacity cost.

[0013] A capacity configuration optimization system for a multi-energy power system, comprising: A data acquisition module, used to acquire historical load data, meteorological data and renewable energy output data of M regions in power grid operation, and to construct a load-meteorological correlation model and a meteorological-renewable energy output correlation model based on the historical load data, the meteorological data and the renewable energy output data; A future load scenario generation module, used to generate future load scenarios according to the load and weather correlation model and the weather and renewable energy output correlation model; A load risk assessment module is used to construct a load risk assessment model to conduct risk exposure assessment on the future load scenario, obtain a risk score for the future load scenario, and classify the risk score into risk levels according to a preset classification rule; The multi-objective optimization module is used to obtain the initial redundant capacity configuration plan, build a multi-objective optimization model to minimize risk exposure and redundant capacity configuration cost as the optimization goal, establish an optimization objective function in combination with the global redundant capacity cost and the regional redundant capacity cost, and obtain the optimal redundant capacity configuration plan by iteratively updating the optimization objective function; the optimization objective function is: ; in, To optimize the objective function; Dynamic weighting for risk scoring; is the risk scoring function; is the dynamic weight of global redundant capacity cost; is the global redundant capacity cost function; is the dynamic weight of power system reliability; Score system reliability; For the The regional redundant capacity cost function of each region; is the total number of regions; For the The regional redundant capacity cost weight of each region; The redundant capacity configuration module is used to dynamically configure the redundant capacity according to the optimal redundant capacity configuration scheme and based on expert experience.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention constructs a load-meteorological correlation model and a meteorological-renewable energy output correlation model, comprehensively analyzes the relationship between historical load data, meteorological data, and renewable energy output data, effectively captures the uncertainty of load fluctuations and renewable energy power generation, and provides accurate data support for generating conventional load scenarios, sudden load scenarios, and extreme load scenarios. This method effectively improves the accuracy of load forecasting and significantly reduces the risk of redundant capacity configuration deviation due to inaccurate forecasting, thereby enhancing the reliability of future load scenario generation.

[0015] 2. The present invention constructs a load-meteorological correlation model and a meteorological-renewable energy output correlation model to accurately simulate the impact of meteorological conditions on grid load and renewable energy output, generate conventional, sudden and extreme load scenarios, and fully reflect the various operating conditions that the grid may face; then, based on the risk assessment model, the risk exposure assessment and classification of future load scenarios are carried out, which can effectively quantify the risk level of each scenario and ensure that the capacity configuration plan is targeted in high-risk scenarios. This method improves the system's ability to cope with sudden loads and extreme loads while avoiding waste of resources and achieving a balanced optimization of risks and costs.

[0016] 3. The present invention proposes a multi-objective optimization model that aims to minimize risk exposure and redundant capacity configuration costs. By combining global redundant capacity costs and regional redundant capacity costs, the optimization objective function can flexibly respond to load uncertainty and the instability of renewable energy generation, achieve efficient coordination of global and local capacity configuration, and significantly improve the economy and reliability of the system. At the same time, by dynamically adjusting risk weights and configuration schemes, the system can quickly respond to different load scenarios, effectively reduce resource waste and system failure risks, and ensure the safe and stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a capacity configuration optimization method for a multi-energy power system provided by an embodiment of the present invention; Figure 2 A structural diagram of a capacity configuration optimization system for a multi-energy power system provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of generating future load scenarios provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] With the advancement of global energy transformation, the power system is gradually developing towards multi-energy grid connection, especially the widespread application of renewable energy (such as wind and solar energy). Although renewable energy helps to reduce dependence on traditional energy and achieve carbon emissions reduction, the instability and unpredictability of its power generation puts higher demands on the reliability and economy of the power system. The power grid needs to continuously adjust its redundant capacity configuration to cope with the impact of fluctuations in renewable energy generation. In addition, the spatiotemporal uncertainty of load also increases the complexity of the power system, especially in extreme weather conditions or sudden load conditions, the power grid must be able to respond flexibly.

