Panoramic time sequence simulation method, device and equipment for power optimization planning and storage medium

Abstract

The invention particularly relates to a power optimization planning panoramic time sequence simulation method and device, equipment and a storage medium, and the method comprises the steps: obtaining the multi-time-scale operation characteristics of a power system and time sequence data in a preset duration; determining the optimal clustering number of each month based on a preset contour coefficient method, performing clustering dimension reduction according to a preset natural time sequence to obtain a target scene day set, forming a reconstruction time sequence through the sequence, and coupling a preset power system panoramic time sequence operation simulation model and a multi-time scale energy storage operation model to obtain a multi-time scale energy storage operation simulation model. The method comprises the steps of obtaining a power optimization planning panoramic time sequence simulation model based on a time sequence, modeling intraday balanced short-time energy storage and annual balanced seasonal energy storage to obtain a simulation state of energy storage in a target scene day, and restoring according to a mapping relation of a reconstructed time sequence to obtain an energy storage simulation state of an annual fluctuation time sequence. Therefore, the problem that the calculation complexity of annual operation simulation in a traditional power system is high is solved, and the optimization efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of power system optimization, and in particular to a method, device, equipment and storage medium for panoramic time series simulation of power optimization planning. Background Art

[0002] As the main force of the energy system, the low-carbon transformation of the power system is the key to achieving a low-carbon society. With the continuous increase in the penetration rate of renewable energy, the operation of the power system faces many uncertain challenges on the source-load side. Figure 1 As shown, Figure 1 The curve of the proportion of monthly load / wind power of a certain power grid to the total annual load in the relevant technology, the intermittent and seasonal fluctuations of renewable energy such as wind and solar power, and the "double peak" characteristics of load demand in winter and summer make the seasonal supply and demand imbalance of the power system increasingly prominent. The emergence of long-term static and stable environmental conditions further exacerbates the problem of sufficient power supply in the system, which may lead to power safety accidents such as load shedding.

[0003] In the related art, the operation simulation is usually performed based on a typical day (i.e., the target scenario day), mainly considering the power balance within the day or week; Figure 2 As shown, Figure 2 This is a schematic diagram of the operation simulation model based on the 8760 time series in the relevant technology. By simulating the system operation status hour by hour, the uncertainty of renewable energy and load is accurately described. In addition, seasonal energy storage, as a seasonal flexibility resource, has gradually attracted attention. Through technologies such as hydrogen electrolysis and electric-thermal conversion, electrical energy can be converted into other forms of energy (such as hydrogen, heat, etc.) and stored across seasons, thereby smoothing the seasonal imbalance of the system.

[0004] However, although the 8760-hour time series operation simulation throughout the year can accurately describe the seasonal fluctuation characteristics of the system, its computational complexity is extremely high and difficult to solve efficiently. Although the traditional typical day simulation method has high computational efficiency, it cannot model the cross-seasonal energy balance of seasonal energy storage; related technical research focuses on short-term energy storage operation constraints (such as pumped storage and electrochemical energy storage), and the research on long-term scale and annual energy balance of seasonal energy storage participating in system operation simulation is relatively scarce and needs to be solved urgently. Summary of the invention

[0005] The present application provides a panoramic time series simulation method, device, equipment and storage medium for power optimization planning to solve the problems of high complexity of simulation calculations for year-round operation in traditional power systems, thereby improving the efficiency of power system optimization planning.

[0006] The first embodiment of the present application provides a method for simulating a panoramic time series of power optimization planning, comprising the following steps:

[0007] Obtain the multi-time scale operation characteristics of the power system and the time series data within a preset duration.

[0008] Based on the preset silhouette coefficient method, determine the optimal number of clusters for each month according to the time series data, and based on the optimal number of clusters, perform clustering and dimensionality reduction according to the preset natural time series to obtain a set of target scenario days, and form a reconstructed time series through the sequence reconstruction of the target scenario days.

[0009] Based on the reconstructed time series, couple the preset panoramic time series operation simulation model of the power system and the preset multi-time scale energy storage operation model to obtain a panoramic time series simulation model of power optimization planning based on time series, and based on the time series-based operation simulation model, model the short-term energy storage for intra-day balance and seasonal energy storage for annual balance to obtain the simulation state of the energy storage on the target scenario days, and restore the simulation state of the energy storage on the target scenario days according to the mapping relationship of the new reconstructed time series to obtain the simulation state of the energy storage of the annual fluctuation time series.

[0010] Optionally, the time series data includes renewable energy and load data. After obtaining the multi-time scale operation characteristics of the power system and the time series data within a preset duration, it further includes:

[0011] Based on the renewable energy and the load data, construct the preset panoramic time series operation simulation model of the power system and the preset multi-time scale energy storage operation model.

[0012] Optionally, the determining the optimal number of clusters for each month according to the preset silhouette coefficient method based on the time series data includes:

[0013] Based on the time series data, calculate the average within-cluster dissimilarity and the minimum between-cluster dissimilarity of the clusters, and obtain the silhouette coefficient for each month according to the average within-cluster dissimilarity and the minimum between-cluster dissimilarity.

[0014] Based on the preset silhouette coefficient method, calculate the optimal number of clusters for each month according to the silhouette coefficient of each month.

