Power system flexibility evaluation method and system based on entropy weight method improved clustering and multivariate variational mode decomposition

Abstract

The invention relates to the field of power system performance evaluation, and discloses a power system flexibility evaluation method and system based on entropy weight method improved clustering and multivariate variational mode decomposition, and the method comprises the steps: obtaining an intra-day net load curve, and carrying out the similarity clustering of the net load curve; decomposing the intra-day net load curves of different clusters, dividing the intra-day net load curves into net load component curves under a plurality of fluctuation frequency bands, and dividing an upward climbing set and a downward climbing set to obtain upward and downward flexibility requirements under each time scale; performing unit combination on the intra-day net load curve to obtain operation states of various types of controllable units of the power system, and calculating upward and downward flexibility resources under different fluctuation time scales according to the operation states, output adjustment characteristics and fluctuation periods of the various types of controllable units; and performing matching analysis on the flexibility resource and the flexibility demand under the same time scale, calculating to obtain a flexibility index of the power system, and weighting to form a power system flexibility assessment comprehensive index.

Description

Technical Field

[0001] This invention relates to the field of power system performance evaluation technology, and in particular to a power system flexibility evaluation method and system based on entropy weight method improved clustering and multivariate variational mode decomposition. Background Technology

[0002] In recent years, with the large-scale installation and grid connection of highly volatile and intermittent distributed energy sources such as wind power and photovoltaics, the operation of power systems faces new challenges. Power system flexibility refers to the ability of a system with high penetration of intermittent power sources to respond quickly to predictable and unpredictable fluctuations in supply and demand on both sides, ensuring the reliable operation of the power system. Unlike traditional stable power sources, the output characteristics of these distributed energy sources exhibit significant weather dependence, and their random fluctuations lead to increased volatility in the system's net load curve (the actual load after deducting renewable energy output from total electricity demand). This places more stringent demands on the regulation capabilities of conventional power sources. Against this backdrop, how to scientifically assess the power system's ability to cope with uncertainties on both the supply and demand sides after a high proportion of renewable energy is integrated has become an important research topic in the energy field, driving continuous innovation in power system flexibility assessment systems and methodologies.

[0003] Existing methods for quantitative assessment of power system flexibility can be divided into: (1) deterministic methods and (2) uncertain methods. Deterministic methods typically employ deterministic quantitative indicators or scoring tables, or use the inherent properties of components or systems, or physical quantities that need to be obtained through simulation operation, as flexibility indicators. Uncertainty indicators mainly include interval-type and probabilistic types: interval-type indicators measure system flexibility by the range of variation of flexibility-related physical quantities. The main idea of ​​probabilistic types is to use statistical parameters or their mathematical transformations to describe the probability distribution of power system physical quantities (such as renewable energy output, net load, and flexibility margin) to reflect the characteristics of system flexibility in a certain aspect.

[0004] The existing assessment methods have the following problems: (1) Deterministic indicators ignore the impact of uncertainty and cannot fully reflect the flexibility characteristics of the system; interval indicators are inconvenient in describing the proportion of high distributed energy. Probabilistic indicators use low-order statistics such as mean, variance and quantile to characterize the system flexibility, which can reflect the central position and dispersion of the data, but ignore the high-dimensional characteristics of the distribution such as skewness and kurtosis, and cannot clearly reflect the potential and risk of the system's flexibility adjustment. (2) Most of the above methods do not consider the fluctuation of source load at multiple time scales, or the consideration of multiple time scales is only based on the first difference of the net load curve, which cannot reflect the differences in flexibility demand at different time scales. Correspondingly, it is also difficult to effectively distinguish the differences in the adjustment characteristics of the system's controllable units at different time scales in terms of flexibility resources. The resulting flexibility assessment indicators lack the distinguishability in adjustment rate, which is not conducive to the planning of flexible power sources. Summary of the Invention

[0005] To address the aforementioned problems, the purpose of this invention is to provide a power system flexibility assessment method and system based on entropy weighting improved clustering and multivariate variational mode decomposition, which can improve prediction stability and prediction accuracy.

[0006] To achieve the above objectives, in a first aspect, the technical solution adopted by the present invention is as follows: a power system flexibility assessment method based on entropy weight method-improved clustering and multivariate variational mode decomposition, comprising: acquiring intraday net load curves; performing similarity clustering on the net load curves using an entropy weight method-improved clustering algorithm to simplify the sample data into representative typical patterns; decomposing the intraday net load curves of different clusters based on a multivariate variational mode decomposition algorithm, dividing them into net load component curves under multiple fluctuation frequency bands, and dividing them into upward ramp sets and downward ramp sets through waveform identification to obtain the upward and downward flexibility requirements at each time scale; combining the intraday net load curves of each typical pattern with generating units to obtain the corresponding operating status of each type of controllable generating unit in the power system, and calculating the upward and downward flexibility resources at different fluctuation time scales based on the operating status, output regulation characteristics, and fluctuation cycle of each type of controllable generating unit; performing matching analysis on the flexibility resources and flexibility requirements at the same time scale, calculating the flexibility missing probability, flexibility abundance expectation, and flexibility insufficiency expectation at each time scale as flexibility indicators of the power system, and weighting them to form a comprehensive power system flexibility assessment index.

