Multi-target-based wind and light energy storage power station capacity prediction system

Abstract

The invention discloses a wind and light energy storage power station capacity prediction system based on multiple targets, and relates to the technical field of new energy power generation. The multi-target data acquisition module determines a target area according to power grid topology, load and wind and light resource distribution, equipment is installed to acquire data, a reliable data basis is provided for subsequent modules, data is ensured to be accurate and effective, error and abnormal data interference is avoided, and the balance management module monitors a power grid in real time. Analyzing the wind-solar power generation and electrical load balance relationship, and adjusting the operation mode of the energy storage power station according to the relationship; the power prediction module constructs and optimizes a model, accurately predicts power, calculates energy storage charging and discharging power, ensures stable operation of a power grid and improves energy utilization efficiency, and the scheduling decision module formulates a scheduling scheme in combination with multi-aspect information to ensure safe, stable and economic operation of the power grid; the risk assessment module comprehensively considers various risk factors, assesses the operation performance of each power station, matches a coping strategy and an early warning mechanism, and enhances the operation reliability and safety of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy power generation, and particularly relates to a capacity prediction system for a wind-solar energy storage power station based on multiple objectives. Background Art

[0002] Wind power generation and photovoltaic power generation are intermittent and volatile, which bring great impacts to the stable operation of the power grid. The traditional thermal reserve mode of thermal power units is not applicable to wind power and photovoltaic power, so it is necessary to configure energy storage installations as thermal reserve capacity to participate in power grid peak shaving. For example, a smart optimal scheduling method for grid-connected wind-solar energy storage power generation disclosed in the patent with the publication number CN104283236B includes: 1) collecting historical wind power and photovoltaic power data to obtain the prediction error distributions of wind power and photovoltaic power output at different time scales of day-ahead and intra-day; 2) using the existing wind power output prediction system and photovoltaic power output prediction system to respectively obtain the predicted wind power and photovoltaic power output values according to meteorological data (light, temperature); 3) determining the wind power output scenario value and photovoltaic power output scenario value by combining steps 1) and 2); 4) establishing a smart scheduling model for grid-connected wind-solar energy storage power generation, and the smart scheduling model for grid-connected wind-solar energy storage power generation includes a scheduling optimization objective function model and a scheduling constraint condition model; the scheduling optimization objective function model includes day-ahead, intra-day, and real-time scheduling models; 5) optimizing to obtain the first day-ahead combined scheduling value of wind-solar energy storage power generation; 6) obtaining the intra-day combined scheduling value of wind-solar energy storage power generation and the intra-day scheduling values of wind, light, and energy storage power generation; 7) obtaining the real-time scheduling value of wind-solar energy storage.

[0003] Although the above patent schedules and arranges wind-solar energy storage power generation according to different time scales and meteorological data to cope with the uncertainty of wind-solar power generation, there are still the following problems:

[0004] 1. In the prior art, the predicted wind power and photovoltaic power output values are mainly obtained based on historical data and existing prediction systems. The prediction accuracy of the power generation power sequences of wind power and photovoltaic power generation with high intermittency and volatility is limited, and it is difficult to accurately capture the real-time changes of wind-solar power generation, resulting in a large deviation between the subsequent scheduling arrangements and the actual situation, and it is impossible to effectively ensure the stable operation of the power grid.

[0005] 2. In the actual operation of a wind-solar energy storage power station, it is necessary to comprehensively consider the optimization of objectives such as maximizing economic benefits, minimizing energy storage costs, and minimizing the impact on the power grid. The optimization of a single objective is difficult to balance the requirements of all aspects, and it may cause adverse effects in other aspects while achieving good results in some aspects. Summary of the Invention

[0006] The object of the present invention is to provide a capacity prediction system for a wind-solar energy storage power station based on multiple objectives. By collecting accurate data through multi-objective data acquisition, and through balance management and power prediction, the stability of the power grid is ensured and the energy utilization efficiency is improved. Finally, through scientific scheduling and risk prevention and control, the safe and economic operation of the power grid is ensured, and the effectiveness and reliability of the capacity prediction system for the wind-solar energy storage power station are comprehensively improved, so as to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A capacity prediction system for a wind-solar energy storage power station based on multiple objectives, comprising:

[0009] A multi-objective data acquisition module, configured to determine the target area of the wind-solar energy storage power station, collect the power station operation data of the target area power grid, and preprocess the collected power station operation data;

[0010] A balance management module, configured to obtain the preprocessed power station operation data, monitor the target area power grid in real time, analyze the balance relationship between wind-solar power generation and electricity load in the target area, judge whether there is a power imbalance situation, and adjust the operation mode of the energy storage power station in the target area power grid according to the analysis result;

[0011] A power prediction module, configured to construct a power prediction model based on the obtained power station operation data, predict the photovoltaic, wind power and load power, optimize the power prediction model, and calculate the charge and discharge power of the energy storage at each moment according to the prediction result and the multi-objective optimization algorithm.

