Method and device for identifying operation mode of high-proportion new energy power system

Abstract

The invention provides a high-proportion new energy power system operation mode identification method. The method comprises the steps of obtaining time sequence operation simulation data of a high-proportion new energy power system; determining a plurality of characteristic indexes of a daily operation scene of the high-proportion new energy power system according to the time sequence operation simulation data; performing clustering analysis on each feature index to obtain a clustering analysis result; and determining a typical operation mode of the high-proportion new energy power system according to a clustering analysis result and a pre-constructed typical operation mode screening rule. Compared with a traditional method, the method not only can dynamically adapt to a complex operation environment brought by high-proportion new energy access, but also can remarkably improve the identification capability of potential risks in an extreme operation mode, and extracts a typical operation mode with representativeness and risk revealing capability.

Description

Technical Field

[0001] This application relates to the field of power system operation mode analysis technology, and in particular to a method and device for identifying the operation mode of a high-proportion new energy power system. Background Technology

[0002] With increasing global emphasis on environmental protection and sustainable development, the utilization of new energy sources has become a significant trend in power system development. However, the integration of a high proportion of new energy sources has brought unprecedented challenges to the stable operation of power systems. Due to the intermittency and uncertainty of new energy sources, the operation of power systems has become more complex and variable, altering the operating characteristics of the power grid and posing new requirements for the operation, planning, protection, and stability analysis of power systems. Traditional analysis methods based on experience or static data are no longer sufficient to meet the needs of current power systems.

[0003] Traditional power system operation mode analysis relies heavily on experience, selecting typical operating scenarios for economic, safety, and stability assessments. These scenarios primarily include four typical patterns: high summer / low summer performance, high winter / low winter performance, and so on. However, with the increasing penetration of renewable energy and the proliferation of power electronics, traditional analytical methods are increasingly unable to accurately capture the actual operating conditions of power systems and reveal the risks faced by the system under extreme conditions. Therefore, identifying and extracting typical operating modes has become an urgent issue, and is of great significance for power system planning and grid security and stability analysis. Summary of the Invention

[0004] The purpose of this application is to address at least one of the aforementioned technical deficiencies, particularly the technical deficiency in the prior art where, in the context of high-proportion renewable energy penetration and power electronics, traditional analysis methods cannot accurately capture the actual operating conditions of the power system and are difficult to reveal the risks faced by the system under extreme conditions.

[0005] Firstly, this application provides a method for identifying the operation mode of a high-proportion renewable energy power system, the method comprising:

[0006] Obtain time-series operation simulation data of high-proportion renewable energy power systems;

[0007] Based on time-series simulation data, several characteristic indicators of daily operation scenarios for high-proportion renewable energy power systems were determined;

[0008] Cluster analysis was performed on each characteristic indicator to obtain the cluster analysis results;

[0009] Based on the cluster analysis results, and in accordance with the pre-constructed typical operation mode screening rules, typical operation modes of high-proportion renewable energy power systems are determined. The typical operation mode screening rules are used to screen typical operation modes that are representative and have risk disclosure capabilities. Typical operation modes are used to represent the key operational behavior characteristics of high-proportion renewable energy power systems under different operating scenarios.

[0010] In one embodiment, the step of obtaining time-series operation simulation data of a high-proportion renewable energy power system includes:

[0011] Acquire basic data of a high-proportion renewable energy power system and input the basic data into power system simulation software for simulation to obtain time-series operation simulation data.

[0012] In one embodiment, the time-series simulation data includes all time periods of the day, daily peak load periods, daily off-peak load periods, wind power output, photovoltaic power output, and power system load. The characteristic indicators include minimum wind power output during peak load periods, minimum photovoltaic power output during peak load periods, maximum wind power output during off-peak load periods, average photovoltaic power output throughout the day, daily wind power generation ratio, and daily photovoltaic power generation ratio.

[0013] The steps for determining multiple characteristic indicators of a high-proportion renewable energy power system's daily operation scenario based on time-series simulation data include:

[0014] Based on wind power output and daily peak load periods, determine the minimum wind power output during peak load periods, and based on photovoltaic power output and daily peak load periods, determine the minimum photovoltaic power output during peak load periods.

[0015] Determine the maximum wind power output during the off-peak period based on wind power output and daily off-peak load times;

[0016] The average daily photovoltaic output is determined based on the photovoltaic power generation output and all time periods of the day.

[0017] The daily percentage of wind power generation is determined based on wind power output, power system load, and all time periods of the day; similarly, the daily percentage of photovoltaic power generation is determined based on photovoltaic power output, power system load, and all time periods of the day.

[0018] In one embodiment, the step of performing cluster analysis on each feature index to obtain the cluster analysis results includes:

[0019] Determine the optimal number of clusters, and based on the optimal number of clusters, use the K-means clustering algorithm to perform cluster analysis on each feature index to obtain the cluster analysis results.

[0020] In one embodiment, the step of determining the optimal number of clusters includes:

[0021] Determine multiple candidate cluster numbers and calculate the silhouette coefficient corresponding to each candidate cluster number;

[0022] The candidate cluster number corresponding to the silhouette coefficient with the largest value is taken as the optimal cluster number.

[0023] In one embodiment, typical operating modes include a high-penetration renewable energy operating mode, a conventional power generation low-operation mode, and a high short-circuit current risk operating mode. The high-penetration renewable energy operating mode is used to represent the operating characteristics of power supply and demand fluctuations, voltage and frequency regulation problems caused by a high proportion of renewable energy access. The conventional power generation low-operation mode is used to represent the operating characteristics of low conventional unit operating capacity, which leads to insufficient peak shaving and reserve capacity. The high short-circuit current risk operating mode is used to represent the operating characteristics of conventional unit capacity being much greater than the load, resulting in increased short-circuit current levels and increased risks of equipment damage and system instability.