[0007] Furthermore, a clustering algorithm based on the entropy weight method is used to perform similarity clustering on the net load curves, including: The intraday net load curve is formed by the historical output curves of load, wind power, and photovoltaic power, and the daily average load, daily maximum load, and daily maximum fluctuation difference are selected as load characteristic indicators. The entropy weight method is used to assign weights to different characteristic indicators, and the clustering algorithm is improved by setting an initial weight vector to generate new cluster centers; After generating new cluster centers, the contribution of each characteristic index to the cluster centers is calculated, resulting in a similarity matrix with m evaluation objects and k evaluation indicators. The attraction matrix and belonging matrix are updated based on the similarity matrix and the damping coefficient. It is then determined whether the attraction matrix and belonging matrix have converged. If they have converged, the clustering result is output; otherwise, the attraction matrix and belonging matrix are updated again.

[0008] Furthermore, based on the multivariate variational mode decomposition algorithm, the intraday net load curves of different clusters are decomposed into net load component curves under multiple fluctuation frequency bands, including: Based on the defined IMF set of load data decomposition, determine the input data. ; Input data The input is a constructed multivariate variational mode decomposition model, used to analyze the input data. Construct the optimal constrained variational problem; The constrained variational problem is unconstrained by using the Lagrangian function to solve the variational problem and determine whether the solution converges. If it does not converge, the alternating direction multiplier iterative algorithm is used to solve the unconstrained problem. The IMF set and the center frequency set of the IMF components are iteratively updated to obtain the set number of IMF components to obtain a new IMF set. The newly determined input data is then input into the multivariate variational mode decomposition model. If convergence occurs, the upward and downward flexibility requirements are determined. Combining the characteristics of distributed generation fluctuations and the distribution of power regulation rates, the intraday net load curve is divided into net load component curves in three frequency bands.

[0009] Furthermore, by identifying waveforms and dividing them into upward and downward climbing sets, the upward and downward flexibility requirements at various time scales are obtained, including: The net load component curves of the three frequency bands are divided into two ramp subsets, one upward and one downward, by waveform identification. Among them, the amplitude of the climbing segment in the climbing subset represents the flexibility requirement of the climbing segment, the duration represents the fluctuation period of the climbing segment, and the fluctuation period length under different fluctuation components should correspond to the time scale defined by them.

[0010] Furthermore, the daily net load curves for each typical mode are combined with the corresponding generating units to obtain the operating status of various types of controllable generating units in the power system, including: The short-term flexibility resources of thermal power plants are constrained by their upward and downward ramp rates and maximum and minimum operating output. On a long-term scale, in addition to the combined cycle gas turbine units in operation, flexibility is also achieved by starting and stopping the gas turbine units and adjusting the output of the coal-fired units that have undergone flexibility modifications. Hydropower units serve as peak-shaving power sources, providing corresponding flexibility resources at various time scales throughout the day based on their operating status. Power electronic energy storage devices are flexible power sources that can be quickly adjusted. Among them, supercapacitors and compressed air energy storage devices are suitable for flexible adjustment on time scales of <15 minutes and 15~60 minutes, respectively. The size of controllable load regulation resources is determined by the size of the electricity load of the demand-side response implementation agency and its own maximum output variation limit.

[0011] Furthermore, a matching analysis is conducted on flexibility resources and flexibility requirements at the same time scale. The probability of flexibility deficiency, the expected abundance of flexibility, and the expected insufficiency of flexibility at each time scale are calculated as flexibility indicators of the power system. These are then weighted to form a comprehensive system flexibility assessment index, including: The probability of lack of flexibility, the expected abundance of flexibility, and the expected insufficiency of flexibility are selected as flexibility indicators for power systems. Based on the flexibility index, the flexibility indexes at each scale are weighted and summed to form the scale-weighted flexibility index. The weight of each scale-weighted flexibility index is determined by the average flexibility demand per unit time, so as to reflect the impact of fluctuations on the overall flexibility of the power system.

[0012] Furthermore, the probability of lack of flexibility, the expected abundance of flexibility, and the expected deficiency of flexibility are selected as flexibility indicators for the power system, specifically: Statistical analysis was conducted on the flexibility-deficient ramp-up segment where flexibility resources were lower than flexibility requirements to obtain the probability of flexibility deficiency. Based on the flexibility abundance, resources and demand of a certain ramp segment, as well as the total installed capacity of the power system, set the flexibility abundance expectation; By statistically analyzing the flexibility deficit in the insufficient flexibility ramp-up section, the expected value of the insufficient flexibility sample is calculated to reflect the severity of the flexibility gap in the power system.