[0015] Optionally, after calculating the optimal number of clusters for each month according to the silhouette coefficient of each month, it further includes:

[0016] Evaluate the clustering compactness index for each month according to the optimal number of clusters for each month.

[0017] Based on the evaluation results, take the month with the lowest clustering compactness index among all the clustering compactness indices as the reference month, and calculate the clustering scaling coefficient for each month based on the reference month month by month.

[0018] Adjust the optimal number of clusters for each month according to the clustering scaling coefficient for each month.

[0019] Optionally, the preset silhouette coefficient method is as follows:

[0020]

[0021]

[0022] k = x(d);

[0023] where N is the optimal number of clusters vector matrix for each month, is the optimal number of clusters for the m-th month, M is the total number of months, is the number of clusters for each month calculated based on the silhouette coefficient method, |D| is the total number of days, N m is the optimization variable of the number of clusters for the m-th month, Θ m,d is the time series vector for the d-th day of the m-th month, is the clustering compactness index of the clustering result within the m-th month, Ω m,i is the i-th clustering set for the m-th month, is the time series vector corresponding to the target scenario day for the d-th day of the m-th month, κ m is the clustering scaling coefficient for the m-th month, d is the d-th natural day within the month, and k is the k-th target scenario day within the month.

[0024] An embodiment of the second aspect of the present application provides a power optimization planning panoramic time series simulation device, including:

[0025] An acquisition module, configured to acquire the multi-time scale operation characteristics of the power system and the time series data within a preset duration;

[0026] A clustering module, configured to determine the optimal number of clusters for each month based on the preset silhouette coefficient method according to the time series data, and perform clustering dimensionality reduction according to the preset natural time series based on the optimal number of clusters to obtain a target scenario day set, and form a new reconstructed time series through sequence reconstruction of the target scenario days;

[0027] A modeling module, configured to couple the preset power system panoramic time series operation simulation model and the preset multi-time scale energy storage operation model based on the new reconstructed time series to obtain a power optimization planning panoramic time series simulation model based on the time series, and perform modeling on the short-term energy storage for intra-day balance and the seasonal energy storage for annual balance based on the time series-based operation simulation model to obtain the simulated state of the energy storage on the target scenario days, and restore the simulated state of the energy storage on the target scenario days according to the mapping relationship of the new reconstructed time series to obtain the simulated state of the energy storage of the annual fluctuation time series.

[0028] Optionally, the time series data includes renewable energy and load data, and the obtaining module is further configured to:

[0029] Based on the renewable energy and the load data, construct the preset panoramic time series operation simulation model of the power system and the preset multi-time scale energy storage operation model.

[0030] Optionally, the clustering module is specifically configured to:

[0031] Based on the time series data, calculate the mean of the within-cluster dissimilarity and the minimum of the between-cluster dissimilarity of the clustering, and obtain the silhouette coefficient of each month according to the mean of the within-cluster dissimilarity and the minimum of the between-cluster dissimilarity;

[0032] Based on the preset silhouette coefficient method, calculate the optimal number of clusters for each month according to the silhouette coefficient of each month.

[0033] Optionally, the clustering module is further configured to:

[0034] Evaluate the clustering compactness index of each month according to the optimal number of clusters of each month;

[0035] Based on the evaluation result, take the month with the lowest clustering compactness index among all the clustering compactness indexes as the reference month, and calculate the clustering scaling coefficient of each month month by month based on the reference month;

[0036] Adjust the optimal number of clusters of each month according to the clustering scaling coefficient of each month.

[0037] Optionally, the preset silhouette coefficient method is:

[0038]

[0039] k = x(d);

[0040] Where N is the optimal number of clusters vector matrix for each month, is the optimal number of clusters in the m-th month, M is the total number of months, is the number of clusters for each month calculated based on the silhouette coefficient method, |D| is the total number of days, N m is the optimization variable of the number of clusters in the m-th month, Θ m,d is the time series vector on the d-th day of the m-th month, is the clustering compactness index of the clustering result within the m-th month, Ω m,i is the i-th clustering set in the m-th month, is the time series vector of the target scenario day corresponding to the d-th day of the m-th month, κ mis the clustering scaling coefficient for the m-th month, d is the d-th natural day within a month, and k is the k-th target scenario day within a month.

[0041] An embodiment of the third aspect of the present application provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are configured to execute the power optimization planning panoramic time series simulation method as described in the above embodiments.

[0042] An embodiment of the fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored, and the program is executed by a processor to implement the power optimization planning panoramic time series simulation method as described in the above embodiments.

[0043] An embodiment of the fifth aspect of the present application provides a computer program product, the computer program product stores a computer program, and when the program is executed by a processor, it implements the power optimization planning panoramic time series simulation method as described in the above embodiments.

[0044] Thus, the multi-time scale operation characteristics of the power system and the time series data within a preset duration are obtained; the optimal clustering number for each month is determined based on the preset silhouette coefficient method, the target scenario day set is obtained by clustering and dimensionality reduction according to the preset natural time series, and a reconstructed time series is formed through its sequence. The preset power system panoramic time series operation simulation model and the multi-time scale energy storage operation model are coupled to obtain a power optimization planning panoramic time series simulation model based on the time series. The short-term energy storage for intra-day balance and the seasonal energy storage for annual balance are modeled to obtain the simulated state of the energy storage on the target scenario days, and the simulated state of the energy storage for the annual fluctuation time series is restored according to the mapping relationship of the reconstructed time series. Thus, the problems such as high computational complexity of the annual operation simulation in the traditional power system are solved, and the efficiency of the power system optimization planning is improved.