[0020] The present invention proposes a capacity configuration optimization method for a multi-energy power system to achieve dynamic optimization of redundant capacity configuration. This method is applied to a capacity configuration optimization system for a multi-energy power system. For specific method flow chart and system structure diagram, please refer to Figure 1 and Figure 2 In order to illustrate that the method and system of the present invention can play a role in dynamically optimizing redundant capacity configuration, the effectiveness of the present invention will be described from two embodiments below.

[0021] Embodiment 1 In the embodiment of the present application, the method and system proposed by the present invention are used to describe in detail the dynamic optimization process of redundant capacity configuration. In the embodiment of the present application, the dynamic optimization of redundant capacity configuration is aimed at the redundant capacity configuration optimization of the power grid A in a certain area. Figure 1 and Figure 2 The content describes in detail the redundant capacity configuration optimization process of the power system; among them, Figure 1The specific process of the method proposed in the present invention includes: obtaining historical load data, meteorological data and renewable energy output data of M areas in power grid operation; constructing a load-meteorological correlation model and a meteorological-renewable energy output correlation model according to the historical load data, the meteorological data and the renewable energy output data; generating future load scenarios according to the load-meteorological correlation model and the meteorological-renewable energy output correlation model; constructing a risk assessment model to conduct risk exposure assessment on the future load scenarios, obtain risk scores of the future load scenarios, and classify the risk scores into risk levels according to preset classification rules; obtaining an initial redundant capacity configuration plan, constructing a multi-objective optimization model to minimize risk exposure and redundant capacity configuration cost as optimization goals, establishing an optimization objective function in combination with global redundant capacity cost and regional redundant capacity cost, and obtaining an optimal redundant capacity configuration plan by iteratively updating the optimization objective function; and dynamically configuring redundant capacity according to expert experience according to the optimal redundant capacity configuration plan. Figure 2 The structure diagram of the system proposed in the present invention includes: a data acquisition module, a future load scenario generation module, a load risk assessment module, a multi-objective optimization module and a redundant capacity configuration module; combined with Figure 1 and Figure 2 The following is a description of the contents: Obtain historical load data, meteorological data and renewable energy output data of M areas in power grid operation; the historical load data include: load demand, load curve and peak-to-valley value changes; the meteorological data include: temperature, humidity, wind speed, wind direction and sunshine duration; the renewable energy output data include: wind turbine power output and photovoltaic power station power generation.

[0022] Specifically, historical load data, meteorological data and renewable energy output data of the three areas of the region A power grid are obtained; the historical load data include: the hourly load demand, load curve and peak-to-valley value changes in the three areas from 2019 to 2021; the meteorological data include: the temperature, humidity, wind speed, wind direction and sunshine duration in the three areas from 2019 to 2021; the renewable energy output data include: the power output of wind turbines and the power generation of photovoltaic power stations in the three areas from 2019 to 2021.

[0023] By acquiring historical load data, meteorological data, and renewable energy output data from multiple regions, reliable basic support can be provided for the optimized dispatching, load forecasting, redundant capacity configuration, and system reliability assessment of the power grid. These data help analyze the load change trend, the impact of meteorological conditions on energy output, and the contribution of renewable energy, thereby improving the operational efficiency, stability, and security of the power grid. After analyzing these data, accurate risk assessment, optimized decision-making, and power grid dispatching can be achieved, thereby improving the power grid's ability to respond and reliability in the face of extreme weather and emergencies.

[0024] Preferably, the load and weather correlation model includes: a load and weather correlation analysis layer, a load and weather time series feature extraction layer and a load and weather change prediction layer; The load and meteorological correlation analysis layer evaluates the correlation between different meteorological conditions in the meteorological data and the historical load data through a correlation analysis method, and selects highly correlated meteorological data; The load and meteorological time series feature extraction layer extracts load and meteorological time series features based on the highly correlated meteorological data; The load change with meteorology prediction layer predicts the time trend of load change through a time series model according to the load and meteorological time series characteristics, and obtains the prediction result of load change with meteorology.