[0012] Further, the multi-objective data acquisition module specifically includes:

[0013] A target area determination unit, configured to determine the target area that needs to be capacity predicted and managed according to the topological structure, load distribution and wind-solar resource distribution of the power grid, construct a target area power grid based on the wind power station, photovoltaic power station, energy storage power station and substation in the target area, and install data acquisition devices at each key node in the target area power grid;

[0014] A data acquisition unit, configured to collect the power station operation data of the target area power grid in real time based on the data acquisition device, screen and preprocess the obtained power station operation data, identify and remove obviously incorrect or abnormal data, and transmit the preprocessed power station operation data to the balance management module through a wireless communication network.

[0015] Further, the multi-objective data acquisition module specifically further includes:

[0016] Extract the operating load corresponding to each unit operation monitoring period of the wind power station, photovoltaic power station, energy storage station, and substation within the target area; wherein, the value range of the unit operation monitoring period is 24h - 72h;

[0017] Extract the corresponding moments of the peak operating loads of the wind power station, photovoltaic power station, energy storage station, and substation in each unit operation monitoring period;

[0018] According to the corresponding moments of the peak operating loads of the wind power station, photovoltaic power station, energy storage station, and substation in each unit operation monitoring period, obtain the peak operating load moment offset rates corresponding to the wind power station, photovoltaic power station, energy storage station, and substation;

[0019] Use the peak operating load moment offset rates corresponding to the wind power station, photovoltaic power station, energy storage station, and substation to obtain the average peak operating load moment offset rates of the wind power station, photovoltaic power station, energy storage station, and substation corresponding to each unit operation monitoring period;

[0020] Use the average peak operating load moment offset rates of the wind power station, photovoltaic power station, energy storage station, and substation corresponding to each unit operation monitoring period to set the data collection frequencies of the data collection devices in the next unit operation monitoring period.

[0021] Further, using the average peak operating load moment offset rates of the wind power station, photovoltaic power station, energy storage station, and substation corresponding to each unit operation monitoring period to set the data collection frequencies of the data collection devices in the next unit operation monitoring period, specifically including:

[0022] Retrieve the average peak operating load moment offset rates of all unit operation monitoring periods that the operation has experienced, and generate a moment offset rate set;

[0023] Retrieve the adjustment ratios corresponding to the maximum data collection frequency adjustment amplitudes of all unit operation monitoring periods that the operation has experienced, and generate a frequency change ratio set;

[0024] Retrieve the maximum and minimum values of the average peak operating load moment offset rates from the moment offset rate set;

[0025] Retrieve the frequency change ratios corresponding to the maximum and minimum values of the average peak operating load moment offset rates from the frequency change ratio set;

[0026] Use the maximum and minimum values of the average peak operating load moment offset rates and the frequency change ratios corresponding to the maximum and minimum values to set the data collection frequencies of the data collection devices in the next unit operation monitoring period.

[0027] Furthermore, the balance management module includes:

[0028] A regional power grid monitoring unit, configured to monitor the operating status of the target regional power grid in real time based on the pre-processed power station operation data, and draw corresponding curve graphs based on the key indicators of the operating status of the target regional power grid;

[0029] An energy storage analysis unit, configured to extract the energy storage power station operation data from the power station operation data, calculate the average output of wind and solar power generation according to a pre-set calculation period based on the energy storage power station operation data, and at the same time, calculate the load power consumption within the same period based on the drawn curve graphs;

[0030] A balance adjustment unit, configured to calculate the difference between power generation and power supply based on the average output of wind and solar power generation and the load power consumption, judge the operating mode of the energy storage power station based on the calculation result, match the corresponding control instructions according to the judgment result, and regulate the energy storage power station.

[0031] Furthermore, the regional power grid monitoring unit specifically includes:

[0032] Obtain the pre-processed power station operation data, including power generation data, load data, and power grid status data, and determine the effective operating status of the target regional power grid;

[0033] Cluster the power station operation data according to the data characteristics of the power station operation data, construct a key indicator data group, and draw a comprehensive power generation curve, a load power curve, and a voltage fluctuation curve based on the target values of the key indicator data group;

[0034] Calculate the power balance parameter and the voltage stability parameter based on the drawn comprehensive power generation curve, load power curve, and voltage fluctuation curve, and evaluate the real-time balance state between wind and solar power generation and the load based on the calculation result.

[0035] Furthermore, the power prediction module includes:

[0036] A prediction model construction unit, configured to construct a BP neural network model based on the genetic algorithm, and train the BP neural network model based on the obtained historical power station operation data to establish a power prediction model;

[0037] A model optimization unit, configured to calculate the feature vector similarity between the sample data and the predicted data, continuously optimize the weight coefficients and thresholds of the neural network using the genetic algorithm, and output the optimized power prediction model;

[0038] The charge and discharge power calculation unit is configured to input the real-time collected power station operation data into the optimized power prediction model for prediction, obtain the predicted values of photovoltaic, wind power, and load power, and determine the charge and discharge power of the energy storage at each moment based on the multi-objective optimization algorithm.