[0045] Additional aspects and advantages of the present application will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present application. Description of the Drawings

[0046] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description of the embodiments in conjunction with the drawings, where:

[0047] Figure 1 is a schematic diagram of the curve of the monthly electricity consumption of a certain power grid load / wind power in the related art accounting for the total annual load electricity consumption;

[0048] Figure 2 is a schematic diagram of the operation simulation model based on the 8760 time series in the related art;

[0049] Figure 3 It is a flowchart of a panoramic time - series simulation method for power optimization planning provided according to an embodiment of the present application;

[0050] Figure 4 It is a schematic diagram of the operation simulation of energy storage at different time scales based on the reconstructed time series for a panoramic time - series simulation method for power optimization planning provided according to an embodiment of the present application;

[0051] Figure 5 It is a schematic diagram of the adaptive time - series clustering dimensionality reduction and reconstruction method for a panoramic time - series simulation method for power optimization planning provided according to an embodiment of the present application;

[0052] Figure 6 It is a flowchart of the internal software calculation for a panoramic time - series simulation method for power optimization planning provided according to an embodiment of the present application;

[0053] Figure 7 It is a schematic diagram of the topology of the power system in northwest China for a panoramic time - series simulation method for power optimization planning provided according to an embodiment of the present application;

[0054] Figure 8 It is a schematic diagram of the optimal number of clusters per month for a panoramic time - series simulation method for power optimization planning provided according to an embodiment of the present application;

[0055] Figure 9 It is a schematic diagram of the balance of short - term energy storage and seasonal energy storage at different time scales for a panoramic time - series simulation method for power optimization planning provided according to an embodiment of the present application;

[0056] Figure 10 It is a schematic diagram of a panoramic time - series simulation device for power optimization planning provided according to an embodiment of the present application;

[0057] Figure 11 It is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. Detailed implementation manners

[0058] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application, and should not be construed as a limitation to the present application.

[0059] The power optimization planning panoramic time - series simulation method, device, equipment, and storage medium according to the embodiments of the present application will be described below with reference to the accompanying drawings. In view of the problem of high computational complexity in the annual operation simulation of traditional power systems mentioned in the above - mentioned background technology, the present application provides a power optimization planning panoramic time - series simulation method. In this method, the multi - time - scale operation characteristics of the power system and the time - series data within a preset duration are obtained; based on the preset silhouette coefficient method, the optimal number of clusters for each month is determined, and the target scenario - day set is obtained by clustering and dimension reduction according to the preset natural time - series, and a reconstructed time - series is formed through its sequence. The preset power system panoramic time - series operation simulation model and the multi - time - scale energy - storage operation model are coupled to obtain a power optimization planning panoramic time - series simulation model based on the time - series. The short - term energy storage for intra - day balance and the seasonal energy storage for annual balance are modeled to obtain the simulated state of the energy storage on the target scenario - day, and the simulated state of the energy storage for the annual fluctuation time - series is restored according to the mapping relationship of the reconstructed time - series. Thus, problems such as high computational complexity in the annual operation simulation of traditional power systems are solved, and the efficiency of power system optimization planning is improved.

[0060] Specifically, Figure 3 FIG. is a schematic flowchart of a power optimization planning panoramic time - series simulation method provided by an embodiment of the present application.

[0061] As Figure 3 shown, the power optimization planning panoramic time - series simulation method includes the following steps:

[0062] In step S301, the multi - time - scale operation characteristics of the power system and the time - series data within a preset duration are obtained.

[0063] Optionally, in some embodiments, the time - series data includes renewable energy and load data. After obtaining the multi - time - scale operation characteristics of the power system and the time - series data within a preset duration, it further includes: based on the renewable energy and load data, constructing a preset power system panoramic time - series operation simulation model and a preset multi - time - scale energy - storage operation model.

[0064] Among them, the multi - time - scale operation characteristics are to observe and analyze the operation modes and characteristics of the power system at different time spans; the time - series data within a preset duration is a series of measurement values collected at fixed time intervals within a specific time period.

[0065] Specifically, collect time-series data related to the power system within a preset time duration, including but not limited to renewable energy and load data for each hour of the 8,760 hours in a year, including the power generation of traditional energy sources (such as thermal power and hydropower) and renewable energy sources (such as wind energy and solar energy), meteorological data, market data, etc., to ensure that the collected data has sufficient resolution to capture the required multi-time scale characteristics (e.g., hourly, daily, monthly, or annual data); clean and preprocess the collected data, fill in or delete missing values, identify and correct outliers, and standardize / normalize the data for subsequent analysis; based on the preprocessed data, extract multi-time scale operation characteristics, and store the extracted characteristics and the original time-series data for subsequent analysis and model training.

[0066] In step S302, based on the preset silhouette coefficient method, determine the optimal number of clusters for each month according to the time-series data, and based on the optimal number of clusters, perform clustering and dimensionality reduction according to the preset natural time series to obtain a set of target scenario days, and form a reconstructed time series through the sequence reconstruction of the target scenario days.