[0013] Secondly, the technical solution adopted by this invention is as follows: a power system flexibility assessment system based on entropy weight method-improved clustering and multivariate variational mode decomposition, comprising: a clustering module, which acquires intraday net load curves and performs similarity clustering on the net load curves using an entropy weight method-improved clustering algorithm to simplify the sample data into representative typical patterns; and a flexibility requirement determination module, which decomposes the intraday net load curves of different clusters based on a multivariate variational mode decomposition algorithm, dividing them into net load component curves under multiple fluctuation frequency bands, and dividing them into upward and downward ramp sets through waveform identification to obtain the upward and downward ramp sets at each time scale. The system includes a downward flexibility requirement module; a flexibility resource determination module, which combines units based on the intraday net load curves of various typical modes, obtains the operating status of various types of controllable units in the corresponding power system, and calculates upward and downward flexibility resources under different fluctuation time scales based on the operating status, output regulation characteristics, and fluctuation cycle of each type of controllable unit; and an evaluation module, which performs matching analysis on flexibility resources and flexibility requirements under the same time scale, calculates the probability of flexibility deficiency, expected flexibility abundance, and expected flexibility insufficiency at each time scale as flexibility indicators of the power system, and weights them to form a comprehensive indicator for power system flexibility assessment.

[0014] Thirdly, the technical solution adopted by the present invention is: a computer-readable storage medium for storing one or more programs, wherein the one or more programs include instructions, which, when executed by a computing device, cause the computing device to perform any of the methods described above.

[0015] Fourthly, the technical solution adopted by the present invention is: a computing device comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing any of the methods described above.

[0016] The present invention has the following advantages due to the adoption of the above technical solutions: 1. This invention is accurate. It introduces a multivariate variational mode decomposition (MVMD) data decomposition and reconstruction method into flexibility assessment. By simultaneously decomposing multivariate load flexibility data using MVMD, it effectively reduces the prediction complexity of the original sequence while preserving the coupling characteristics of each modal component. This improves the model's generalization ability at the data level, achieves synchronous decomposition of each load, simplifies the prediction process, and enhances the model's prediction stability and accuracy.

[0017] 2. This invention is objective. This invention uses an AP clustering algorithm based on the entropy weight method to perform similarity clustering of net load curves. The entire process requires no human intervention, thus making it more objective. Traditional clustering algorithms require inputting the number of clusters, leading to subjective interference with the clustering results; some algorithms randomly select cluster centers or search directions, which may cause the clustering results to get stuck in local optima.

[0018] 3. This invention is effective. This invention can effectively distinguish power system flexibility assessment indicators that address flexibility needs across multiple time scales. Existing methods consider multiple time scales based on the first-order difference of the net load curve, failing to reflect the differences in flexibility needs across different time scales. Consequently, they also struggle to effectively differentiate the regulation characteristics of controllable units at different time scales in terms of flexibility resources. This invention overcomes these shortcomings. Attached Figure Description

[0019] Figure 1 This is a flowchart of the power system flexibility assessment method based on entropy weight method improved clustering and multivariate variational mode decomposition in an embodiment of the present invention; Figure 2 This is a clustering flowchart in an embodiment of the present invention; Figure 3 This is a flowchart of the multivariate variational mode decomposition in an embodiment of the present invention; Figure 4 This is a flowchart of the controllable operating state modeling in an embodiment of the present invention. Detailed Implementation

[0020] To address the technical shortcomings of existing assessment methods, this invention provides a power system flexibility assessment method and system based on improved clustering and multivariate variational mode decomposition using the entropy weight method, which can effectively distinguish flexibility demands across multiple time scales. First, a daily net load curve is generated based on historical load, wind power, and photovoltaic output curves. The net load curve is then clustered based on similarity using an improved clustering algorithm using the entropy weight method. Next, a multivariate mode decomposition algorithm is employed to decompose the net load curve across multiple time scales, obtaining a set of time series components that independently reflect the fluctuation characteristics of different frequency bands, reflecting the upward and downward flexibility demands at different time scales. Then, adjustment capacity models for different types of controllable units within the system are established under different adjustment rate ranges, forming multi-time scale flexibility resource adjustment ranges. Through matching analysis of flexibility resources and demands at the same time scale, this invention can easily and intuitively calculate indicators such as the probability of flexibility loss, expected flexibility abundance, and expected flexibility deficiency, and finally, a weighted comprehensive index for system flexibility assessment is formed.