[0039] Further, the model optimization unit specifically includes:

[0040] Determine the number of weights and thresholds of the BP neural network model, and perform real-number coding on the weight coefficients and thresholds of the BP neural network model to form the chromosomes of the genetic algorithm;

[0041] Each chromosome includes a set of initial weight and threshold combinations of the BP neural network model, constituting the initial chromosome population;

[0042] Input each set of weights and thresholds in the initial chromosome population into the BP neural network model to obtain the output result, compare the output result with the actual output of the training sample, determine the output error, and use it as the fitness function value of the genetic algorithm;

[0043] Based on the fitness function value, screen the chromosomes, perform multiple rounds of iterative optimization, and select the weight and threshold combination corresponding to the chromosome with the largest fitness function value from the final chromosome population as the initial weights and initial thresholds of the optimized BP neural network model;

[0044] Input the optimized initial weights and initial thresholds into the BP neural network model to output the power prediction model optimized based on the genetic algorithm.

[0045] Further, determining the charge and discharge power of the energy storage at each moment based on the multi-objective optimization algorithm specifically includes:

[0046] Obtain the predicted values of photovoltaic, wind power, and load power output by the optimized power prediction model, and perform normalization processing on the predicted values of different ranges and types of power;

[0047] Construct a multi-objective optimization scheduling model, calculate the multi-objective function value corresponding to each chromosome in combination with the preset constraint conditions, and convert the multi-objective function into a single-objective function for evaluation according to the set weight coefficients;

[0048] Based on the evaluation results, in combination with the real-time operation status of the regional target power grid and the real-time equipment status of the energy storage power station, determine the charge and discharge power of the energy storage at each moment.

[0049] Further, it also includes:

[0050] The scheduling decision-making module is configured to construct a scheduling decision-making scheme for the wind-solar energy storage power station according to the charge and discharge power of the energy storage at each moment and the real-time balance status of wind-solar power generation and load, in combination with the operation requirements and constraint conditions of the power grid.

[0051] A risk assessment module, configured to perform risk assessment on the operation process of the target area power grid, evaluate the operation performance of each power station in the target area power grid under different risk scenarios, and match corresponding risk response strategies and warning mechanisms.

[0052] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0053] Through the multi-objective data acquisition module, the target area is determined according to the power grid topology, load and wind-solar resource distribution, equipment is installed to collect data, and after screening and preprocessing, it is transmitted to provide a reliable data basis for subsequent modules, ensuring the accuracy and effectiveness of the data, avoiding interference from incorrect and abnormal data, the balance management module monitors the power grid in real time, analyzes the balance relationship between wind-solar power generation and power consumption load, and adjusts the operation mode of the energy storage power station accordingly; the power prediction module constructs and optimizes the model, accurately predicts the power, calculates the charge and discharge power of the energy storage, ensures the stable operation of the power grid, improves the energy utilization efficiency, the dispatching decision-making module formulates a dispatching plan by combining various information to ensure the safe, stable and economic operation of the power grid; the risk assessment module comprehensively considers various risk factors, evaluates the operation performance of each power station, and matches the response strategies and warning mechanisms to enhance the reliability and security of the power grid operation. Description of the Drawings

[0054] Figure 1 It is a module diagram of the wind-solar energy storage power station capacity prediction system based on multi-objectives of the present invention. Detailed Embodiments

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0056] In order to solve the technical problems that the prior art has limited accuracy in predicting wind-solar power generation, it is difficult to capture real-time changes, resulting in a large deviation between dispatching and reality; and most are single-objective optimizations, unable to take into account the technical problems of multi-faceted requirements such as economic benefits, energy storage costs and power grid impacts, please refer to Figure 1 , the following technical solutions are provided in this embodiment:

[0057] A multi-objective based wind-solar energy storage power station capacity prediction system, comprising:

[0058] The multi-objective data acquisition module is configured to determine the target area of the wind-solar energy storage power station, collect the power grid operation data of the target area, including the power generation operation data of the wind power station, the photovoltaic power station (including distributed photovoltaics), the energy storage power station, and the regional substation, where the power generation operation data includes the power generation power, the power generation amount, the load power, the meteorological data, etc., and preprocess the collected power generation operation data to remove the data with obvious errors or anomalies;

[0059] The balance management module is configured to obtain the preprocessed power generation operation data of the power grid, monitor the target area power grid in real time, analyze the balance relationship between wind-solar power generation and the electricity load in the target area, judge whether there is a power imbalance situation, and adjust the operation mode of the energy storage power station in the target area power grid according to the analysis result, and at the same time conduct the matching judgment of the energy storage capacity to provide support for the stable operation of the power grid and the reasonable planning of the energy storage power station;

[0060] The power prediction module is configured to construct a power prediction model based on the obtained power generation operation data, predict the photovoltaic power, wind power, and load power, and optimize the power prediction model to improve the prediction accuracy, and calculate the charge and discharge power of the energy storage at each moment according to the prediction result and the multi-objective optimization algorithm;

[0061] The dispatching decision-making module is configured to construct a dispatching decision-making scheme for the wind-solar energy storage power station according to the charge and discharge power of the energy storage at each moment and the real-time balance state of the wind-solar power generation and the load, combined with the operation requirements and constraints of the power grid, including the selection of the power generation and energy storage modes of the energy storage power station, the access and distribution of the wind-solar power generation, etc., to ensure the safe, stable, and economic operation of the power grid;

[0062] The risk assessment module is configured to conduct a risk assessment on the operation process of the target area power grid, consider factors such as the uncertainty of wind-solar power generation, the failure risk of energy storage devices, and power grid failures, analyze the influence degree of these factors on the operation of each power station in the target area power grid, evaluate the operation performance of each power station in the target area power grid under different risk scenarios, and match the corresponding risk response strategies and early warning mechanisms.