[0067] Among them, the preset silhouette coefficient method is a calculation method based on the target silhouette coefficient, which is a modified silhouette coefficient method obtained by considering time correlation and adjusting the dissimilarity calculation method to optimize the cluster number selection criterion; the preset natural time series refers to a data series arranged in the natural order of time.

[0068] Specifically, based on the time-series data of 8,760 hours in a year, use the preset silhouette coefficient method to calculate the number of clusters for each month, and determine the optimal number of clusters for each month according to the change trend of the silhouette coefficient; usually, as the number of clusters increases, the silhouette coefficient will first rise and then fall, and the optimal number of clusters is usually a value where the silhouette coefficient is the largest or near the peak; perform K-means clustering on the time-series data of each month based on the optimal number of clusters to obtain a set of target scenario days for each month; convert the original time series data into a simplified representation form composed of these target scenario days; based on the selected target scenario days, reconstruct a new reconstructed time series in the natural time order, and use the reconstructed time series for further analysis or simulation.

[0069] Optionally, in some embodiments, determining the optimal number of clusters for each month based on the preset silhouette coefficient method includes: based on the time-series data, calculate the mean within-cluster dissimilarity and the minimum between-cluster dissimilarity of the clusters, and obtain the silhouette coefficient for each month according to the mean within-cluster dissimilarity and the minimum between-cluster dissimilarity; based on the preset silhouette coefficient method, calculate the optimal number of clusters for each month according to the silhouette coefficient for each month.

[0070] Optionally, in some embodiments, the preset silhouette coefficient method is:

[0071]

[0072] k = zx(d);

[0073] Where N is the vector matrix of the optimal number of clusters per month, is the optimal number of clusters in the m-th month, M is the total number of months, is the number of clusters per month calculated based on the silhouette coefficient method, |D| is the total number of days, N m is the optimization variable of the number of clusters in the m-th month, Θ m,d is the time series vector on the d-th day of the m-th month, is the clustering compactness index of the clustering result within the m-th month, Ω m,i is the i-th clustering set in the m-th month, is the time series vector corresponding to the target scenario day on the d-th day of the m-th month, κ m is the clustering scaling coefficient in the m-th month, d is the d-th natural day within the month, and k is the k-th target scenario day within the month.

[0074] It can be understood that by calculating the average within-cluster dissimilarity and the minimum between-cluster dissimilarity, the silhouette coefficient of the time series vector is solved, and finally, the average of the silhouette coefficients of all vector points in the space is taken to obtain the silhouette coefficient for each month based on the current clustering result; the following is the calculation formula for the target silhouette coefficient:

[0075]

[0076] Where a(Θ d ) is the average within-cluster dissimilarity, Θ j is the time series vector in the same clustering cluster Ω d as Θ x , b(Θ d ) is the minimum between-cluster dissimilarity, where Θ q is the time series vector in a different clustering cluster Ω d from Θ y ; sl(Θ d ) is the silhouette coefficient of the vector Θ d .

[0077] It should be noted that when the silhouette coefficient reaches the maximum value or shows a significant plateau, the corresponding number of clusters is the optimal number of clusters for that month. If multiple numbers of clusters result in similar maximum silhouette coefficients, a suitable number of clusters can be selected as the optimal number of clusters according to actual requirements (such as computational complexity, model interpretability, etc.).

[0078] In step S303, based on the reconstructed time series, a preset panoramic time-series operation simulation model of the power system and a preset multi-time-scale energy storage operation model are coupled to obtain a panoramic time-series simulation model for power optimization planning based on the time series. Based on the operation simulation model based on the time series, short-term energy storage for intraday balance and seasonal energy storage for annual balance are modeled to obtain the simulated state of the energy storage on the target scenario day. The simulated state of the energy storage on the target scenario day is restored according to the mapping relationship of the new reconstructed time series to obtain the simulated state of the energy storage for the annual fluctuation time series.

[0079] Specifically, as Figure 4 shown, Figure 4 FIG. is a schematic diagram of the operation simulation of energy storage with different time scales based on the reconstructed time series for a power optimization planning panoramic time-series simulation method provided by an embodiment of the present application. The preset panoramic time-series operation simulation model of the power system models the SOC of the energy storage throughout the year; the preset multi-time-scale energy storage operation model divides the SOC of the energy storage into an intraday component and a daily component, respectively establishes the energy conversion process and constraint conditions, and models the set of target scenario days. The charging and discharging behaviors of the seasonal energy storage are the same within the same target scenario day, and there is an accumulative relationship in the storage amount during the day; the fluctuation curves of the SOC of the short-term energy storage are the same within the same target scenario day, and the intraday energy balance is ensured; the hourly energy state conversion process of the intraday component of the energy storage is modeled, the initial value of the intraday component of the SOC of the energy storage is set to 0, and the upper and lower limit constraint conditions of the charging and discharging power of the energy storage during the day and the upper and lower limit constraint conditions of the intraday component of the SOC of the energy storage are set. As Figure 5 shown, Figure 5 FIG. is a schematic diagram of the adaptive time series clustering dimensionality reduction and reconstruction method for a power optimization planning panoramic time-series simulation method provided by an embodiment of the present application. The mapping relationship of the reconstructed time series means that in the clustering dimensionality reduction, each day will be assigned to a target scenario day, and there is a mapping relationship from the original time series to the set of target scenario days; for each day in the original time series, corresponding to the target scenario day to which it belongs, the simulated state of the energy storage on the target scenario day is used as the simulated state of the energy storage on this day.