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0022] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0023] In one embodiment of the present invention, a power system flexibility assessment method based on entropy weighting and improved clustering and multivariate variational mode decomposition is provided. In this embodiment, as... Figure 1 As shown, the method includes the following steps: 1) Obtain the intraday net load curve, and perform similarity clustering on the net load curve using a clustering algorithm based on the entropy weight method, simplifying a large amount of sample data into a few representative typical patterns; 2) Based on the multivariate variational mode decomposition algorithm, the intraday net load curves of different clusters are decomposed into net load component curves under multiple fluctuation frequency bands. The upward ramp set and downward ramp set are divided by waveform identification to obtain the upward and downward flexibility requirements at each time scale. 3) Combine the intraday net load curves of each typical mode to obtain the operating status of each type of controllable unit in the power system, and calculate the upward and downward flexibility resources under different fluctuation time scales based on the operating status, output regulation characteristics and fluctuation period of each type of controllable unit. 4) Perform a matching analysis on flexibility resources and flexibility requirements at the same time scale, calculate the probability of flexibility deficiency, expected flexibility abundance, and expected flexibility insufficiency at each time scale as flexibility indicators of the power system, and weight them to form a comprehensive indicator for power system flexibility assessment.

[0024] In step 1) above, if Figure 2 As shown, the net load curves are clustered based on similarity using a clustering algorithm improved by the entropy weight method, including the following steps: 1.1) Selecting Net Load Characteristic Indicators. The output curves of different distributed systems under different scenarios have certain differences, and the output characteristics can reflect their internal output patterns. Therefore, in this embodiment, the intraday net load curve is formed by the historical output curves of load, wind power, and photovoltaic, and the daily average load, daily maximum load, and daily maximum fluctuation difference are selected as load characteristic indicators.

[0025] 1.2) Assign weights to different characteristic indicators. The entropy weight method is used to assign weights to different characteristic indicators. By setting an initial weight vector, the clustering algorithm is improved to generate new cluster centers.

[0026] Specifically, the improved clustering algorithm is as follows: The initial weight vector is set as... , The number of characteristic indicators, This represents the weight of the k-th characteristic indicator.

[0027] 1.3) After generating new cluster centers, calculate the contribution of each characteristic index to the cluster center to obtain a similarity matrix S with m evaluation objects and k evaluation indicators.

[0028] In this embodiment, the contribution of each characteristic index to the cluster center is calculated. : (1) In the formula, The number of cluster centers; Let j be the value of the characteristic index of the j-th cluster center, where j = 1, 2, ..., m; Let the output characteristic index value be any value selected from the j-th class. This forms a matrix with m evaluation objects and k evaluation indices. .

[0029] In this embodiment, Euclidean distance is used to measure contribution, and a smaller value is better. The improved Euclidean distance formula is as follows: (2) In the formula, For users and Euclidean distance formula based on photovoltaic power output characteristic indicators; The number of characteristic indicators; Weighted by contribution level; The photovoltaic output of user p; The photovoltaic output of user q. (Measured by Euclidean distance) Construct a similarity matrix S.

[0030] 1.4) Update the attraction matrix R and the membership matrix A using the similarity matrix S, and update them again based on the damping coefficient. Determine if both the attraction matrix R and the membership matrix A have converged. If they have converged, output the clustering result; otherwise, update the attraction matrix R and the membership matrix A again. Updating the damping coefficient helps prevent numerical oscillations during iteration, which could lead to convergence problems.

[0031] In this embodiment, attractiveness reflects the degree to which sample k is suitable as the cluster center of sample i. Belonging reflects the degree to which sample i chooses sample k as its cluster center.

[0032] In the P algorithm, the similarity matrix S is usually taken as the negative of the distance:

[0033] In the formula, Represents the elements of the similarity matrix S; Indicates user and users Euclidean distance based on photovoltaic power output characteristics.

[0034] During the iteration process, information is transferred between sample points, updating the following two matrices: (1) Update the elements of the attraction matrix R:

[0035] in, Let t be the element of the updated attraction matrix R, and t represent the number of iterations. This is the affiliation value from the previous round. The formula indicates that... The value is equal to the similarity between i and k minus i's ability to choose other competing points.

[0036] (2) Update the affiliation matrix A:

[0037] in, This indicates the suitability of sample (i.e., user) i in choosing sample (i.e., user k) as its cluster center at time t+1; This represents the degree to which samples i' other than samples i and k consider sample k to be suitable as their cluster center in the (t+1)th iteration, i.e., the positive support from other samples for k; This represents the self-responsibility value of sample k in the (t+1)th iteration, reflecting the degree to which sample k believes it is suitable to be a cluster center.

[0038] In step 2) above, this embodiment uses a multivariate variational mode decomposition (MVMD) algorithm to simultaneously decompose a highly volatile multivariate load time series into net load component curves under multiple fluctuation frequency bands. For example... Figure 3 As shown, the intraday net load curves of different clusters are decomposed based on the MVMD method, and divided into net load component curves under multiple fluctuation frequency bands, including the following steps: 2.1) Define the input data. Based on the defined IMF set of load data decomposition, determine the input data. , specifically The input is multivariate load data.