[0063] In this embodiment, the multi-objective data acquisition module accurately determines the target area, comprehensively collects various power station operation data and preprocesses it. The balance management module monitors the power grid in real time, analyzes the balance relationship between wind and solar power generation and electricity load, flexibly adjusts the operation mode and capacity ratio of the energy storage power station, and ensures the stability of the power grid. The power prediction module constructs and optimizes the model, accurately predicts the power, calculates the charge and discharge power of the energy storage in combination with the algorithm, and provides forward-looking operation data. The dispatching decision-making module formulates a comprehensive dispatching decision-making plan according to the operation requirements of the power grid to ensure the safe, stable and economic operation of the power grid. The risk assessment module comprehensively considers various risk factors, evaluates the operation of the power grid, analyzes the degree of influence, evaluates the operation performance and matches the response strategies and warning mechanisms to comprehensively enhance the reliability and security of the power grid operation.

[0064] In this embodiment, the multi-objective data acquisition module specifically includes:

[0065] The target area determination unit is configured to determine the target area that needs to be capacity predicted and managed according to the topological structure, load distribution and wind and solar resource distribution of the power grid, that is, the scope covered by one or more substations, and construct a target area power grid based on the wind power station, photovoltaic power station, energy storage power station and substation in the target area. Install data acquisition devices at each key node in the target area power grid;

[0066] The data acquisition unit is configured to collect the power station operation data of the target area power grid in real time based on the data acquisition device, including relevant data such as power generation power, power generation amount, load power, voltage, current, meteorological data (such as light intensity, wind speed, temperature, etc.), screen and preprocess the obtained power station operation data, identify and remove obvious errors or abnormal data, such as data loss, data mutation, etc., perform normalization processing on the data, and convert different types and ranges of data to the same scale for subsequent analysis and processing, and transmit the preprocessed power station operation data to the balance management module through the wireless communication network.

[0067] Further, the multi-objective data acquisition module specifically further includes:

[0068] Extract the operation load of each unit operation monitoring period corresponding to the wind power station, photovoltaic power station, energy storage power station and substation in the target area; wherein, the value range of the unit operation monitoring period is 24h - 72h;

[0069] Extract the corresponding moment of the peak value of the operation load of the wind power station, photovoltaic power station, energy storage power station and substation in each unit operation monitoring period;

[0070] According to the peak moment of the operating load of the wind power station, photovoltaic power station, energy storage power station, and substation corresponding to each unit's operation monitoring cycle, obtain the offset rate of the peak moment of the operating load of the wind power station, photovoltaic power station, energy storage power station, and substation;

[0071] Among them, the offset rate of the peak moment of the operating load is obtained through the following formula:

[0072]

[0073] Among them, R represents the offset rate of the peak moment of the operating load; T c represents the time interval between the peak moment of the operating load in the current unit operation monitoring cycle and the peak moment of the operating load in the previous unit operation monitoring cycle; T represents the time length corresponding to the unit operation monitoring cycle;

[0074] Use the offset rate of the peak moment of the operating load of the wind power station, photovoltaic power station, energy storage power station, and substation corresponding to each unit's operation monitoring cycle to obtain the average value of the offset rate of the peak moment of the operating load of the wind power station, photovoltaic power station, energy storage power station, and substation corresponding to each unit's operation monitoring cycle;

[0075] Use the average value of the offset rate of the peak moment of the operating load of the wind power station, photovoltaic power station, energy storage power station, and substation corresponding to each unit's operation monitoring cycle to set the data collection frequency of each data collection device in the next unit operation monitoring cycle.

[0076] The technical effects of the above technical solution are as follows: By calculating the peak moment offset rate of the operating load, it is possible to quantify the changes in the peak moments of the operating loads of each power station within different unit operating monitoring cycles. Based on this, the time patterns of the load changes of each power station can be accurately captured. For example, it is found that the peak moments of wind power stations and photovoltaic power stations show periodic or random offsets due to natural conditions (such as wind speed and light changes), and the peak load change patterns of energy storage power stations and substations due to factors such as electricity demand, providing an accurate basis for subsequent data collection and system dispatching. Using the average value of the peak moment offset rates of the operating loads of each power station to set the data collection frequency for the next unit operating monitoring cycle realizes the dynamic adaptive adjustment of the data collection frequency. When the average offset rate is large, it means that the peak moment of the load changes violently. At this time, increasing the data collection frequency can obtain the equipment operation data more timely and accurately, avoiding data omission; when the average offset rate is small and the peak moment of the load is relatively stable, the collection frequency is appropriately reduced to reduce data redundancy and processing pressure, improving the overall efficiency and resource utilization rate of the data collection system. Accurately grasping the peak moment offset of the operating load of each power station and reasonably adjusting the data collection frequency helps to detect potential abnormalities in equipment operation in advance. For example, an abnormal offset in the peak moment of the load may indicate equipment failure or sudden changes in external conditions. By collecting data at high frequencies, it can be monitored and warned in time, facilitating maintenance personnel to take measures to ensure the stable operation of wind power stations, photovoltaic power stations, energy storage power stations and substations, reducing the risk of system failures, and improving the reliability of the entire power supply system. This technical solution takes into account the peak moment offsets of various types of power stations, which is conducive to realizing the collaborative optimization of different energy generation, energy storage and power transformation links. By analyzing the offset rate and its average value, the intermittent power generation of wind power and photovoltaic power can be better coordinated with energy storage and power transformation, making the power supply more stable and efficient, and improving the comprehensive utilization efficiency of multiple energies. In terms of performance indicators, it is reflected in the improvement of the stability and continuity of power supply and the energy utilization efficiency.