[0080] Therefore, this software uses the preset silhouette coefficient method to adaptively calculate the optimal number of clusters for each month. Using different numbers of clusters each month can effectively reflect the seasonal fluctuations in the system's monthly electricity consumption within the year, and can also reflect the differences in the discrete distribution of time series data within each month. Through the clustering selection of typical days (i.e., target scenario days) each month, the extraction and modeling of multi-time scale operating characteristics of the power system can be finally realized. This application reconstructs the target scenario day time series based on the target scenario day data obtained by clustering, which can effectively reduce the scale of the optimization problem. Based on the reconstructed time series, a panoramic time series operation simulation of the power system is performed, while considering the system operation constraints within and between days. By using a small number of decision variables, the seasonal fluctuation characteristics of the power system throughout the year can be accurately portrayed in the optimization model. The embodiment of the present application divides the energy storage SOC into an intraday component and an inter-day component, and establishes the energy conversion process of the intraday component of the energy storage and the net incremental constraint of the intraday component after the energy storage is simulated through intra-day operation according to the characteristics of the intraday component and the inter-day component respectively; by modeling the intraday component and the inter-day component, modeling is performed for the short-term energy storage with intra-day balance and the seasonal energy storage operation state with annual balance.

[0081] Optionally, in some embodiments, after calculating the optimal number of clusters for each month based on the silhouette coefficient of each month, it also includes: evaluating the cluster compactness index of each month based on the optimal number of clusters for each month; based on the evaluation result, taking the month with the lowest cluster compactness index as the base month, and calculating the cluster scaling coefficient of each month based on the base month; adjusting the optimal number of clusters for each month according to the cluster scaling coefficient of each month.

[0082] Among them, the cluster compactness index is the ratio or difference between the intra-cluster distance and the inter-cluster distance, which is used to measure the compactness within a clustering result and the degree of separation from other clusters.

[0083] Therefore, for each month's data and its corresponding optimal number of clusters, K-means clustering is performed to calculate the mean intra-cluster dissimilarity and the minimum inter-cluster dissimilarity. The month with the lowest cluster compactness index is found from all months as the base month. The lower the cluster compactness index, the higher the similarity between the points in the cluster, and the greater the difference between clusters, the better the clustering effect. Based on the base month, the cluster scaling coefficient is calculated for other months month by month to help adjust the optimal number of clusters in other months to make it closer to the clustering effect of the base month. The cluster scaling coefficient is defined by comparing the cluster compactness index of each month with the base month; according to the calculated cluster scaling coefficient, the optimal number of clusters for each month is appropriately adjusted.

[0084] In order to facilitate the understanding of the technical personnel in the field of this application, the software part of this application will be described in detail below. Figure 6 As shown, Figure 6This is the internal software calculation flowchart of an embodiment of the present application. The internal software calculation mainly includes several processes such as data reading, simulation calculation, result statistics, and result output. In the actual execution process, the embodiment of the present application reads input data from the file GTEP_in.csv, verifies the read data to ensure the integrity and accuracy of the data; performs temporal aggregation on the original data to obtain a typical day (i.e., the target scenario day) curve and a mapping matrix, and adds corresponding constraint conditions to the model according to the year to be optimized, which may include power grid operation rules, energy storage limitations, etc.; based on the extracted typical day information, a temporal acceleration optimization model is established, and this model is used to simulate and optimize the operating state of the power system, check whether constraints have been added to all years to be optimized, if not, continue to process the next optimization year, if the constraints for the current year have been added, then find the next year to be optimized, and repeat the above steps; use an optimization algorithm to solve the established temporal acceleration optimization model to obtain the optimal solution; output the obtained result to the file GTEP_out.csv; through this process, panoramic temporal simulation of power system optimization planning can be effectively carried out, and detailed optimization results can be output.

[0085] It should be noted that the software of the embodiment of the present application supports a maximum of 300 nodes, a maximum of 500 units, and a maximum of 150 lines. It supports temporal aggregation operation simulation with a minimum time of one year and a maximum time of 10 years. The software running time (optimal hardware configuration) is that the operation simulation time of the power system for one year does not exceed 2 hours; at the same time, the system running time and the required memory increase with the increase in the number of units, the number of load nodes, and the length of the simulation time series.

[0086] Furthermore, the software of the embodiment of the present application is an.exe format executable file, and the input and output data interact with the outside through the.csv file format. The input and output include the following files, as shown in Table 1:

[0087] Table 1

[0088] Type Name File Format Remarks Input File gtep_in.csv CSV Comma-Separated File Calculation Input Data File Output File gtep_out.csv CSV Comma-Separated File System Calculation Overall Result Information

[0089] Among them, in the xls form of the input file, information is input in rows. The input information for each row contains at most 30 dots, in the form of 3 + X (except for system debugging information), that is, it includes 3 necessary basic information points and at most 27 data contents. The sequence of the 3 necessary basic information points is as shown in Table 2:

[0090] Table 2

[0091]

[0092] The input data types represented by different data type labels are shown in Table 3 as follows:

[0093] Table 3

[0094]

[0095]

[0096] The descriptions of the output information in each column are shown in Table 4 as follows:

[0097] Table 4

[0098]

[0099] Thus, in the embodiments of this application, through the program call method, it illustrates how to call the executable file (gtep.exe) of the software and introduces the setting of the storage paths of the input and output files; through the input data format, it details the data structure of the input file gtep_in.csv, including the input content, column labels, descriptions, units, and value ranges of different data types, etc.; through the output data format, it details the data structure of the output file gtep_out.csv, including the column numbers, names, descriptions, etc. of the output data; through the data type label table, it lists the input content corresponding to different data type labels to help users correctly identify and input different types of data. The embodiments of this application ensure that users can correctly input data and parse the output results through the program call method and the description of the input and output data formats.