[0024] In one embodiment, the selection criteria for high-penetration new energy operation modes are:

[0025]

[0026] The selection criteria for the conventional power supply's low-power operation mode are:

[0027]

[0028] The screening criteria for high short-circuit current risk operating modes are as follows:

[0029]

[0030] in, Screening criteria representing high-penetration new energy operation modes; Indicates daily operation scenario, This represents the wind power output at time t. This represents the photovoltaic power output at time t. This represents the power system load at time t. The screening criteria represent the typical power supply's low-power operation mode. The power supply's on-time capacity is represented by time t. Screening criteria representing operating modes with high short-circuit current risk.

[0031] Secondly, this application provides a high-proportion renewable energy power system operation mode identification device, the device comprising:

[0032] The time-series operation simulation data acquisition module is used to acquire time-series operation simulation data of high-proportion renewable energy power systems;

[0033] The feature index determination module is used to determine multiple feature indicators for daily operation scenarios of high-proportion new energy power systems based on time-series simulation data.

[0034] The cluster analysis result determination module is used to perform cluster analysis on various feature indicators and obtain the cluster analysis results;

[0035] The typical operation mode determination module is used to determine the typical operation mode of a high-proportion renewable energy power system based on the cluster analysis results and according to the pre-built typical operation mode screening rules. The typical operation mode screening rules are used to select typical operation modes that are representative and have risk disclosure capabilities. The typical operation mode is used to represent the key operational behavior characteristics of the high-proportion renewable energy power system under different operating scenarios.

[0036] Thirdly, this application provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of any of the high-proportion new energy power system operation mode identification methods described in the above embodiments.

[0037] Fourthly, this application provides a computer device, including: one or more processors, and a memory;

[0038] The memory stores computer-readable instructions, which, when executed by one or more processors, perform the steps of any of the high-proportion renewable energy power system operation mode recognition methods described in the above embodiments.

[0039] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0040] The method for identifying the operation modes of high-proportion renewable energy power systems provided in this application comprehensively captures the dynamic operation characteristics of the system at different time scales by acquiring time-series simulation data of the high-proportion renewable energy power system, avoiding the limitations of traditional methods that rely solely on experience or static data. Furthermore, by extracting multiple characteristic indicators from daily operation scenarios, this method can accurately characterize the complex operation state of the power system under the background of high-proportion renewable energy access, solving the technical problem that traditional methods cannot accurately capture the system's operation characteristics. Through cluster analysis of characteristic indicators, this method can classify massive amounts of operation data into representative pattern clusters, thereby effectively screening out typical operation modes that reflect the system's operation patterns. Finally, based on pre-constructed typical operation mode screening rules for selecting representative and risk-revealing typical operation modes, this method can extract representative and risk-revealing typical operation modes, providing a scientific basis for power system planning, operation optimization, and stability analysis. It can be understood that typical operation modes represent the key operational behavior characteristics of high-proportion renewable energy power systems under different operation scenarios. Compared with traditional methods, this method can not only dynamically adapt to the complex operating environment brought about by high-proportion renewable energy access, but also significantly improve the ability to identify potential risks under extreme operating conditions. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart illustrating the method for identifying the operation mode of a high-proportion renewable energy power system provided in this application embodiment;

[0043] Figure 2 Example graphs showing the variation of silhouette coefficients under different cluster numbers k provided in embodiments of this application;

[0044] Figure 3 Example diagrams of daily operation scenarios provided in embodiments of this application;

[0045] Figure 4 Example diagram of the identification results of high penetration rate operation mode of new energy provided in the embodiments of this application;

[0046] Figure 5 A schematic diagram of the structure of the high-proportion new energy power system operation mode identification device provided in the embodiments of this application;

[0047] Figure 6This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] This application provides a method for identifying the operation mode of a high-proportion renewable energy power system. The following embodiments illustrate this method using a computer device as an example. It is understood that the computer device can be any device with data processing capabilities, including but not limited to a single server, server cluster, personal laptop, desktop computer, etc. Figure 1 As shown, the method may include the following steps:

[0050] S101: Obtain time-series operation simulation data of high-proportion new energy power systems.

[0051] Among them, a high-proportion renewable energy power system refers to a power system in which the installed capacity of renewable energy reaches a relatively high level. Such systems typically exhibit characteristics such as a high proportion of renewable energy generation, a high degree of power electronics, and high AC / DC coupling, resulting in complex operating characteristics and significant control challenges. Time-series operation simulation data refers to data generated sequentially over time during the operation of a power system, used to describe the system's operating status and changing patterns. This data may include renewable energy power forecast curves, system load forecast curves, bus load forecast curves, network topology, unit generating capacity, and power plant operation information.

[0052] In this step, firstly, the computer equipment receives or schedules historical operating data of the new energy power system, including information such as wind power, photovoltaic output, and load curves, and preprocesses the data, such as data cleaning, noise reduction, and missing value imputation. Subsequently, the computer equipment generates future time-series operating data based on physical models or data-driven models, such as time series prediction models based on deep learning, to simulate the system's operation under different meteorological conditions, load demand changes, and other factors.

[0053] In specific embodiments, the computer equipment can deploy a multi-layered simulation architecture. One layer is for refined modeling based on large-scale power system simulation software, and the other layer utilizes lightweight machine learning models for rapid prediction, enabling the generation of highly timely operational data in a short period. For example, in situations of drastic wind speed changes, the computer equipment can access historical data and wind power curves, combined with short-term prediction models, to quickly calculate the power output of the wind farm for the next 10 minutes, thereby providing auxiliary decision support for grid dispatch. Furthermore, the computer equipment can integrate edge computing capabilities, allowing the acquisition of time-series operational simulation data to proceed not only from the central server but also from local computing units at wind farms and photovoltaic power plants. This ensures that data acquisition and simulation calculations for local renewable energy power systems remain intact even in the event of faults or communication interruptions, enhancing the system's robustness.

[0054] It is understandable that obtaining time-series operation simulation data of high-proportion renewable energy power systems aims to accurately reflect the dynamic characteristics of renewable energy power systems and avoid the problem of traditional static data analysis methods failing after the penetration rate of renewable energy increases.