[0077] Specifically, the data collection frequencies of each data collection device for the next unit operating monitoring cycle are set using the average value of the peak moment offset rates of the operating loads of the wind power station, photovoltaic power station, energy storage power station and substation corresponding to each unit operating monitoring cycle, which specifically includes:

[0078] Retrieve the average values of the peak moment offset rates of the operating loads of all unit operating monitoring cycles that have been run, and generate a moment offset rate set;

[0079] Retrieve the adjustment ratio corresponding to the maximum value of the data collection frequency adjustment amplitude of all unit operating monitoring cycles that have been run, and generate a frequency change ratio set;

[0080] Retrieve the maximum and minimum values from the average offset rate at the peak operating load moment in the set of moment offset rates;

[0081] Retrieve the frequency change ratios corresponding to the maximum and minimum values from the average offset rate at the peak operating load moment in the set of frequency change ratios;

[0082] Use the maximum and minimum values from the average offset rate at the peak operating load moment and the frequency change ratios corresponding to the maximum and minimum values to set the data collection frequencies of each data collection device in the next unit operating monitoring cycle.

[0083] Among them, the data collection frequencies of each data collection device in the next unit operating monitoring cycle are obtained through the following formula:

[0084]

[0085] Among them, f represents the data collection frequency of each data collection device in the next unit operating monitoring cycle; f 0 represents the preset initial data collection frequency; P max and P min respectively represent the maximum and minimum values from the average offset rate at the peak operating load moment; f max and f min respectively represent the maximum and minimum values from the average offset rate at the peak operating load moment and the frequency change ratios corresponding to the maximum and minimum values; x represents the frequency sensitivity adjustment coefficient, and its value range is 0.3 - 0.7. Among them, Utilize the periodicity and value range (-1 to 1) of the sine function to perform an associative mapping between the change range of the offset rate (P max −P min ) and the frequency change ratio range (f max −f min ) to make the adjustment of the collection frequency fluctuate within a suitable interval and achieve smooth and reasonable frequency adjustment.

[0086] The technical effects of the above technical solution are as follows: Based on the average deviation rate at the peak moment of historical operating load and its corresponding frequency change ratio, this solution dynamically adjusts the data acquisition frequency for the next unit operation monitoring period. It can achieve adaptive optimization of the acquisition frequency according to the characteristics of load fluctuations in different power stations. When the deviation rate fluctuates greatly, the acquisition frequency is increased to promptly capture changes in the equipment operating state; when the fluctuation is small, the frequency is decreased to reduce resource waste and improve data acquisition efficiency and system performance. By considering the maximum and minimum values of historical operating data and the corresponding frequency change ratio, it ensures that the data acquisition frequency can accurately match the actual requirements under different operating conditions. The collected data can more accurately reflect the operating state of the power station, reduce data errors and the risk of missed acquisitions, provide a reliable data basis for subsequent data analysis, fault diagnosis, and system regulation, and thus improve the reliability and stability of the entire energy system operation. Reasonably adjusting the data acquisition frequency enables the system to better cope with natural factors such as wind and light and external disturbances such as changes in electricity load. Timely obtaining accurate data facilitates maintenance personnel to quickly discover and handle potential problems, reduce the impact of disturbances on the power station operation, enhance the stability and anti-interference ability of the energy system, and ensure the continuity and security of power supply.

[0087] On the other hand, the above formula realizes the non-linear adjustment of the data acquisition frequency by using the sine function. Compared with linear adjustment, it can adjust the acquisition frequency more delicately and flexibly according to the change of the deviation rate at the peak moment of the operating load. When the deviation rate changes little, the frequency adjustment range is small; when the deviation rate changes greatly, the frequency adjustment range increases, making the frequency adjustment more in line with the actual operating requirements, improving the accuracy and adaptability of data acquisition, and optimizing the performance indicators. The formula limits the adjustment range of the acquisition frequency by setting conditions (0.63P max ≥P min ) and calculating based on the maximum and minimum values of historical deviation rates and the corresponding frequency change ratio. It avoids excessive fluctuations or unreasonable values of the acquisition frequency, ensures the stability of the data acquisition system operation, and thus improves the reliability and stability of the performance indicators of the entire energy system.