[0100] Next, the power optimization planning panoramic time series simulation method will be described in detail in combination with a specific embodiment of this application.

[0101] As Figure 7 shown Figure 7 is the schematic diagram of the topology of the Northwest Power System of China for a power optimization planning panoramic time series simulation method provided by the embodiments of this application. The Northwest Power System of China is based on the Northwest Power Grid. In the embodiments of this application, the Northwest Power Grid is divided into five regions according to provincial divisions, and the regions are connected by transmission lines. On the basis of the original system, electrochemical energy storage and seasonal energy storage are respectively considered to be installed at different nodes in the power grid. The key parameters and information in the system are shown in Table 5 as follows:

[0102] Table 5

[0103]

[0104] The load and resource endowment data of each region are shown in Table 6, including the peak load of each region, the average utilization hours of wind power and photovoltaic power. It can be seen from the figure that the peak load in Region 1 is the highest, reaching 110 GW; the peak load in Region 3 is the smallest, only 27 GW. In terms of the endowment of renewable energy resources, the utilization hours of wind power resources are the highest in Region 1, reaching 2,645 hours, and the utilization hours of wind power in Region 3 are the lowest, only 1,862 hours. For photovoltaic resources, the photovoltaic resources in Region 4 are the richest, reaching 1,752 hours. On the contrary, the utilization hours of photovoltaic power in Region 3 are the lowest, only 1,511 hours.

[0105] Table 6

[0106]

[0107] Organize the above example data into a corresponding csv file according to the input format of this software, and call this software to perform cloud energy storage operation simulation, and the corresponding output results can be obtained. As Figure 8 shown, Figure 8 is a schematic diagram of the optimal clustering number for each month of a panoramic time-series simulation method for power optimization planning provided by an embodiment of the present application. Among them, the clustering number is the highest on the typical day in June, reaching 5 days. This reflects that the fluctuation characteristics of the wind and light output curves are the largest in June, and more clustering days are required to describe their randomness.

[0108] Figure 9 (a) is a schematic diagram of the balance of short-term energy storage at different time scales of a panoramic time-series simulation method for power optimization planning provided by an embodiment of the present application. The short-term energy storage SOC maintains energy balance within a day, showing the characteristics of "one charge and two discharges" in the selected typical days. Among them, the red column represents energy storage charging, and the purple column represents energy storage discharging. The system's daily power peak shaving demand is realized through the daily charge and discharge of short-term energy storage; Figure 9 (b) is a schematic diagram of the balance of seasonal energy storage at different time scales of a panoramic time-series simulation method for power optimization planning provided by an embodiment of the present application. The seasonal energy storage shows an overall characteristic of "two charges and two discharges" throughout the year. There are two rounds of continuous charging behaviors in spring (March - May) and autumn and winter (September - December), while energy storage discharging begins to occur in the two peak power load seasons of summer (July - August) and winter (January - February). The charging and discharging of energy storage show obvious seasonal fluctuation rules.

[0109] Therefore, the embodiment of the present application can adaptively calculate the optimal clustering number for each month, reflect the random fluctuation characteristics of different months, effectively model the operation status of short-term energy storage and seasonal energy storage, accurately depict the balance characteristics of energy storage at different time scales, and the software operation results verify its practicability and reliability in power system optimization planning, providing strong support for the planning and operation of power systems.

[0110] According to the power optimization planning panoramic time series simulation method proposed in the embodiments of the present application, the multi-time scale operation characteristics of the power system and the time series data within a preset duration are obtained; based on the preset silhouette coefficient method, the optimal number of clusters for each month is determined, and the target scenario day set is obtained by clustering and dimensionality reduction according to the preset natural time series, and a reconstructed time series is formed through its sequence. The preset power system panoramic time series operation simulation model and the multi-time scale energy storage operation model are coupled to obtain a power optimization planning panoramic time series simulation model based on the time series. The short-term energy storage for intraday balance and the seasonal energy storage for annual balance are modeled to obtain the simulated state of the energy storage on the target scenario day, and the simulated state of the energy storage for the annual fluctuation time series is restored according to the mapping relationship of the reconstructed time series. Thereby, problems such as high computational complexity in the annual operation simulation of traditional power systems are solved, and the efficiency of power system optimization planning is improved.

[0111] Next, a power optimization planning panoramic time series simulation device proposed in the embodiments of the present application is described with reference to the accompanying drawings.

[0112] Figure 10 It is a block diagram of the power optimization planning panoramic time series simulation device according to the embodiments of the present application.