[0055] S102: Based on time-series simulation data, determine multiple characteristic indicators for the daily operation scenario of a high-proportion renewable energy power system.

[0056] The daily operation scenario refers to the typical operating state of the power system within a day, reflecting a combination of new energy output, load characteristics, and grid operation features at different times. Characteristic indicators are a set of parameters used to describe and quantify the key characteristics of the daily operation scenario, such as the proportion of new energy output, peak-to-valley load difference, grid frequency fluctuation range, and voltage deviation. These indicators can be used to identify grid operation modes and analyze stability risks.

[0057] In this step, the computer equipment first analyzes the time-series simulation data, extracting information such as renewable energy output, load changes, grid frequency, and voltage fluctuations. Since the raw data may contain missing values, outliers, or measurement errors, the computer equipment performs data cleaning, such as using interpolation to fill in missing values, removing outliers through statistical analysis, and employing data smoothing algorithms to reduce measurement noise, ensuring data accuracy and usability. To extract characteristic indicators of the daily operation scenario, the computer equipment needs to divide the time-series data into time windows, such as by hour, peak / valley periods, or load characteristics. Subsequently, the data from each time period is aggregated and calculated, such as obtaining the average, maximum, minimum, and standard deviation of renewable energy output, as well as statistical characteristics such as load change rate and frequency deviation range, providing a foundation for subsequent analysis.

[0058] Computer equipment calculates multiple characteristic indicators based on data aggregation results. For example, it calculates the renewable energy penetration rate as the proportion of renewable energy output to total power generation in the system to assess the impact of renewable energy on the system; it calculates the load peak-valley difference to reflect load fluctuation characteristics; it calculates the grid frequency fluctuation range and voltage deviation to analyze the stability of the power system; and it calculates the load power factor to measure the power quality of the system. These characteristic indicators can quantitatively describe the characteristics of different daily operating scenarios, providing data support for subsequent pattern recognition.

[0059] It is understandable that by extracting feature indicators based on time-series simulation data using computer equipment, the key characteristics of daily operation scenarios of high-proportion new energy power systems can be comprehensively quantified, avoiding the subjectivity and uncertainty of traditional experience-based analysis methods.

[0060] S103: Perform cluster analysis on each characteristic index to obtain the cluster analysis results.

[0061] Cluster analysis is a data analysis method that divides data into different categories or groups by calculating the similarity between different data points. This results in data points within the same category having high similarity, while data points in different categories show greater differences. Cluster analysis can help identify and understand the characteristic patterns of power systems under different operating conditions. The cluster analysis result is the output of the clustering process, representing the grouping of characteristic indicators according to similarity, and can be used to further determine typical operating modes.

[0062] In this step, before cluster analysis, the computer equipment can preprocess each feature indicator to ensure it is suitable for the clustering algorithm. First, the data can be standardized to eliminate dimensional differences between different indicators. For example, the numerical ranges of renewable energy penetration rate and load peak-valley difference may differ; standardization transforms them into dimensionless data, preventing certain indicators from having an excessive impact on the clustering results. Standardization often uses the Z-score standardization method, which involves subtracting the mean from the data for each indicator and dividing by the standard deviation.

[0063] Depending on the nature of the characteristic indicators and the amount of data, computer equipment can select appropriate clustering algorithms for analysis. Common clustering algorithms include hierarchical clustering and density-based spatial clustering. Hierarchical clustering methods calculate the similarity between characteristic indicators and progressively merge or split clusters, making them suitable for datasets without a predefined number of clusters. Density-based spatial clustering, on the other hand, effectively identifies noisy data and irregularly shaped clusters by dividing dense regions. In power system data, since data points may have different density distributions, density-based spatial clustering algorithms can better identify and separate irregular operating patterns, thus achieving better clustering results in complex power systems.

[0064] Computer equipment performs clustering calculations based on a selected clustering algorithm. For example, when using hierarchical clustering, the equipment calculates a distance matrix based on the similarity between feature indicators and progressively merges similar indicators into clusters until a termination condition is met. The final clustering result is a category label for each feature indicator, representing their similarity during system operation. These results will help analyze system characteristics under different operating conditions, such as the impact of different renewable energy penetration rates on the power grid and the impact of load fluctuations on system stability.

[0065] In one example, taking a regional renewable energy power system as an example, computer equipment acquires a week's worth of time-series operational simulation data and calculates characteristic indicators such as renewable energy penetration rate, load peak-valley difference, and grid frequency fluctuation. Next, the computer equipment standardizes these characteristic indicators to ensure the data are within the same unit of measurement. Then, a density-based spatial clustering algorithm is used for cluster analysis, classifying these characteristic indicators into different categories, such as "low load state," "high load state," and "abnormal frequency fluctuation state." The results of the cluster analysis are stored as category labels and provide data support for the subsequent selection of typical operating modes.

[0066] It is understandable that cluster analysis of various characteristic indicators can extract operating patterns with similar characteristics from complex power system operation data, avoiding the uncertainty and subjectivity of traditional experience-based manual classification methods. Cluster analysis can help identify the characteristic behaviors of the power grid, providing a scientific basis for subsequent selection of typical operating patterns and risk prediction, thereby improving the operating efficiency, reliability, and stability of the power system.

[0067] S104: Based on the cluster analysis results, and in accordance with the pre-constructed typical operation mode screening rules, determine the typical operation mode of the high-proportion new energy power system.

[0068] Typical operation mode screening rules are used to screen typical operation modes that are representative and have risk disclosure capabilities. Typical operation modes are used to represent the key operational behavior characteristics of high-proportion renewable energy power systems under different operating scenarios.

[0069] The typical operating mode screening rules refer to a predefined set of rules used to select representative and typical operating states from the cluster analysis results. These rules are typically designed based on historical power system data, operational experience, or expert knowledge to identify key operating modes, such as high load and high penetration rates. Typical operating modes, on the other hand, refer to a set of patterns that represent the typical operating behavior of a high-proportion renewable energy power system under specific conditions. These patterns reflect the key characteristics of the system under different operating scenarios and can effectively reveal the stability, efficiency, and security of the power system.