[0088] In this embodiment, the balance management module includes:

[0089] The regional power grid monitoring unit is configured to monitor the operating state of the target regional power grid in real time based on the preprocessed power station operating data, including parameters such as power generation power, load power, voltage, and frequency, draw corresponding curve graphs based on the key indicators of the operating state of the target regional power grid, and determine the real-time balance relationship between wind-solar power generation and electricity load, specifically including:

[0090] Obtain the pre - processed power station operation data, including power generation data (wind power generation power, photovoltaic power generation power, charge - discharge power of energy storage power station), load data (real - time load power of substation, historical load curve), and power grid status data (bus voltage, system frequency, line current), and determine the effective operation status of the target area power grid;

[0091] Cluster - process the power station operation data according to the data characteristics of the power station operation data, construct a key index data group, and draw a comprehensive power generation curve (superposition of wind power, photovoltaic power, and energy storage output), a load power curve (including comparison of real - time value and historical average), and a voltage fluctuation curve (marking the threshold lines of the upper and lower limits of the rated value) based on the target values of the key index data group;

[0092] Calculate the power balance parameter and voltage stability parameter based on the drawn comprehensive power generation curve, load power curve, and voltage fluctuation curve, and evaluate the real - time balance state of wind - solar power generation and load based on the calculation results;

[0093] The energy storage analysis unit is configured to extract the energy storage power station operation data from the power station operation data, including parameters such as the power station's electricity quantity, charge - discharge power, state of charge, etc., calculate the average output of wind - solar power generation based on the energy storage power station operation data according to a pre - set calculation period. At the same time, calculate the load electricity quantity within the same period based on the drawn curve graph;

[0094] The balance adjustment unit is configured to calculate the difference between the power generation and supply electricity quantity based on the average output of wind - solar power generation and the load electricity quantity, judge the operation mode of the energy storage power station based on the calculation results, match the corresponding control instructions according to the judgment results, and regulate the energy storage power station;

[0095] In this embodiment, when the difference between the power generation and supply electricity quantity increases, it is determined that the electricity load is greater than the wind - solar power generation output, and the energy storage power station starts the power generation mode to transmit electric energy to the power grid to supplement the power gap of the power grid; when the difference between the power generation and supply electricity quantity decreases, it is determined that the wind - solar power generation output is greater than the electricity load, and the energy storage power station starts the energy storage mode to store the excess electric energy to maintain the power balance of the power grid. At the same time, during the adjustment process, the equipment status of the energy storage power station (such as battery temperature, charge - discharge depth, etc.) is monitored in real - time to ensure that the charge - discharge operation is carried out within the safety threshold, prevent over - charge and over - discharge from damaging the energy storage equipment, and ensure the efficient and safe operation of the energy storage power station, thereby realizing the dynamic power balance and stable operation of the target area power grid.

[0096] In this embodiment, the power prediction module includes:

[0097] The prediction model construction unit is configured to construct a BP neural network model based on the genetic algorithm, train the BP neural network model based on the obtained historical power station operation data, establish a power prediction model, and determine the structure and parameters of the network;

[0098] The model optimization unit is configured to calculate the feature vector similarity between the sample data and the prediction data, continuously optimize the weight coefficients and thresholds of the neural network using the genetic algorithm, and output the optimized power prediction model, improving the prediction accuracy and generalization ability of the model and establishing a model with high precision and high matching degree;

[0099] The charge and discharge power calculation unit is configured to input the power station operation data collected in real time into the optimized power prediction model for prediction, obtain the predicted values of the photovoltaic, wind power, and load power, and determine the charge and discharge power of the energy storage at each moment based on the multi-objective optimization algorithm.

[0100] In this embodiment, the model optimization unit specifically includes:

[0101] Determine the number of weight values and thresholds of the BP neural network model, and perform real number encoding on the weight coefficients and thresholds of the BP neural network model to form the chromosomes of the genetic algorithm;

[0102] Each chromosome includes a set of initial weight values and threshold combinations of the BP neural network model, constituting the initial chromosome population;

[0103] Input each set of weight values and thresholds in the initial chromosome population into the BP neural network model to obtain the output result, compare the output result with the actual output of the training sample, determine the output error, and use it as the fitness function value of the genetic algorithm;

[0104] Based on the fitness function value, screen the chromosomes, perform multiple rounds of iterative optimization, and select the weight value and threshold combination corresponding to the chromosome with the largest fitness function value from the final chromosome population as the initial weight value and initial threshold of the optimized BP neural network model;

[0105] Input the optimized initial weight value and initial threshold into the BP neural network model, and output the power prediction model optimized by the genetic algorithm for predicting the photovoltaic, wind power, and load power. Compared with the unoptimized BP neural network model, it has higher prediction accuracy and stronger generalization ability.

[0106] In this embodiment, determining the charge and discharge power of the energy storage at each moment based on the multi-objective optimization algorithm specifically includes:

[0107] Obtain the predicted values of the photovoltaic, wind power, and load power output by the optimized power prediction model, which reflect the changes in the power generation of photovoltaic and wind power and the electricity load in the future period of time, and perform normalization processing on the predicted values of different ranges and types of power, unifying the predicted values of different ranges and types of power to the same scale for subsequent comprehensive calculation and comparison in the multi-objective optimization algorithm;

[0108] With the core objective of meeting the requirements of the dispatching plan by the combined output of wind, light, and energy storage and minimizing the operation cost of energy storage, a multi-objective optimal dispatching model is constructed. The multi-objective function values corresponding to each chromosome are calculated in combination with the preset constraint conditions, and the multi-objective function is transformed into a single-objective function for evaluation according to the set weight coefficients;