[0113] As Figure 10 shown, the power optimization planning panoramic time series simulation device 1000 includes: an acquisition module 100, a clustering module 200, and a modeling module 300.

[0114] Among them, the acquisition module 100 is used to acquire the multi-time scale operation characteristics of the power system and the time series data within a preset duration;

[0115] The clustering module 200 is used to determine the optimal number of clusters for each month based on the preset silhouette coefficient method according to the time series data, and based on the optimal number of clusters, perform clustering and dimensionality reduction according to the preset natural time series to obtain a target scenario day set, and form a new reconstructed time series through the reconstruction of the sequence of the target scenario days;

[0116] The modeling module 300 is used to couple the preset power system panoramic time series operation simulation model and the preset multi-time scale energy storage operation model based on the new reconstructed time series to obtain a power optimization planning panoramic time series simulation model based on the time series, and model the short-term energy storage for intraday balance and the seasonal energy storage for annual balance based on the time series-based operation simulation model to obtain the simulated state of the energy storage on the target scenario day, and restore the simulated state of the energy storage for the annual fluctuation time series according to the mapping relationship of the new reconstructed time series.

[0117] Optionally, the time series data includes renewable energy and load data, and the acquisition module 100 is further configured to: based on the renewable energy and load data, construct a preset panoramic time series operation simulation model of the power system and a preset multi-time scale energy storage operation model.

[0118] Optionally, the clustering module 200 is specifically configured to: based on the time series data, calculate the mean of the within-cluster dissimilarity and the minimum of the between-cluster dissimilarity of the clusters, and obtain the silhouette coefficient of each month according to the mean of the within-cluster dissimilarity and the minimum of the between-cluster dissimilarity; based on the preset silhouette coefficient method, calculate the optimal number of clusters for each month according to the silhouette coefficient of each month.

[0119] Optionally, the clustering module 200 is further configured to: evaluate the clustering compactness index of each month according to the optimal number of clusters of each month; based on the evaluation result, use the month with the lowest clustering compactness index among all the clustering compactness indexes as the reference month, and calculate the clustering scaling coefficient of each month month by month based on the reference month; adjust the optimal number of clusters of each month according to the clustering scaling coefficient of each month.

[0120] Optionally, the preset silhouette coefficient method is:

[0121]

[0122] k = χ N (d);

[0123] where N is the optimal number of clusters vector matrix for each month, is the optimal number of clusters in the m-th month, M is the total number of months, is the number of clusters for each month calculated based on the silhouette coefficient method, |D| is the total number of days, N m is the optimization variable of the number of clusters in the m-th month, Θ m,d is the time series vector on the d-th day of the m-th month, is the clustering compactness index of the clustering result within the m-th month, Ω m,i is the i-th clustering set in the m-th month, is the time series vector of the target scenario day corresponding to the d-th day of the m-th month, κ m is the clustering scaling coefficient of the m-th month, d is the d-th natural day within the month, and k is the k-th target scenario day within the month.

[0124] It should be noted that the foregoing explanation of the embodiments of the panoramic time series simulation method for power optimization planning also applies to the panoramic time series simulation device for power optimization planning in this embodiment, and will not be elaborated here.

[0125] The power optimization planning panoramic time-series simulation device proposed according to the embodiments of the present application obtains the multi-time scale operation characteristics of the power system and the time-series data within a preset duration; determines the optimal number of clusters for each month based on the preset silhouette coefficient method, clusters and reduces the dimension according to the preset natural time series to obtain a set of target scenario days, and forms a reconstructed time series through its sequence. Coupling the preset power system panoramic time-series operation simulation model and the multi-time scale energy storage operation model, a power optimization planning panoramic time-series simulation model based on the time series is obtained, modeling the short-term energy storage for intra-day balance and the seasonal energy storage for annual balance, obtaining the simulated state of the energy storage on the target scenario days, and restoring the simulated state of the energy storage of the annual fluctuation time series according to the mapping relationship of the reconstructed time series. Thus, problems such as high computational complexity in the annual operation simulation of traditional power systems are solved, and the efficiency of power system optimization planning is improved.

[0126] Figure 11 The structural schematic diagram of the electronic device provided by the embodiments of the present application. The electronic device may include:

[0127] A memory 1101, a processor 1102, and a computer program stored on the memory 1101 and executable on the processor 1102.

[0128] When the processor 1102 executes the program, it implements the power optimization planning panoramic time-series simulation method provided in the above embodiments.

[0129] Further, the electronic device further includes:

[0130] A communication interface 1103 for communication between the memory 1101 and the processor 1102.

[0131] The memory 1101 is used to store a computer program executable on the processor 1102.

[0132] The memory 1101 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0133] If the memory 1101, the processor 1102, and the communication interface 1103 are implemented independently, the communication interface 1103, the memory 1101, and the processor 1102 can be interconnected via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 11 only a thick line is used in Figure 11 , but it does not mean that there is only one bus or one type of bus.

[0134] Optionally, in a specific implementation, if the memory 1101, the processor 1102, and the communication interface 1103 are integrated on a single chip, the memory 1101, the processor 1102, and the communication interface 1103 can communicate with each other through an internal interface.

[0135] The processor 1102 may be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0136] The embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above power optimization planning panoramic timing simulation method is implemented.