[0070] In this step, the computer equipment first needs to load the results of the previous clustering analysis. Each cluster contains a set of characteristic indicators representing different operating states of the power system. The computer equipment imports the clustering results into memory and assigns a unique identifier to each category. The computer equipment then analyzes the clustering results according to pre-designed typical operating mode screening rules. These screening rules may include specific thresholds, such as a renewable energy penetration rate exceeding a certain percentage or grid frequency fluctuations within a certain range. The computer equipment iterates through each category in the clustering results, determining whether it meets these rules. For example, a certain cluster might correspond to a state of "high load, high renewable energy penetration rate." If this category meets the preset rules, it is considered one of the typical operating modes.

[0071] When a category meets the screening rules, the computer equipment identifies that category as the corresponding typical operating mode. For each category that meets the screening criteria, the computer equipment generates a corresponding identifier or label and saves these typical modes to a database or other storage system for later use. If a category meets the screening rules under multiple conditions, it may be further refined or divided into multiple typical modes. Ultimately, the computer equipment outputs all typical operating modes that meet the screening rules, serving as the basis for power system planning, operation optimization, or risk assessment. These modes support power system design, stability analysis, and emergency response.

[0072] In one example, within a high-proportion renewable energy power system, computer equipment acquired a week's worth of cluster analysis results. The results categorized the system's operating scenarios into several types, such as "low penetration rate, high load," "high penetration rate, low load," and "high penetration rate, large frequency fluctuations." Based on preset screening rules, the computer equipment determined that the power system enters a "high load, high penetration rate" state when the renewable energy penetration rate exceeds 50% and the load exceeds 80%. This state is designated as a typical operating mode and output as key data for system analysis. These typical modes will provide a basis for optimizing power system operation.

[0073] It is understandable that by combining cluster analysis results with preset screening rules, representative typical operating modes can be accurately identified, which helps to reveal the key operating status of high-proportion new energy power systems.

[0074] In the above embodiments, by acquiring time-series simulation data of a high-proportion renewable energy power system, the dynamic operating characteristics of the system at different time scales can be comprehensively captured, avoiding the limitations of traditional methods that rely solely on experience or static data. Furthermore, by extracting multiple characteristic indicators from daily operating scenarios, this method can accurately characterize the complex operating state of the power system under the background of high-proportion renewable energy access, solving the technical problem that traditional methods cannot accurately capture system operating characteristics. Through cluster analysis of characteristic indicators, this method can classify massive amounts of operating data into representative pattern clusters, thereby effectively screening out typical operating patterns that reflect the system's operating rules. Finally, based on pre-constructed typical operating pattern screening rules, this method can extract typical operating patterns with representativeness and risk disclosure capabilities, providing a scientific basis for power system planning, operation optimization, and stability analysis. Compared with traditional methods, this method can not only dynamically adapt to the complex operating environment brought about by high-proportion renewable energy access, but also significantly improve the ability to identify potential risks under extreme operating conditions.

[0075] In one embodiment, the step of obtaining time-series operation simulation data of a high-proportion renewable energy power system includes:

[0076] Acquire basic data of a high-proportion renewable energy power system and input the basic data into power system simulation software for simulation to obtain time-series operation simulation data.

[0077] The basic data for a high-proportion renewable energy power system refers to the initial data set including various parameters of the power system. This initial data set includes annual load curves, renewable energy output curves, power transmission / reception curves, and unit parameter information. This data reflects the actual operating status of the power system at different time periods and is used for power system simulation. "Power system simulation software" refers to software tools that simulate and predict power system operation using computer models. Simulation software can simulate the behavior of the power system under various operating conditions using the input basic data and output time-series operational simulation data, including time-series information such as grid voltage, current, and power flow direction.

[0078] Specifically, the computer equipment first needs to extract basic data from various data sources related to the high proportion of new energy power systems. This data may come from real-time monitoring systems, meteorological data sources, historical operational data, or forecasting models. The computer equipment will access these data sources according to predetermined data interfaces, automatically importing relevant system parameters into local data storage. After acquiring the basic data, the computer equipment will perform data cleaning, noise reduction, and formatting. If the data contains missing or outliers, the computer equipment will process it using interpolation or other data imputation methods to ensure the integrity and consistency of the input data. Simultaneously, the computer equipment may need to perform unit conversion or normalization of the data to meet the requirements of the simulation software.

[0079] The computer equipment formats the cleaned and processed basic data according to the interface specifications of the simulation software and inputs it into the power system simulation software. Upon receiving this data, the simulation software simulates the power system based on its built-in mathematical models, algorithms, and physical laws. Within the simulation software, the computer equipment initiates the simulation process by setting specific simulation parameters, such as simulation time, step size, and load variation range. The simulation software calculates the state changes of the power system in real time over a specified period, based on the basic data and the mathematical model of the power system, including information such as power generation, load demand, grid load, power flow, and system stability. After the simulation is complete, the computer equipment extracts time-series simulation data from the simulation software. This data is typically a time series result, detailing the dynamic changes of various important parameters of the power system during the simulation. The equipment stores this data in a database for subsequent analysis or decision-making.

[0080] In one example, in a high-proportion renewable energy power system, the computer equipment first acquires the basic data of the power system, including the annual load curve, renewable energy output curve, power transmission / receiving curve, and unit parameter information. Then, the data is preprocessed, such as format conversion, missing data imputation, and consistency checks. The processed data is then input into the simulation software, which performs a time-series simulation of 8760 hours throughout the year to calculate the output and operating status of each unit at different times.

[0081] Furthermore, to ensure the accuracy of the analysis, this method employs existing mature power system simulation software, inputting basic power system data for operational simulation. This process analyzes future operational scenarios and extracts representative typical daily operational scenarios. Since practical applications require safety and stability verification of future operational states, it is unnecessary to verify all data individually. Instead, cluster analysis is used to classify massive amounts of operational data and filter out typical scenarios that reflect the system's operational patterns. During safety and stability analysis, verification is directly performed on these typical daily operational scenarios, ensuring both the representativeness of the analysis and improving computational efficiency.