[0109] In this embodiment, for the target of the combined output of wind, light, and energy storage within the dispatching plan range, the sum of the squares of the deviations between the predicted combined output of wind, light, and energy storage and the dispatching plan is calculated to measure the target of the combined output of wind, light, and energy storage tracking the dispatching plan range. The smaller the deviation, the better the tracking effect. It includes the energy storage cost minimization function, comprehensively considering the investment cost, operation and maintenance cost of the energy storage device, and the efficiency loss cost during the charge and discharge process. The investment cost can be amortized to each time period according to a certain depreciation life; the operation and maintenance cost is related to the number of charge and discharge times, operation time, etc.; the charge and discharge efficiency loss cost is calculated according to the charge and discharge power and efficiency; at the same time, combined with the electricity market price, the economic benefits of charge and discharge in different time periods are calculated to make the power generation output more in line with the principle of maximizing economic benefits. For example, during high electricity price periods, priority is given to arranging energy storage discharge and wind and light power generation for grid connection; during low electricity price periods, energy storage charging is carried out;

[0110] Based on the evaluation results, combined with the real-time operation status of the regional target power grid and the real-time equipment status of the energy storage power station, the charge and discharge power of the energy storage at each moment is determined.

[0111] In this embodiment, the prediction model construction unit constructs a BP neural network model with the help of the genetic algorithm and trains it with historical data to determine the network structure and parameters, laying a foundation for power prediction. The model optimization unit iteratively optimizes the weight coefficients and thresholds by calculating the feature vector similarity and using the genetic algorithm, greatly improving the prediction accuracy and generalization ability of the model, establishing a high-precision and high-matching model, and making the prediction results more in line with the actual situation; the charge and discharge power calculation unit inputs the real-time data into the optimized model to obtain the prediction value, unifies the scale through normalization processing, constructs a multi-objective optimal dispatching model, comprehensively considers the objectives such as the combined output of wind, light, and energy storage within the dispatching plan range, energy storage cost minimization, and realizing the maximization of economic benefits by combining the electricity market price, calculates the multi-objective function value and transforms it into a single-objective function for evaluation, and then determines the charge and discharge power of the energy storage at each moment according to the real-time status of the regional target power grid and the energy storage power station, which can accurately plan the charge and discharge of the energy storage power station, ensure the stable operation of the power grid, improve the energy utilization efficiency, and at the same time ensure the economy and safety.

[0112] As mentioned above, it is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered within the protection scope of the present invention.

Claims

1. A multi-objective wind-solar-energy-storage power station capacity prediction system, characterized in that: include: A multi-target data acquisition module is configured to determine the target area of ​​the wind, solar and energy storage power station, collect the power station operation data of the power grid in the target area, and pre-process the collected power station operation data, extract the unit cycle operation load of each power station, calculate the peak moment deviation rate and the average value, and adjust the data acquisition frequency of each data acquisition device based on the calculation results; The balance management module is configured to obtain pre-processed power station operation data, conduct real-time monitoring of the target area power grid, analyze the balance relationship between wind and solar power generation and power load in the target area, determine whether there is a power imbalance, and adjust the operation mode of the energy storage power station in the target area power grid according to the analysis results; The power prediction module is configured to build a power prediction model based on the acquired power station operation data, predict photovoltaic, wind power and load power, optimize the power prediction model, and calculate the charging and discharging power of energy storage at each moment according to the prediction results and the multi-objective optimization algorithm.

2. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 1, characterized in that: Multi-target data acquisition module, specifically including: The target area determination unit is configured to determine the target area that needs capacity prediction and management according to the topological structure of the power grid, the load distribution, and the distribution of wind and solar resources, and to construct the target area power grid based on the wind power station, photovoltaic power station, energy storage station, and substation set up in the target area, and to install data acquisition equipment at each key node in the target area power grid; The data acquisition unit is configured to collect the power station operation data of the target area power grid in real time based on the data acquisition equipment, screen and preprocess the acquired power station operation data, identify and remove obviously erroneous or abnormal data, and transmit the preprocessed power station operation data to the balance management module through the wireless communication network.

3. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 1 or 2, characterized in that: The multi-target data acquisition module specifically includes: Extract the operating load corresponding to each unit operation monitoring cycle of the wind power station, photovoltaic power station, energy storage station and substation in the target area; wherein the value range of the unit operation monitoring cycle is 24h-72h; Extract the corresponding time of the peak operating load of wind power stations, photovoltaic power stations, energy storage stations and substations in each unit operation monitoring cycle; According to the corresponding time of the operating load peak of the wind power station, photovoltaic power station, energy storage station and substation in each unit operation monitoring cycle, the offset rate of the operating load peak time corresponding to the wind power station, photovoltaic power station, energy storage station and substation is obtained; The peak moment deviation rates of the operating loads corresponding to the wind power station, photovoltaic power station, energy storage station and substation are used to obtain the average value of the peak moment deviation rates of the operating loads of the wind power station, photovoltaic power station, energy storage station and substation corresponding to each unit operation monitoring cycle; The data collection frequency of each data collection device in the next unit operation monitoring cycle is set using the average value of the peak moment deviation rate of the operating load of the wind power station, photovoltaic power station, energy storage station and substation corresponding to each unit operation monitoring cycle.

4. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 3 is characterized in that: The data collection frequency of each data collection device in the next unit operation monitoring cycle is set by using the average value of the peak moment deviation rate of the operating load of the wind power station, photovoltaic power station, energy storage station and substation corresponding to each unit operation monitoring cycle, including: Retrieve the average value of the peak moment deviation rate of the operating load of all unit operating monitoring cycles that have been experienced, and generate a moment deviation rate set; Retrieve the adjustment ratio corresponding to the maximum value of the adjustment amplitude of the data acquisition frequency of all unit operation monitoring cycles that have been experienced, and generate a frequency change ratio set; Retrieving the maximum and minimum values ​​of the average values ​​of the deviation rates at the peak moment of the operating load from the set of moment deviation rates; Retrieving the frequency change ratios corresponding to the maximum and minimum values ​​of the average values ​​of the deviation rate at the peak moment of the operating load from the frequency change ratio set; The data collection frequency of each data collection device in the next unit operation monitoring cycle is set by using the maximum value and the minimum value in the average value of the deviation rate at the peak time of the operation load and the frequency change ratio corresponding to the maximum value and the minimum value.

5. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 2, characterized in that: Balance management module, including: A regional power grid monitoring unit is configured to monitor the operating status of the target regional power grid in real time based on the preprocessed power station operating data, and draw a corresponding curve graph based on key indicators of the operating status of the target regional power grid; The energy storage analysis unit is configured to extract the energy storage power station operation data from the power station operation data, calculate the average output of wind and solar power generation according to a preset calculation cycle based on the energy storage power station operation data, and calculate the load power in the same cycle based on the drawn curve graph; The balancing and regulating unit is configured to calculate the difference in power generation and supply based on the average output of wind and solar power generation and the load power, determine the operation mode of the energy storage power station based on the calculation result, match the corresponding control instructions according to the determination result, and regulate the energy storage power station.

6. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 5, characterized in that: Regional power grid monitoring unit, including: Obtain pre-processed power station operation data, including power generation data, load data, and grid status data, and determine the effective operation status of the target area grid; Clustering the power plant operation data according to its data characteristics, constructing a key indicator data group, and drawing a comprehensive power generation curve, a load power curve, and a voltage fluctuation curve based on the target values ​​of the key indicator data group; The power balance parameters and voltage stability parameters are calculated based on the drawn comprehensive power generation curve, load power curve and voltage fluctuation curve, and the real-time balance status of wind and solar power generation and load is evaluated based on the calculation results.

7. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 6, characterized in that: Power prediction module, including: A prediction model building unit is configured to build a BP neural network model based on a genetic algorithm, and train the BP neural network model based on the acquired historical power plant operation data to establish a power prediction model; a model optimization unit configured to calculate the feature vector similarity between the sample data and the prediction data, continuously optimize the weight coefficient and threshold of the neural network using a genetic algorithm, and output an optimized power prediction model; The charging and discharging power calculation unit is configured to input the power station operation data collected in real time into the optimized power prediction model for prediction, obtain the predicted values ​​of photovoltaic, wind power and load power, and determine the charging and discharging power of energy storage at each moment based on the multi-objective optimization algorithm.

8. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 7, characterized in that: Model optimization unit, specifically including: Determine the weights and thresholds of the BP neural network model, and encode the weight coefficients and thresholds of the BP neural network model into real numbers to form chromosomes of the genetic algorithm; Each chromosome includes a set of initial weights and threshold combinations of the BP neural network model, forming an initial chromosome population; Input each set of weights and thresholds in the initial chromosome population into the BP neural network model to obtain the output result, compare the output result with the actual output of the training sample, and determine the output error as the fitness function value of the genetic algorithm; The chromosomes are screened based on the fitness function value, and multiple rounds of iterative optimization are performed. The weight and threshold combination corresponding to the chromosome with the largest fitness function value is selected from the final chromosome population as the initial weight and initial threshold of the optimized BP neural network model; The optimized initial weights and initial thresholds are input into the BP neural network model, and the power prediction model optimized by the genetic algorithm is output.

9. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 8, characterized in that: The charging and discharging power of energy storage at each moment is determined based on a multi-objective optimization algorithm, including: Obtain the predicted values ​​of photovoltaic, wind power and load power output by the optimized power prediction model, and normalize the power prediction values ​​of different ranges and types; Construct a multi-objective optimization scheduling model, calculate the multi-objective function value corresponding to each chromosome in combination with the preset constraints, and convert the multi-objective function into a single objective function for evaluation according to the set weight coefficient; Based on the evaluation results, combined with the real-time operating status of the regional target power grid and the real-time equipment status of the energy storage power station, the charging and discharging power of the energy storage at each moment is determined.

10. The multi-objective wind-solar energy storage power station capacity prediction system according to claim 9, characterized in that: Also includes: The dispatching decision module is configured to construct a dispatching decision plan for the wind-solar-energy storage power station based on the energy storage charging and discharging power at each moment and the real-time balance state of wind-solar-energy generation and load, combined with the operation requirements and constraints of the power grid; The risk assessment module is configured to conduct risk assessment on the operation process of the target area power grid, evaluate the operation performance of each power station in the target area power grid under different risk scenarios, and match corresponding risk response strategies and early warning mechanisms.

Images