[0039] Specifically, definition For the IMF set of load data decomposition, then the input data It can be represented as: (3) 2.2) Input data The input is a constructed multivariate variational mode decomposition model, used to analyze the input data. Construct the optimal constrained variational problem.

[0040] Specifically, since MVMD requires that the sum of the bandwidths of the extracted modes be minimized and that the extracted IMFs can accurately reconstruct the original signal, the optimal constrained variational problem for the actual sequence is as follows: (4) In the formula, and These correspond to the set of IMFs after multivariate load decomposition and the center frequency of each IMF component, respectively. It is the Dirac function; This is the convolution operator; V represents the set of multivariate load data; J represents the number of original input signals; K represents the Jacobian matrix; and t represents time.

[0041] 2.3) The constrained variational problem is unconstrained by using the Lagrangian function to solve the variational problem and determine whether the solution converges. If it does not converge, the alternating direction multiplier iterative algorithm is used to solve the unconstrained problem, iteratively updating the IMF set and the center frequency set of the IMF components to obtain the set number of IMF components to obtain a new IMF set, and then inputting the newly determined input data into the multivariate variational mode decomposition model.

[0042] In this embodiment, the constrained variational problem is made unconstrained using the Lagrangian function: (5) In the formula, For Lagrange factors; This is a penalty factor.

[0043] In this embodiment, the alternating direction multiplier iterative algorithm is used to solve the unconstrained problem and iteratively update... and After obtaining the specified number of IMF components, the new modal components can be expressed as: (6) In the formula, They are respectively The estimated value after Fourier transform.

[0044] This invention combines the characteristics of distributed generation fluctuations with the distribution of conventional power supply regulation rates to divide net load fluctuations into three frequency bands: fast (<15min), medium (15~60min), and slow (>1h).

[0045] 2.4) If convergence occurs, determine the upward and downward flexibility requirements. Combining the distributed generation fluctuation characteristics and power regulation rate distribution, divide the intraday net load curve into net load component curves of 3 frequency bands.

[0046] In this embodiment, after decomposing the flexibility requirements into fluctuation components at different time scales, the fluctuations at different scales correspond to different adjustment rate requirements. By identifying and dividing the waveform into upward and downward ramp sets, the upward and downward flexibility requirements at each time scale are obtained. The specific implementation method is as follows: Waveform identification was used to split the net load component curves of the three frequency bands into two ramp subsets: upward and downward.

[0047] In this subset, the amplitude of the climbing segment represents the flexibility requirement of that segment, and the duration represents the fluctuation period of that segment. The length of the fluctuation period under different fluctuation components should correspond to their defined time scale. Climbing slope When there is an upward need for flexibility; conversely... This is due to the need for downward flexibility. This represents the power value of the net load component.

[0048] In step 3) above, the flexibility resources of a power system including distributed energy sources such as wind power and photovoltaics mainly include thermal power plants with flexibility modifications, hydropower stations with regulation capabilities, power electronic energy storage devices, and controllable loads. Modeling is performed on various flexibility resources. Specifically, such as... Figure 4 As shown, the daily net load curves for each typical mode are combined with the corresponding controllable units of the power system to obtain the operating status of each type of controllable unit. This includes the following steps: 3.1) The short-term flexibility resources of thermal power plants are constrained by their upward and downward ramp rates and maximum and minimum operating output; on a long time scale, in addition to the online combined cycle gas turbine units, flexibility is also adjusted by starting and stopping gas turbine units and regulating the output of coal-fired units that have undergone flexibility modifications.

[0049] In this embodiment, the short-term flexibility resource is: (7) In the formula, , and These are the unit's maximum operating output, minimum operating output, and current output, respectively. These are the unit's upward and downward ramp rates, respectively. This refers to the time spent climbing the hill. For the upward flexibility resources (upward adjustment capacity) of thermal power units; For the downward flexibility resources (capacity reduction) of thermal power units.

[0050] In this embodiment, flexibility adjustment is achieved by starting and stopping the gas turbine unit and adjusting the output of the coal-fired unit that has undergone flexibility modification, as detailed below: (8) In the formula, These represent the current output and maximum regulating capacity of the peak-shaving unit, respectively.

[0051] 3.2) Hydropower units, as high-performance peak-shaving power sources, are characterized by rapid response, flexible operation, and convenient start-up and shutdown, and can provide corresponding flexible resources according to their operating status at various time scales throughout the day.

[0052] Specifically: (9) In the formula, , These are the maximum power generation and water storage capacity of the hydropower station, respectively. These represent the water volume and the upper and lower limits of the water storage capacity at time t, respectively. Adjust the time interval for climbing; For hydropower stations, this refers to upward flexibility resources (upward adjustment capacity). For hydropower stations, this refers to their downward flexibility resources (downward adjustment capacity).