[0082] In this embodiment, by acquiring basic data from a high-proportion renewable energy power system and inputting it into simulation software, the computer equipment can comprehensively simulate the dynamic behavior of the power system under various operating conditions. This process ensures that the simulation is based on real-time and accurate system data, reflecting the operational performance of the renewable energy system under different conditions. After the data is input into the simulation software, high-precision time-series operational simulation data can be generated.

[0083] In one embodiment, the time-series simulation data includes all time periods of the day, daily peak load periods, daily off-peak load periods, wind power output, photovoltaic power output, and power system load. The characteristic indicators include minimum wind power output during peak load periods, minimum photovoltaic power output during peak load periods, maximum wind power output during off-peak load periods, average photovoltaic power output throughout the day, daily wind power generation ratio, and daily photovoltaic power generation ratio.

[0084] The steps for determining multiple characteristic indicators of a high-proportion renewable energy power system's daily operation scenario based on time-series simulation data include:

[0085] Based on wind power output and daily peak load periods, determine the minimum wind power output during peak load periods, and based on photovoltaic power output and daily peak load periods, determine the minimum photovoltaic power output during peak load periods.

[0086] Determine the maximum wind power output during the off-peak period based on wind power output and daily off-peak load times;

[0087] The average daily photovoltaic output is determined based on the photovoltaic power generation output and all time periods of the day.

[0088] The daily percentage of wind power generation is determined based on wind power output, power system load, and all time periods of the day; similarly, the daily percentage of photovoltaic power generation is determined based on photovoltaic power output, power system load, and all time periods of the day.

[0089] Specifically, the computer equipment first acquires time-series simulation data from power system simulation software and preprocesses the data, including format conversion, data cleaning, and time alignment, to ensure the accuracy of the calculations. Next, the computer equipment extracts data from the daily peak load periods and calculates the minimum output of wind and solar power during these periods, recording them as the minimum wind and solar power outputs during peak load periods to assess the minimum supply level of renewable energy generation under high load demand. Subsequently, the computer equipment extracts data from the daily off-peak load periods and calculates the maximum wind power output during these periods, thus identifying the maximum power supply capacity of wind power under low load demand. To measure the overall output level of solar power generation, the computer equipment extracts the solar power output from data across all time periods each day and calculates its average value to obtain the daily average solar power output. Finally, to assess the penetration of renewable energy into the system load, the computer equipment calculates the total daily wind and solar power generation and divides it by the total load of the power system to obtain the daily wind power generation share and the daily solar power generation share.

[0090] In one example, the minimum output of wind and solar power during peak load periods reflects the support capacity of new energy sources for the system during peak load periods, and its expression is:

[0091]

[0092]

[0093] in, and These represent the wind power and photovoltaic power output at time t, respectively. This indicates the peak load period of the day. This indicates the minimum output of wind power during peak load periods. This indicates the minimum output of photovoltaic power during peak load periods.

[0094] The maximum wind power output during off-peak hours reflects the absorption pressure exerted on the system by new energy sources during nighttime off-peak hours, and its expression is:

[0095]

[0096] in, Indicates the period of lowest daily load. This indicates the maximum wind power output during periods of low load.

[0097] The average daily photovoltaic output reflects the average photovoltaic output level, and its expression is:

[0098]

[0099] Among them, T D It represents all time periods of the day. This indicates the average daily photovoltaic output.

[0100] The daily wind and solar power generation ratio represents the proportion of daily wind power generation and daily solar power generation to the load. The higher the ratio, the more serious the problem of renewable energy consumption in the system, and the greater the peak-shaving pressure. Its expression is:

[0101]

[0102]

[0103] in, The load represents time t. This indicates the ratio of daily wind power generation to load. This indicates the ratio of daily photovoltaic power generation to load.

[0104] In this embodiment, by filtering data during peak and off-peak load periods and extracting extreme values, the impact of renewable energy generation on grid dispatch can be accurately assessed, and potential power supply shortages or surpluses can be identified. Calculating the average output of photovoltaic power can reflect the stability of photovoltaic power generation, while calculating the daily power generation ratio of wind and photovoltaic power can measure the penetration of renewable energy and its impact on the grid.

[0105] In one embodiment, the step of performing cluster analysis on each feature index to obtain the cluster analysis results includes:

[0106] Determine the optimal number of clusters, and based on the optimal number of clusters, use the K-means clustering algorithm to perform cluster analysis on each feature index to obtain the cluster analysis results.

[0107] The optimal number of clusters refers to the number of clusters determined by a specific evaluation method during cluster analysis. This ensures that the data maintains good discriminative power in category division while avoiding over-clustering or under-clustering. The K-means clustering algorithm is a partitioning-based method that classifies data points according to their similarity. Through iterative optimization, it aims to make data points within each category as similar as possible, while maximizing the differences between categories.

[0108] Specifically, the computer equipment first loads time-series simulation data of a high-proportion renewable energy power system and preprocesses the characteristic indicators, including normalization, to ensure that data from different units can be compared within the same numerical range. Subsequently, the computer equipment uses methods such as the elbow method or DBI (Davies-Bouldin Index) to calculate the clustering quality corresponding to different numbers of clusters, and determines the optimal number of clusters to ensure the rationality and stability of the cluster analysis.

[0109] After determining the optimal number of clusters, the computer initializes the K-means clustering algorithm, randomly selecting initial cluster centers or using the K-means++ method. Then, the computer calculates the distance between each feature index data point and each cluster center, assigning each data point to the nearest cluster center based on Euclidean distance or other distance metrics. After data point assignment, the computer recalculates the center point of each cluster and iterates this process until the change in cluster centers is less than a set convergence threshold or the set maximum number of iterations is reached. Finally, the computer outputs the cluster analysis results.