[0137] The embodiments of the present application further provide a computer program product, the computer program product stores a computer program, and when the program is executed by a processor, the above power optimization planning panoramic timing simulation method is implemented.

[0138] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0139] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of this application, the meaning of "N" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0140] Any process or method description shown in a flowchart or described in other ways herein can be understood to represent a module, segment, or portion of code including one or more N executable instructions for implementing a customized logical function or process, and the scope of the preferred embodiments of this application includes additional implementations, where the functions may be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of this application pertain.

[0141] It should be understood that each part of this application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following techniques well known in the art or a combination of them can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0142] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

Claims

1. A panoramic time-sequence simulation method for power optimization planning, characterized in that, The method includes the following steps: Obtain the multi-time scale operation characteristics of the power system and the time series data within a preset duration; Based on the preset silhouette coefficient method, determine the optimal number of clusters for each month according to the time series data, and based on the optimal number of clusters, perform clustering and dimensionality reduction according to the preset natural time series to obtain a set of target scenario days, and form a reconstructed time series through the sequence reconstruction of the target scenario days; Based on the reconstructed time series, couple the preset panoramic time series operation simulation model of the power system and the preset multi-time scale energy storage operation model to obtain a panoramic time series simulation model for power optimization planning based on the time series, and based on the operation simulation model based on the time series, model the short-term energy storage for intra-day balance and the seasonal energy storage for annual balance to obtain the simulated state of the energy storage on the target scenario days, and restore the simulated state of the energy storage on the target scenario days according to the mapping relationship of the new reconstructed time series to obtain the simulated state of the energy storage for the annual fluctuation time series.

2. The method according to claim 1, wherein The time series data includes renewable energy and load data. After obtaining the multi-time scale operation characteristics of the power system and the time series data within a preset duration, it further includes: Based on the renewable energy and the load data, construct the preset panoramic time series operation simulation model of the power system and the preset multi-time scale energy storage operation model.

3. The method according to claim 1, wherein The step of determining the optimal number of clusters for each month according to the preset silhouette coefficient method based on the time series data includes: Based on the time series data, calculate the mean within-cluster dissimilarity and the minimum between-cluster dissimilarity of the clusters, and obtain the silhouette coefficient for each month according to the mean within-cluster dissimilarity and the minimum between-cluster dissimilarity; Based on the preset silhouette coefficient method, calculate the optimal number of clusters for each month according to the silhouette coefficient for each month.

4. The method according to claim 3, wherein After calculating the optimal number of clusters for each month according to the silhouette coefficient for each month, it further includes: Evaluate the clustering compactness index for each month according to the optimal number of clusters for each month; Based on the evaluation results, take the month with the lowest clustering compactness index among all the clustering compactness indices as the reference month, and calculate the clustering scaling coefficient for each month based on the reference month month by month; Adjust the optimal number of clusters for each month according to the clustering scaling coefficient for each month.

5. The method according to claim 3, characterized in that The preset silhouette coefficient method is: k = χ N (d); Among them, N is the optimal clustering number vector matrix for each month, is the optimal clustering number for the m-th month, M is the total number of months, is the clustering number for each month calculated based on the silhouette coefficient method, |D| is the total number of days, N m is the optimization variable of the clustering number for the m-th month, Θ m,d is the time series vector for the d-th day of the m-th month, is the clustering compactness index of the clustering result within the m-th month, Ω m,i is the i-th clustering set for the m-th month, is the time series vector corresponding to the target scenario day for the d-th day of the m-th month, κ m is the clustering scaling coefficient for the m-th month, d is the d-th natural day within the month, and k is the k-th target scenario day within the month.

6. A panoramic time-series simulation device for power optimization planning, characterized in that, It includes: An acquisition module for obtaining the multi-time scale operation characteristics of the power system and the time series data within a preset duration; A clustering module for determining the optimal number of clusters for each month according to the time series data based on the preset silhouette coefficient method, and based on the optimal number of clusters, performing clustering and dimensionality reduction according to the preset natural time series to obtain a set of target scenario days, and forming a new reconstructed time series through the sequence reconstruction of the target scenario days; A modeling module, configured to couple a preset panoramic time-series operation simulation model of a power system and a preset multi-time-scale energy storage operation model based on the new reconstructed time series, so as to obtain a panoramic time-series simulation model of power optimization planning based on the time series, and model short-term energy storage for intraday balance and seasonal energy storage for annual balance based on the operation simulation model based on the time series, so as to obtain the simulated state of the energy storage on the target scenario day, and restore the simulated state of the energy storage on the target scenario day according to the mapping relationship of the new reconstructed time series to obtain the simulated state of the energy storage of the annual fluctuation time series.

7. The device according to claim 6, characterized in that, The time series data includes renewable energy and load data. After the obtaining module, it is further configured to: Based on the renewable energy and the load data, construct the preset panoramic time-series operation simulation model of the power system and the preset multi-time-scale energy storage operation model.

8. An electronic device, characterized in that, It includes: A memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the program to implement the panoramic time-series simulation method for power optimization planning according to any one of claims 1-5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to be used to implement the panoramic time-series simulation method for power optimization planning according to any one of claims 1-5.

10. A computer program product storing a computer program, characterized in that, When the program is executed by the processor, it implements the panoramic time-series simulation method for power optimization planning according to any one of claims 1-5.

Images