[0053] 3.3) Power electronic energy storage devices are flexible power sources that can be quickly adjusted. Among them, supercapacitors and compressed air energy storage devices are suitable for flexible adjustment on time scales of <15min and 15~60min, respectively. The amount of flexible resources they provide can be quantified by equation (9).

[0054] Among them, power electronic energy storage devices have different application scenarios depending on the energy storage technology. Supercapacitors have fast dynamic response and long cycle life, but are more expensive and are suitable for smoothing high-frequency but low-energy fluctuations within 15 minutes; compressed air energy storage devices are low-cost, have large capacity, and slow dynamic response, but can smooth energy-concentrated fluctuations on the order of tens of minutes.

[0055] 3.4) The size of controllable load regulation resources is determined by the size of the electricity load of the demand-side response implementation agency and its own maximum output variation limit.

[0056] Specifically: (10) In the formula, These represent the current output and maximum output of the controllable load, respectively. These are the limits for the maximum upward and downward output variation of the controllable load itself, and their values ​​are related to the specific load characteristics. Upward flexibility resources provided for controllable loads; Downward flexibility resources provided for manageable loads.

[0057] In step 4) above, a matching analysis is performed on the flexibility resources and flexibility requirements at the same time scale. The probability of flexibility deficiency, the expected abundance of flexibility, and the expected insufficiency of flexibility at each time scale are calculated as flexibility indicators of the power system. These are then weighted to form a comprehensive system flexibility assessment index, including the following steps: 4.1) Select the probability of lack of flexibility, the expected abundance of flexibility, and the expected insufficiency of flexibility as flexibility indicators of the power system.

[0058] In this embodiment, the probability of lack of flexibility, the expected abundance of flexibility, and the expected insufficiency of flexibility are selected as flexibility indicators of the power system. The specific process is as follows: 4.1.1) Statistical analysis is performed on the ramp segment where flexibility resources are lower than flexibility requirements to obtain the probability of flexibility deficiency.

[0059] Specifically, the probability of lack of flexibility for: (11) In the formula, The number of climbing sections with insufficient flexibility; This represents the total number of ramp sections in the system.

[0060] 4.1.2) Based on the flexibility abundance, resources and demand of a certain ramp section, as well as the total installed capacity of the power system, set the flexibility abundance expectation to reflect the overall flexibility of the system.

[0061] Specifically, flexibility abundance expectation for: (12) In the formula, and These represent the flexibility abundance, resources, and demand for the i-th climbing segment, respectively. This represents the total number of system installations.

[0062] 4.1.3) By statistically analyzing the flexibility deficit in the insufficient flexibility ramp section, the expected value of the insufficient flexibility sample is calculated to reflect the severity of the power system flexibility gap.

[0063] Specifically, the expected value for samples with insufficient flexibility is:

[0064] (13) In the formula, Let be the size of the flexibility deficit for the i-th climbing segment.

[0065] 4.2) To intuitively reflect the overall flexibility of the system, a scale-weighted flexibility index is formed by weighting and summing the flexibility indices at each scale based on the flexibility index. The weights of the weighted flexibility indicators at each scale are determined by the average flexibility demand per unit time, in order to reflect the impact of fluctuations on the overall flexibility of the power system.

[0066] Specifically: (14) in: (15) In the formula: These are, respectively, the scale-weighted probability of insufficient flexibility, the expectation of flexibility abundance, and the expectation of insufficient flexibility, which comprehensively reflect flexibility across k scales. Assign weights to scale i; Let i be the size of the flexibility requirement per unit time at scale i; Let be the size of the flexibility deficit for the i-th climbing segment; Let be the expected abundance of flexibility for the i-th climbing segment; Let be the probability of lack of flexibility in the i-th climbing segment.

[0067] In one embodiment of the present invention, a power system flexibility assessment system based on entropy weight method improved clustering and multivariate variational mode decomposition is provided, comprising: The clustering module obtains the intraday net load curve and performs similarity clustering on the net load curve using a clustering algorithm based on the entropy weight method, simplifying the sample data into representative typical patterns. The flexibility requirement determination module decomposes the intraday net load curves of different clusters based on the multivariate variational mode decomposition algorithm, dividing them into net load component curves under multiple fluctuation frequency bands. By identifying the waveform, it divides the upward ramp set and the downward ramp set, and obtains the upward and downward flexibility requirements at each time scale. The flexibility resource determination module combines the units according to the intraday net load curves of each typical mode, obtains the corresponding operating status of each type of controllable unit in the power system, and calculates the upward and downward flexibility resources under different fluctuation time scales based on the operating status, output regulation characteristics and fluctuation period of each type of controllable unit. The assessment module performs matching analysis on flexibility resources and flexibility requirements at the same time scale, calculates the probability of flexibility deficiency, expected flexibility abundance, and expected flexibility insufficiency at each time scale as flexibility indicators of the power system, and weights them to form a comprehensive indicator for power system flexibility assessment.