[0025] Specifically, the load and meteorological correlation analysis layer evaluates the correlation by calculating the Pearson correlation coefficient between different meteorological conditions and historical load data; sets a correlation threshold, selects meteorological data above the correlation threshold, and obtains high-correlation meteorological data; The load and meteorological time series feature extraction layer uses a sliding window method to extract the time series features between the load and the highly correlated meteorological data; The load change prediction layer predicts the trend of load change with weather through LSTM network to obtain the future load demand prediction result.

[0026] The embodiment of the present application can effectively reveal the impact of meteorological conditions on power grid load demand by constructing a load and meteorological correlation model, combining correlation analysis, time series feature extraction and LSTM network prediction. First, the correlation analysis layer evaluates the correlation between meteorological data and historical load data through the Pearson correlation coefficient, and selects highly correlated meteorological factors; then, the time series feature extraction layer uses the sliding window method to extract the time series features between load and meteorological data; finally, the load change with meteorological prediction layer predicts future load change trends through the LSTM network, thereby achieving accurate prediction of power grid load demand, improving the intelligent level of power grid dispatching and management, and enhancing the ability to cope with meteorological fluctuations.

[0027] Preferably, the meteorological and renewable energy output correlation model comprises: a meteorological and renewable energy correlation feature extraction layer, a wind speed and wind power generation output relationship layer, and a light intensity and photovoltaic power generation output relationship layer; The meteorological and renewable energy related feature extraction layer extracts features related to renewable energy output in the meteorological data based on the meteorological data, including: meteorological and wind power generation related features and meteorological and photovoltaic power generation related features; The wind speed and wind power generation output relationship layer is based on the meteorological and wind power generation correlation characteristics, combined with the wind turbine power curve, to fit the mapping relationship between wind speed and wind power generation output; The light intensity and photovoltaic power generation output relationship layer is based on the meteorological and photovoltaic power generation correlation characteristics, combined with the photoelectric conversion fitting of the photovoltaic components.

[0028] Specifically, the wind speed and wind power generation output relationship layer fits the relationship between wind speed and wind power generation output in different regions through a polynomial fitting model; The light intensity and photovoltaic power generation output relationship layer uses a linear regression model to fit the relationship between light intensity and photovoltaic power generation power in different areas.

[0029] The embodiment of the present application effectively captures the impact of meteorological factors on renewable energy power generation output by constructing a meteorological and renewable energy output correlation model. First, the meteorological and renewable energy correlation feature extraction layer extracts features related to wind power generation and photovoltaic power generation from meteorological data; then, the wind speed and wind power generation output relationship layer fits the relationship between wind speed and wind power generation power through a polynomial, and the light intensity and photovoltaic power generation output relationship layer fits the relationship between light intensity and photovoltaic power generation power through a linear regression model. This method can accurately simulate the impact of meteorological conditions on renewable energy output and provide a scientific basis for optimizing the dispatch and operation of power systems.

[0030] Preferably, the future load scenario is generated according to the load-meteorological correlation model and the meteorological-renewable energy output correlation model; for specific steps, please refer to Figure 3 The steps include: combining the load and meteorological correlation model and the meteorological and renewable energy output correlation model, generating future load scenarios through a scenario generation algorithm; the future load scenarios include: future conventional load scenarios, future sudden load scenarios and future extreme load scenarios; Specifically, the future conventional load scenario is generated based on the prediction results of historical load data and meteorological data; the future sudden load scenario is simulated according to the load fluctuation of the load curve in the historical load data and the sudden load scenario in combination with meteorological emergencies; the future extreme load scenario is generated by considering extreme meteorological events and insufficient output of renewable energy; The conventional load scenario, the sudden load scenario and the extreme load scenario are converted into a data model of a future load scenario through a Monte Carlo simulation algorithm.