[0110] In this embodiment, by first determining the optimal number of clusters, the poor clustering effect caused by blindly setting the number of categories can be avoided, thus improving the scientific nature of data classification. The K-means clustering algorithm can reasonably divide the operating characteristics of the new energy power system based on data similarity, making the characteristics of each operating scenario clearer and revealing the typical features under different operating states.

[0111] In one embodiment, the step of determining the optimal number of clusters includes:

[0112] Determine multiple candidate cluster numbers and calculate the silhouette coefficient corresponding to each candidate cluster number;

[0113] The candidate cluster number corresponding to the silhouette coefficient with the largest value is taken as the optimal cluster number.

[0114] The candidate cluster number refers to a set of possible cluster numbers set before performing cluster analysis to determine the optimal number of clusters. These candidate values ​​are usually selected within a reasonable range to allow for subsequent evaluation of the clustering effect under different cluster numbers. The silhouette coefficient is an indicator used to evaluate the quality of clustering results, with a value ranging from -1 to 1. A higher silhouette coefficient indicates better clustering results, meaning that samples within the same category have high similarity, while samples between different categories have high discriminative power.

[0115] Specifically, the computer equipment first receives or loads time-series simulation data of the new energy power system and preprocesses the characteristic indicators, such as normalization, to ensure that data of different dimensions can be calculated on the same scale. Subsequently, the equipment sets a set of candidate cluster numbers, for example, an integer range from 2 to 10, as the number of possible categories, and performs cluster analysis for each candidate cluster number.

[0116] For each candidate cluster number, the computer device executes a clustering algorithm and calculates the silhouette coefficient corresponding to that cluster number based on the clustering results. Specifically, the computer device first calculates the average distance from each sample point to the center of its class (intra-class similarity), and then calculates the average distance from that sample point to the nearest centers of other classes (inter-class separation). These calculations yield the silhouette coefficient for each data point, and the average silhouette coefficient for the entire dataset is further calculated. After calculating the silhouette coefficients for all candidate cluster numbers, the computer device compares these values ​​and selects the candidate cluster number corresponding to the largest silhouette coefficient as the final optimal cluster number.

[0117] In this embodiment, by calculating the silhouette coefficients of multiple candidate cluster numbers and selecting the candidate cluster number corresponding to the maximum value as the optimal cluster number, it can be ensured that the final selected number of categories can maximize the compactness of intra-cluster data and the separation of inter-cluster data, thereby improving the quality of cluster analysis.

[0118] In one embodiment, typical operating modes include a high-penetration renewable energy operating mode, a conventional power generation low-operation mode, and a high short-circuit current risk operating mode. The high-penetration renewable energy operating mode is used to describe the operating characteristics of power supply and demand fluctuations, voltage and frequency regulation problems caused by a high proportion of renewable energy access. The conventional power generation low-operation mode is used to describe the operating characteristics of low conventional unit operating capacity, which leads to insufficient peak shaving and reserve capacity. The high short-circuit current risk operating mode is used to describe the operating characteristics of conventional unit capacity being much greater than the load, resulting in increased short-circuit current levels and increased risks of equipment damage and system instability.

[0119] Based on cluster analysis results from daily operation scenarios, and further combined with the characteristics of various safety and stability issues, a typical operation mode screening index system was constructed. This index system is used to extract representative typical modes from each operation scenario, effectively capturing the key operational characteristics of the system under different scenarios, and also covering high-risk scenarios related to system safety and stability, providing a scientific and reliable basis for the selection of operation modes in power system safety and stability analysis.

[0120] Within this framework, focusing on stability risks within the power system, three typical operating modes were selected for study: High-Penetration Renewable Energy Operation Mode (HPRE mode), Conventional Power Generation with Low Operating Capacity (LCPS mode), and High Short-Circuit Current Risk Operation Mode (HSCR mode). HPRE mode addresses the issue that significant fluctuations in renewable energy generation can disrupt power supply-demand balance, voltage stability, and frequency regulation; therefore, analysis of this mode is necessary to ensure the safe and stable operation of the power system. LCPS mode reflects the challenges arising from a reduced reliance on conventional power and an increased reliance on renewable energy. This shift may lead to insufficient peak-shaving and reserve capacity, impacting power supply-demand balance and emergency response; therefore, analyzing this mode is crucial for optimizing power source structure and improving grid flexibility. HSCR mode typically occurs when the operating capacity of conventional units far exceeds the system load, resulting in relatively high short-circuit current levels in the grid. In this situation, grid equipment needs to withstand greater current surges, increasing the risk of equipment damage and grid instability; therefore, identifying this mode is crucial for grid equipment protection and operational safety.

[0121] In one embodiment, the screening criteria for high-penetration new energy operation modes are:

[0122]

[0123] The selection criteria for the conventional power supply's low-power operation mode are:

[0124]

[0125] The screening criteria for high short-circuit current risk operating modes are as follows:

[0126]

[0127] in, Screening criteria representing high-penetration new energy operation modes; Indicates daily operation scenario, This represents the wind power output at time t. This represents the photovoltaic power output at time t. This represents the power system load at time t. The screening criteria represent the typical power supply's low-power operation mode. The power supply's on-time capacity is represented by time t. Screening criteria representing operating modes with high short-circuit current risk.

[0128] Understandably, to achieve accurate screening of typical operating modes, computer equipment first extracts key characteristic indicators such as daily peak load periods, off-peak load periods, wind power output, photovoltaic power output, and power system load based on time-series simulation data of the power system. Then, cluster analysis is used to classify each operating scenario and determine representative operating modes. Building on this, a screening index system is constructed by combining safety and stability analysis to identify high-risk scenarios that may occur during system operation, ensuring that these scenarios fully reflect the stability characteristics of the power system under different operating states. In this way, key risk points under different operating scenarios can be effectively identified, and in-depth research on different operating modes can further guide the formulation of power system safety protection measures, improving the grid's adaptability to environments with a high proportion of renewable energy.