[0068] In the above embodiments, the net load curves are clustered based on similarity using a clustering algorithm improved by the entropy weight method, including: The intraday net load curve is formed by the historical output curves of load, wind power, and photovoltaic power, and the daily average load, daily maximum load, and daily maximum fluctuation difference are selected as load characteristic indicators. The entropy weight method is used to assign weights to different characteristic indicators, and the clustering algorithm is improved by setting an initial weight vector to generate new cluster centers; After generating new cluster centers, the contribution of each characteristic index to the cluster centers is calculated, resulting in a similarity matrix with m evaluation objects and k evaluation indicators. The attraction matrix and belonging matrix are updated based on the similarity matrix and the damping coefficient. It is then determined whether the attraction matrix and belonging matrix have converged. If they have converged, the clustering result is output; otherwise, the attraction matrix and belonging matrix are updated again.

[0069] In the above embodiments, the intraday net load curves of different clusters are decomposed based on the multivariate variational mode decomposition algorithm, and divided into net load component curves under multiple fluctuation frequency bands, including: Based on the defined IMF set of load data decomposition, determine the input data. ; Input data The input is a constructed multivariate variational mode decomposition model, used to analyze the input data. Construct the optimal constrained variational problem; The constrained variational problem is unconstrained by using the Lagrangian function to solve the variational problem and determine whether the solution converges. If it does not converge, the alternating direction multiplier iterative algorithm is used to solve the unconstrained problem. The IMF set and the center frequency set of the IMF components are iteratively updated to obtain the set number of IMF components to obtain a new IMF set. The newly determined input data is then input into the multivariate variational mode decomposition model. If convergence occurs, the upward and downward flexibility requirements are determined. Combining the characteristics of distributed generation fluctuations and the distribution of power regulation rates, the intraday net load curve is divided into net load component curves in three frequency bands.

[0070] In the above embodiments, by dividing the upward climbing set and the downward climbing set through waveform identification, the upward and downward flexibility requirements at each time scale are obtained, including: The net load component curves of the three frequency bands are divided into two ramp subsets, one upward and one downward, by waveform identification. Among them, the amplitude of the climbing segment in the climbing subset represents the flexibility requirement of the climbing segment, the duration represents the fluctuation period of the climbing segment, and the fluctuation period length under different fluctuation components should correspond to the time scale defined by them.

[0071] In the above embodiments, the intraday net load curves of each typical mode are combined with the corresponding units to obtain the operating status of each type of controllable unit in the power system, including: The short-term flexibility resources of thermal power plants are constrained by their upward and downward ramp rates and maximum and minimum operating output. On a long-term scale, in addition to the combined cycle gas turbine units in operation, flexibility is also achieved by starting and stopping the gas turbine units and adjusting the output of the coal-fired units that have undergone flexibility modifications. Hydropower units serve as peak-shaving power sources, providing corresponding flexibility resources at various time scales throughout the day based on their operating status. Power electronic energy storage devices are flexible power sources that can be quickly adjusted. Among them, supercapacitors and compressed air energy storage devices are suitable for flexible adjustment on time scales of <15 minutes and 15~60 minutes, respectively. The size of controllable load regulation resources is determined by the size of the electricity load of the demand-side response implementation agency and its own maximum output variation limit.

[0072] In the above embodiments, a matching analysis is performed on flexibility resources and flexibility requirements at the same time scale. The probability of flexibility deficiency, the expected abundance of flexibility, and the expected insufficiency of flexibility at each time scale are calculated as flexibility indicators of the power system. These are then weighted to form a comprehensive system flexibility assessment index, including: The probability of lack of flexibility, the expected abundance of flexibility, and the expected insufficiency of flexibility are selected as flexibility indicators for power systems. Based on the flexibility index, the flexibility indexes at each scale are weighted and summed to form the scale-weighted flexibility index. The weight of each scale-weighted flexibility index is determined by the average flexibility demand per unit time, so as to reflect the impact of fluctuations on the overall flexibility of the power system.

[0073] In the above embodiments, the probability of lack of flexibility, the expected abundance of flexibility, and the expected deficiency of flexibility are selected as flexibility indicators of the power system, specifically: Statistical analysis was conducted on the flexibility-deficient ramp-up segment where flexibility resources were lower than flexibility requirements to obtain the probability of flexibility deficiency. Based on the flexibility abundance, resources and demand of a certain ramp segment, as well as the total installed capacity of the power system, set the flexibility abundance expectation; By statistically analyzing the flexibility deficit in the insufficient flexibility ramp-up section, the expected value of the insufficient flexibility sample is calculated to reflect the severity of the flexibility gap in the power system.