[0031] The embodiment of the present application combines the load-meteorological correlation model and the meteorological-renewable energy output correlation model, and uses a scenario generation algorithm to generate future load scenarios, covering conventional load scenarios, sudden load scenarios, and extreme load scenarios. Conventional load scenarios are based on the prediction results of historical load and meteorological data, sudden load scenarios take into account load fluctuations and meteorological emergencies, and extreme load scenarios combine extreme meteorological events and insufficient renewable energy output. Through the Monte Carlo simulation algorithm, these load scenarios are converted into data models, which can provide reliable prediction basis for the optimal scheduling, risk assessment and capacity configuration of the power system, and improve the emergency handling capability and reliability of the system. See Table 1.

[0032] Table 1 Prediction accuracy of future load scenarios

[0033] Preferably, constructing a risk assessment model to conduct risk exposure assessment on the future load scenario, obtaining a risk score for the future load scenario, and classifying the risk score into risk levels according to a preset classification rule, the steps include: Based on the future load scenarios, calculate the system load excess probability and renewable energy output shortage probability under future conventional load scenarios, future sudden load scenarios, and future extreme load scenarios; Constructing the risk assessment model, combining the system load excess probability and the renewable energy output under-probability, quantifying the risk exposure under the future load scenario, and obtaining the risk score; The risk score is divided into risk levels according to a preset risk score threshold; the risk levels include: low risk, medium risk and high risk.

[0034] Specifically, the calculation formula of the risk score is: ; in, is the risk scoring function; is the total number of future load scenarios; For the The combined losses of future load scenarios; is the weight of the system load excess risk; is an indicator function, which takes the value of 1 when the probability exceeds the threshold, otherwise it is 0; For the The probability of system load excess under the next future load scenario; is the risk threshold of system load excess probability; The weight of the risk of insufficient output of renewable energy; For the The probability of insufficient output of renewable energy under the future load scenario; The risk threshold for the probability of insufficient output of renewable energy; The risk score thresholds are set as follows: low risk is 0-0.3, medium risk is 0.3-0.6, and high risk is 0.6-1.0; The risk score is classified into risk levels according to the risk score threshold.

[0035] The embodiment of the present application constructs a risk assessment model to conduct risk exposure assessment on the future load scenarios, calculates the probability of system load overload and renewable energy output underload under conventional, sudden and extreme load scenarios, and quantifies risk exposure based on these probabilities. The risk score obtained is divided into low, medium and high risk levels according to the preset risk score threshold. This method provides an accurate risk assessment basis for the capacity configuration and optimization of the power system, helps the system improve its response capability and stability when facing different load scenarios, and ensures the safety and reliability of the power system operation.

[0036] Preferably, an initial redundant capacity configuration scheme is obtained, a multi-objective optimization model is constructed to minimize risk exposure and redundant capacity configuration cost as optimization objectives, an optimization objective function is established in combination with the global redundant capacity cost and the regional redundant capacity cost, and the optimal redundant capacity configuration scheme is obtained by iteratively updating the optimization objective function; the optimization objective function is: ; in, To optimize the objective function; Dynamic weighting for risk scoring; is the risk scoring function; is the dynamic weight of global redundant capacity cost; is the global redundant capacity cost function; is the dynamic weight of power system reliability; Score system reliability; For the The regional redundant capacity cost function of each region; is the total number of regions; For the The regional redundant capacity cost weight of each region; The calculation method of the regional redundant capacity cost function is: according to the historical load data, the meteorological data and the renewable energy output data of the M regions, the redundant capacity configuration amount and the unit cost of each region are calculated to obtain the regional redundant capacity cost function.

[0037] The embodiment of the present application constructs a multi-objective optimization model with the goal of minimizing risk exposure and redundant capacity configuration costs, and optimizes the redundant capacity configuration of the power system by combining the global redundant capacity cost and the regional redundant capacity cost. By establishing an optimization objective function, factors such as risk score, redundant capacity cost and system reliability are comprehensively considered, and the weights are dynamically adjusted to obtain a multi-objective optimization result. In addition, the calculation method of the regional redundant capacity cost takes into account the historical load, meteorological and renewable energy data of different regions to ensure that the model can achieve accurate cost evaluation and optimization according to the characteristics of the specific region. This method can effectively improve the resource allocation efficiency and economy of the power system in the face of different risk situations, refer to Table 2.