[0129] To facilitate understanding of the scheme in this application, specific examples are provided below.

[0130] Taking a certain power system as an example, the region has abundant renewable energy resources. Wind power is developing rapidly and is expected to exceed 30% by 2030. High wind power grid connection poses challenges to system stability. Therefore, this system was chosen as the research object to analyze and identify the operation modes of high-proportion renewable energy power systems through data mining techniques. Figure 2 The figure shows the variation of the silhouette coefficient under different numbers of clusters k. It can be seen from the figure that the silhouette coefficient is the highest when k=5. Therefore, the number of clusters k is selected as 5. Figure 3 The clustering results for daily operation scenarios are shown, with the result of clustering scenario 1 selected for display. Figure 4 The results of extracting the high-penetration new energy operation mode under clustering scenario 1 are shown in the red box in the figure.

[0131] The following describes the high-proportion renewable energy power system operation mode identification device provided in the embodiments of this application. The high-proportion renewable energy power system operation mode identification device described below can be referred to in correspondence with the high-proportion renewable energy power system operation mode identification method described above. Figure 5 As shown, this application provides a device for identifying the operating mode of a high-proportion renewable energy power system. The device includes:

[0132] The timing simulation data acquisition module 201 is used to acquire timing simulation data of a high-proportion renewable energy power system.

[0133] The feature index determination module 202 is used to determine multiple feature indicators for the daily operation scenario of a high-proportion new energy power system based on time-series operation simulation data.

[0134] The cluster analysis result determination module 203 is used to perform cluster analysis on each feature index and obtain the cluster analysis results;

[0135] The typical operation mode determination module 204 is used to determine the typical operation mode of the high-proportion new energy power system based on the cluster analysis results and according to the pre-constructed typical operation mode screening rules. The typical operation mode screening rules are used to screen typical operation modes with representativeness and risk disclosure capabilities. The typical operation mode is used to represent the key operational behavior characteristics of the high-proportion new energy power system under different operating scenarios.

[0136] In one embodiment, the time-series simulation data acquisition module 201 includes:

[0137] The time-series operation simulation data acquisition unit is used to acquire basic data of a high-proportion renewable energy power system and input the basic data into power system simulation software for simulation to obtain time-series operation simulation data.

[0138] In one embodiment, the time-series simulation data includes all time periods of the day, daily peak load periods, daily off-peak load periods, wind power output, photovoltaic power output, and power system load. The characteristic indicators include minimum wind power output during peak load periods, minimum photovoltaic power output during peak load periods, maximum wind power output during off-peak load periods, average photovoltaic power output throughout the day, daily wind power generation ratio, and daily photovoltaic power generation ratio.

[0139] Feature index determination module 202 includes:

[0140] The first characteristic indicator determination unit is used to determine the minimum wind power output during the peak load period based on wind power output and daily peak load time period, and to determine the minimum photovoltaic output during the peak load period based on photovoltaic power output and daily peak load time period.

[0141] The second characteristic indicator determination unit is used to determine the maximum wind power output during the load off-peak period based on wind power output and daily load off-peak time periods.

[0142] The third characteristic indicator determination unit is used to determine the average daily photovoltaic output based on the photovoltaic power generation output and all time periods of the day;

[0143] The fourth characteristic indicator determination unit is used to determine the daily wind power generation ratio based on wind power output, power system load, and all time periods of the day, and to determine the daily photovoltaic power generation ratio based on photovoltaic power output, power system load, and all time periods of the day.

[0144] In one embodiment, the cluster analysis result determination module 203 includes:

[0145] The cluster analysis result determination unit is used to determine the optimal number of clusters, and based on the optimal number of clusters, the K-means clustering algorithm is used to perform cluster analysis on each feature index to obtain the cluster analysis results.

[0146] In one embodiment, the cluster analysis result determination unit includes:

[0147] The silhouette coefficient calculation subunit is used to determine the number of multiple candidate clusters and calculate the silhouette coefficient corresponding to each candidate cluster number.

[0148] The optimal cluster number determination sub-unit is used to select the candidate cluster number corresponding to the silhouette coefficient with the largest value as the optimal cluster number.

[0149] In one embodiment, typical operating modes include a high-penetration renewable energy operating mode, a conventional power generation low-operation mode, and a high short-circuit current risk operating mode. The high-penetration renewable energy operating mode is used to describe the operating characteristics of power supply and demand fluctuations, voltage and frequency regulation problems caused by a high proportion of renewable energy access. The conventional power generation low-operation mode is used to describe the operating characteristics of low conventional unit operating capacity, which leads to insufficient peak shaving and reserve capacity. The high short-circuit current risk operating mode is used to describe the operating characteristics of conventional unit capacity being much greater than the load, resulting in increased short-circuit current levels and increased risks of equipment damage and system instability.

[0150] In one embodiment, the screening criteria for high-penetration new energy operation modes are:

[0151]

[0152] The selection criteria for the conventional power supply's low-power operation mode are:

[0153]

[0154] The screening criteria for high short-circuit current risk operating modes are as follows:

[0155]

[0156] in, Screening criteria representing high-penetration new energy operation modes; Indicates daily operation scenario, This represents the wind power output at time t. This represents the photovoltaic power output at time t. This represents the power system load at time t. The screening criteria represent the typical power supply's low-power operation mode. The power supply's on-time capacity is represented by time t. Screening criteria representing operating modes with high short-circuit current risk.

[0157] In one embodiment, this application also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the high-proportion renewable energy power system operation mode identification method as described in any of the above embodiments.

[0158] In one embodiment, this application also provides a computer device storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the high-proportion renewable energy power system operation mode identification method as described in any of the above embodiments.

[0159] Indicatively, such as Figure 6 As shown, Figure 6 This is a schematic diagram of the internal structure of a computer device 300 provided in an embodiment of this application. The computer device 300 can be provided as a server. (Refer to...) Figure 6 The computer device 300 includes a processing component 302, which further includes one or more processors, and memory resources represented by memory 301 for storing instructions, such as application programs, that can be executed by the processing component 302. The application programs stored in memory 301 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 302 is configured to execute instructions to perform the high-proportion renewable energy power system operation mode recognition method of any of the above embodiments.