[0038] Table 2 Comparison of redundant capacity configuration results

[0039] According to the optimal redundant capacity configuration scheme, redundant capacity is dynamically configured based on expert experience.

[0040] The embodiments of the present application acquire historical load data, meteorological data and renewable energy output data of multiple regions in the operation of the power grid, construct load and meteorological correlation models and meteorological and renewable energy output correlation models, and can accurately predict future conventional, sudden and extreme load scenarios; conduct risk exposure assessment and scoring for different load scenarios based on the risk assessment model, quantify the risk level through grading rules, and provide comprehensive risk analysis; combine global and regional redundant capacity costs to build a multi-objective optimization model, with the goal of minimizing risk exposure and optimizing redundant capacity configuration costs, comprehensively consider system reliability and resource allocation, and generate the optimal configuration plan; finally, dynamically adjust the redundant capacity configuration through expert experience, improve the efficiency and economy of power grid operation, and enhance the reliability and safety of coping with complex load scenarios.

[0041] Embodiment 2 In Example 1, the method proposed in the present invention successfully realizes the dynamic optimization of the redundant capacity configuration of the power grid in region A. To further verify the effectiveness of the present invention, the present application embodiment also proposes a capacity configuration optimization system for a multi-energy power system to dynamically optimize the redundant capacity configuration of the power grid in region B; the system includes: a data acquisition module, a future load scenario generation module, a load risk assessment module, a multi-objective optimization module and a redundant capacity configuration module.

[0042] A data acquisition module, used to acquire historical load data, meteorological data and renewable energy output data of M regions in power grid operation, and to construct a load-meteorological correlation model and a meteorological-renewable energy output correlation model based on the historical load data, the meteorological data and the renewable energy output data; The historical load data includes: load demand, load curve and peak-to-valley value changes; The meteorological data include: temperature, humidity, wind speed, wind direction and sunshine duration; The renewable energy output data include: wind turbine power output and photovoltaic power station power generation.

[0043] Preferably, the load and weather correlation model includes: a load and weather correlation analysis layer, a load and weather time series feature extraction layer and a load and weather change prediction layer; The load and meteorological correlation analysis layer evaluates the correlation between different meteorological conditions in the meteorological data and the historical load data through a correlation analysis method, and selects highly correlated meteorological data; The load and meteorological time series feature extraction layer extracts load and meteorological time series features based on the highly correlated meteorological data; The load change with meteorology prediction layer predicts the time trend of load change through a time series model according to the load and meteorological time series characteristics, and obtains the prediction result of load change with meteorology.

[0044] Preferably, the meteorological and renewable energy output correlation model comprises: a meteorological and renewable energy correlation feature extraction layer, a wind speed and wind power generation output relationship layer, and a light intensity and photovoltaic power generation output relationship layer; The meteorological and renewable energy related feature extraction layer extracts features related to the meteorological data and the output of renewable energy based on the meteorological data, including: meteorological and wind power generation related features and meteorological and photovoltaic power generation related features; The wind speed and wind power generation output relationship layer is based on the meteorological and wind power generation correlation characteristics, combined with the wind turbine power curve, to fit the mapping relationship between wind speed and wind power generation output; The light intensity and photovoltaic power generation output relationship layer is based on the meteorological and photovoltaic power generation correlation characteristics, combined with the photoelectric conversion of photovoltaic components to fit the mapping relationship between light intensity and photovoltaic power generation output.

[0045] Preferably, the future load scenario generation module is used to generate future load scenarios according to the load and weather correlation model and the weather and renewable energy output correlation model, and the steps include: Combining the load and weather correlation model and the weather and renewable energy output correlation model, a future load scenario is generated through a scenario generation algorithm; the future load scenario includes: future conventional load scenario, future sudden load scenario and future extreme load scenario.

[0046] Preferably, the load risk assessment module is used to construct a load risk assessment model to perform risk exposure assessment on the future load scenario, obtain a risk score for the future load scenario, and classify the risk score into risk levels according to a preset classification rule, and the steps include: Based on the future load scenarios, calculate the system load excess probability and renewable energy output shortage probability under future conventional load scenarios, future sudden load scenarios, and future extreme load scenarios; Constructing the risk assessment model, combining the system load excess probability and the renewable energy output under-probability, quantifying the risk exposure under the future load scenario, and obtaining the risk score; The risk score is divided into risk levels according to a preset risk score threshold; the risk levels include: low risk, medium risk and high risk.

[0047] Preferably, the multi-objective optimization module is used to obtain an initial redundant capacity configuration scheme, construct a multi-objective optimization model to minimize risk exposure and redundant capacity configuration cost as optimization objectives, establish an optimization objective function in combination with the global redundant capacity cost and the regional redundant capacity cost, and obtain the optimal redundant capacity configuration scheme by iteratively updating the optimization objective function; the optimization objective function is: ; in, To optimize the objective function; Dynamic weighting for risk scoring; is the risk scoring function; is the dynamic weight of global redundant capacity cost; is the global redundant capacity cost function; is the dynamic weight of power system reliability; Score system reliability; For the The regional redundant capacity cost function of each region; is the total number of regions; For the The regional redundant capacity cost weight of each region; The calculation method of the regional redundant capacity cost function is: according to the historical load data, the meteorological data and the renewable energy output data of the M regions, the redundant capacity configuration amount and the unit cost of each region are calculated to obtain the regional redundant capacity cost function.

[0048] Table 3 shows the comparison of the redundant capacity configuration results of the system proposed by the present invention and the traditional method.

[0049] Table 3 Comparison of redundant capacity configuration results

[0050] Preferably, the redundant capacity configuration module is used to dynamically configure the redundant capacity according to the optimal redundant capacity configuration scheme based on expert experience.

[0051] 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 capacity configuration optimization method for a multi-energy power system, characterized in that: include: Acquire historical load data, meteorological data and renewable energy output data of M regions in power grid operation, and construct a load-meteorological correlation model and a meteorological-renewable energy output correlation model based on the historical load data, the meteorological data and the renewable energy output data; Generate a future load scenario based on the results obtained from the load-meteorological correlation model and the meteorological-renewable energy output correlation model; Constructing a risk assessment model to conduct risk exposure assessment on the future load scenario, obtaining a risk score for the future load scenario, and classifying the risk score into risk levels according to a preset classification rule; The initial redundant capacity configuration scheme is obtained, and a multi-objective optimization model is constructed to minimize risk exposure and redundant capacity configuration cost as the optimization goal. The optimization objective function is established by combining the global redundant capacity cost and the regional redundant capacity cost. The optimal redundant capacity configuration scheme is obtained by iteratively updating the optimization objective function; the optimization objective function is: ； in, To optimize the objective function; Dynamic weighting for risk scoring; is the risk scoring function; is the dynamic weight of global redundant capacity cost; is the global redundant capacity cost function; is the dynamic weight of power system reliability; Score system reliability; For the The regional redundant capacity cost function of each region; is the total number of regions; For the The regional redundant capacity cost weight of each region; According to the optimal redundant capacity configuration scheme, redundant capacity is dynamically configured based on expert experience.

2. A capacity configuration optimization method for a multi-energy power system according to claim 1, characterized in that: The historical load data includes: load demand, load curve and peak-to-valley value changes; The meteorological data include: temperature, humidity, wind speed, wind direction and sunshine duration; The renewable energy output data include: wind turbine power output and photovoltaic power station power generation.

3. The capacity configuration optimization method of a multi-energy power system according to claim 1 is characterized in that: The load and weather correlation model includes: a load and weather correlation analysis layer, a load and weather time series feature extraction layer and a load and weather change prediction layer; The load and meteorological correlation analysis layer evaluates the correlation between different meteorological conditions in the meteorological data and the historical load data through a correlation analysis method, and selects highly correlated meteorological data; The load and meteorological time series feature extraction layer extracts load and meteorological time series features based on the highly correlated meteorological data; The load change with meteorology prediction layer predicts the time trend of load change through a time series model according to the load and meteorological time series characteristics, and obtains the prediction result of load change with meteorology.