[0160] The computer device 300 may also include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate on an operating system stored in memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0161] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0162] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. In this document, "a," "an," "the," "the," and "its" may also include plural forms unless the context clearly indicates otherwise. "Multiple" refers to at least two, such as 2, 3, 5, or 8, etc. "And / or" includes any and all combinations of the related listed items.

[0163] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0164] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying the operation mode of a high-proportion renewable energy power system, characterized in that, The method includes: Obtain time-series operation simulation data of high-proportion renewable energy power systems; Based on the time-series simulation data, several characteristic indicators of the daily operation scenario of the high-proportion new energy power system are determined; Cluster analysis is performed on each of the aforementioned feature indicators to obtain the cluster analysis results; Based on the cluster analysis results, and in accordance with the pre-constructed typical operation mode screening rules, the typical operation modes of the high-proportion new energy power system are determined. The typical operation mode screening rules are used to screen typical operation modes that are representative and have risk disclosure capabilities. The typical operation modes are used to represent the key operational behavior characteristics of the high-proportion new energy power system under different operating scenarios.

2. The method for identifying the operation mode of a high-proportion renewable energy power system according to claim 1, characterized in that, The steps for obtaining time-series operation simulation data of a high-proportion renewable energy power system include: The basic data of the high-proportion new energy power system is obtained, and the basic data is input into the power system simulation software for simulation to obtain the time-series operation simulation data.

3. The method for identifying the operation mode of a high-proportion renewable energy power system according to claim 1, characterized in that, The time-series operation simulation data includes all time periods of the day, daily peak load periods, daily off-peak load periods, wind power output, photovoltaic power output, and power system load. The characteristic indicators include minimum wind power output during peak load periods, minimum photovoltaic power output during peak load periods, maximum wind power output during off-peak load periods, average photovoltaic power output throughout the day, daily wind power generation ratio, and daily photovoltaic power generation ratio. The step of determining multiple characteristic indicators of the daily operation scenario of the high-proportion renewable energy power system based on the time-series simulation data includes: Based on the wind power output and the daily peak load period, determine the minimum wind power output during the peak load period, and based on the photovoltaic power output and the daily peak load period, determine the minimum photovoltaic output during the peak load period. Based on the wind power output and the daily off-peak load period, determine the maximum wind power output during the off-peak load period; The average daily photovoltaic output is determined based on the photovoltaic power generation output and all time periods of the day; The daily wind power generation ratio is determined based on the wind power output, the power system load, and all time periods of the day; and the daily photovoltaic power generation ratio is determined based on the photovoltaic power output, the power system load, and all time periods of the day.

4. The method for identifying the operation mode of a high-proportion renewable energy power system according to claim 1, characterized in that, The step of performing cluster analysis on each of the aforementioned feature indicators to obtain the cluster analysis results includes: Determine the optimal number of clusters, and based on the optimal number of clusters, use the K-means clustering algorithm to perform cluster analysis on each of the aforementioned feature indicators to obtain the cluster analysis results.

5. The method for identifying the operation mode of a high-proportion renewable energy power system according to claim 4, characterized in that, The step of determining the optimal number of clusters includes: Determine multiple candidate cluster numbers and calculate the silhouette coefficient corresponding to each candidate cluster number; The candidate cluster number corresponding to the silhouette coefficient with the largest value is taken as the optimal cluster number.

6. The method for identifying the operation mode of a high-proportion renewable energy power system according to claim 1, characterized in that, The typical operating modes include a high-penetration renewable energy operating mode, a conventional power generation low-operation mode, and a high short-circuit current risk operating mode. The high-penetration renewable energy operating mode is used to describe the operating characteristics of power supply and demand fluctuations and voltage and frequency regulation problems caused by a high proportion of renewable energy access. The conventional power generation low-operation mode is used to describe the operating characteristics of low conventional unit operating capacity, which leads to insufficient peak shaving and reserve capacity. The high short-circuit current risk operating mode is used to describe the operating characteristics of conventional unit capacity being much greater than the load, resulting in increased short-circuit current levels and increased risks of equipment damage and system instability.

7. The method for identifying the operation mode of a high-proportion renewable energy power system according to claim 6, characterized in that, The selection criteria for the high-penetration new energy operation mode are as follows: The selection criteria for the conventional power supply's low-power operation mode are as follows: The screening criteria for the high short-circuit current risk operating mode are as follows: in, The screening criteria represent the high-penetration new energy operation mode described above; Indicates daily operation scenario, This represents the wind power output at time t. This represents the photovoltaic power output at time t. This represents the power system load at time t. The screening criteria represent the typical power supply's low-power-on operating mode. The power supply's on-time capacity is represented by time t. The screening criteria represent the high short-circuit current risk operating mode.

8. A device for identifying the operation mode of a high-proportion renewable energy power system, characterized in that, The device includes: The time-series operation simulation data acquisition module is used to acquire time-series operation simulation data of high-proportion renewable energy power systems; The feature index determination module is used to determine multiple feature indicators of the daily operation scenario of the high-proportion new energy power system based on the time-series operation simulation data. The cluster analysis result determination module is used to perform cluster analysis on each of the aforementioned feature indicators to obtain the cluster analysis results; The typical operation mode determination module is used to determine the typical operation mode of the high-proportion new energy power system according to the cluster analysis results and the pre-constructed typical operation mode screening rules. The typical operation mode screening rules are used to screen typical operation modes with representativeness and risk disclosure capabilities. The typical operation mode is used to represent the key operational behavior characteristics of the high-proportion new energy power system under different operating scenarios.

9. A storage medium, characterized in that: The storage medium stores computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the high-proportion renewable energy power system operation mode identification method as described in any one of claims 1 to 7.

10. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the high-proportion renewable energy power system operation mode identification method as described in any one of claims 1 to 7.

Images