4. The capacity configuration optimization method of a multi-energy power system according to claim 1 is characterized in that: The meteorological and renewable energy output correlation model includes: a meteorological and renewable energy correlation feature extraction layer, a wind speed and wind power generation output relationship layer, and a light intensity and photovoltaic power generation output relationship layer; The meteorological and renewable energy related feature extraction layer extracts features related to renewable energy output in the meteorological data based on the meteorological data, including: meteorological and wind power generation related features and meteorological and photovoltaic power generation related features; The wind speed and wind power generation output relationship layer is based on the meteorological and wind power generation correlation characteristics, combined with the wind turbine power curve, to fit the mapping relationship between wind speed and wind power generation output; The light intensity and photovoltaic power generation output relationship layer is based on the meteorological and photovoltaic power generation correlation characteristics, combined with the photoelectric conversion of the photovoltaic components to fit the mapping relationship between light intensity and photovoltaic power generation output.

5. The capacity configuration optimization method of a multi-energy power system according to claim 1 is characterized in that: The step of generating future load scenarios based on the results obtained from the load-meteorological correlation model and the meteorological-renewable energy output correlation model includes: combining the load-meteorological correlation model and the meteorological-renewable energy output correlation model to generate future load scenarios through a scenario generation algorithm; the future load scenarios include: future conventional load scenarios, future sudden load scenarios and future extreme load scenarios.

6. The capacity configuration optimization method of a multi-energy power system according to claim 1 is characterized in that: Constructing a risk assessment model to conduct risk exposure assessment on the future load scenario, obtaining a risk score for the future load scenario, and classifying the risk score into risk levels according to a preset risk score threshold, the specific steps include: Based on the future load scenarios, calculate the system load excess probability and renewable energy output shortage probability under future conventional load scenarios, future sudden load scenarios, and future extreme load scenarios; Constructing the risk assessment model, combining the system load excess probability and the renewable energy output under-probability, quantifying the risk exposure under the future load scenario, and obtaining the risk score; The risk score is divided into risk levels according to a preset risk score threshold; the risk levels include: low risk, medium risk and high risk.

7. The capacity configuration optimization method of a multi-energy power system according to claim 1, characterized in that: The calculation method of the regional redundant capacity cost function is: according to the historical load data, the meteorological data and the renewable energy output data of the M regions, the redundant capacity configuration amount and the unit cost of each region are calculated to obtain the regional redundant capacity cost function.

8. A capacity configuration optimization system for a multi-energy power system, characterized in that: include: A data acquisition module, used to acquire historical load data, meteorological data and renewable energy output data of M regions in power grid operation, and to construct a load-meteorological correlation model and a meteorological-renewable energy output correlation model based on the historical load data, the meteorological data and the renewable energy output data; A future load scenario generation module, used to generate future load scenarios according to the load and weather correlation model and the weather and renewable energy output correlation model; A load risk assessment module is used to construct a load risk assessment model to conduct risk exposure assessment on the future load scenario, obtain a risk score for the future load scenario, and classify the risk score into risk levels according to a preset classification rule; The multi-objective optimization module is used to obtain the initial redundant capacity configuration plan, build a multi-objective optimization model to minimize risk exposure and redundant capacity configuration cost as the optimization goal, establish an optimization objective function in combination with the global redundant capacity cost and the regional redundant capacity cost, and obtain the optimal redundant capacity configuration plan by iteratively updating the optimization objective function; the optimization objective function is: ； in, To optimize the objective function; Dynamic weighting for risk scoring; is the risk scoring function; is the dynamic weight of global redundant capacity cost; is the global redundant capacity cost function; is the dynamic weight of power system reliability; Score system reliability; For the The regional redundant capacity cost function of each region; is the total number of regions; For the The regional redundant capacity cost weight of each region; The redundant capacity configuration module is used to dynamically configure the redundant capacity according to the optimal redundant capacity configuration scheme and based on expert experience.

Images