[0074] The system provided in this embodiment is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0075] In one embodiment of the present invention, a computing device is provided. This computing device can be a terminal and may include a processor, a communication interface, memory, a display screen, and an input device. The processor, communication interface, and memory communicate with each other via a communication bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. When the computer programs are executed by the processor, they implement the methods described in the above embodiments. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, a management network, NFC (Near Field Communication), or other technologies. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the casing of the computing device, or an external keyboard, touchpad, or mouse. The processor can call logical instructions stored in the memory.

[0076] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0077] In one embodiment of the present invention, a computer program product is provided, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, and when the program instructions are executed by a computer, the computer is able to perform the methods provided in the above-described method embodiments.

[0078] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided, which stores server instructions that cause a computer to perform the methods provided in the above embodiments.

[0079] The computer-readable storage medium provided in the above embodiments has a similar implementation principle and technical effect to the above method embodiments, and will not be described again here.

[0080] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A power system flexibility assessment method based on entropy weighting and improved clustering and multivariate variational mode decomposition, characterized in that, include: The intraday net load curve is obtained, and the net load curve is clustered by similarity using a clustering algorithm based on entropy weight method, simplifying the sample data into representative typical patterns. Based on the multivariate variational mode decomposition algorithm, the intraday net load curves of different clusters are decomposed into net load component curves under multiple fluctuation frequency bands. The upward ramp set and downward ramp set are divided by waveform identification to obtain the upward and downward flexibility requirements at each time scale. The daily net load curves of each typical mode are combined with the units to obtain the corresponding operating status of each type of controllable unit in the power system. Based on the operating status, output regulation characteristics and fluctuation period of each type of controllable unit, the upward and downward flexibility resources under different fluctuation time scales are calculated. A matching analysis of flexibility resources and flexibility requirements at the same time scale is conducted to calculate the probability of flexibility deficiency, the expected abundance of flexibility, and the expected insufficiency of flexibility at each time scale as flexibility indicators of the power system. These are then weighted to form a comprehensive indicator for power system flexibility assessment.

2. The power system flexibility assessment method based on entropy weight method improved clustering and multivariate variational mode decomposition as described in claim 1, characterized in that, The net load curves are clustered based on similarity using a clustering algorithm improved by the entropy weight method, including: The intraday net load curve is formed by the historical output curves of load, wind power, and photovoltaic power, and the daily average load, daily maximum load, and daily maximum fluctuation difference are selected as load characteristic indicators. The entropy weight method is used to assign weights to different characteristic indicators, and the clustering algorithm is improved by setting an initial weight vector to generate new cluster centers; After generating new cluster centers, the contribution of each characteristic index to the cluster centers is calculated, resulting in a similarity matrix with m evaluation objects and k evaluation indicators. The attraction matrix and belonging matrix are updated based on the similarity matrix and the damping coefficient. It is then determined whether the attraction matrix and belonging matrix have converged. If they have converged, the clustering result is output; otherwise, the attraction matrix and belonging matrix are updated again.

3. The power system flexibility assessment method based on entropy weight method improved clustering and multivariate variational mode decomposition as described in claim 1, characterized in that, The intraday net load curves of different clusters are decomposed based on the multivariate variational mode decomposition algorithm, and divided into net load component curves under multiple fluctuation frequency bands, including: Based on the defined IMF set of load data decomposition, determine the input data. ; Input data The input is a constructed multivariate variational mode decomposition model, used to analyze the input data. Construct the optimal constrained variational problem; The constrained variational problem is unconstrained by using the Lagrangian function to solve the variational problem and determine whether the solution converges. If it does not converge, the alternating direction multiplier iterative algorithm is used to solve the unconstrained problem. The IMF set and the center frequency set of the IMF components are iteratively updated to obtain the set number of IMF components to obtain a new IMF set. The newly determined input data is then input into the multivariate variational mode decomposition model. If convergence occurs, the upward and downward flexibility requirements are determined. Combining the characteristics of distributed generation fluctuations and the distribution of power regulation rates, the intraday net load curve is divided into net load component curves in three frequency bands.

4. The power system flexibility assessment method based on entropy weight method improved clustering and multivariate variational mode decomposition as described in claim 3, characterized in that, By segmenting the upward and downward ramp sets through waveform identification, the upward and downward flexibility requirements at various time scales are obtained, including: The net load component curves of the three frequency bands are divided into two ramp subsets, one upward and one downward, by waveform identification. Among them, the amplitude of the climbing segment in the climbing subset represents the flexibility requirement of the climbing segment, the duration represents the fluctuation period of the climbing segment, and the fluctuation period length under different fluctuation components should correspond to the time scale defined by them.

5. The power system flexibility assessment method based on entropy weight method improved clustering and multivariate variational mode decomposition as described in claim 1, characterized in that, By combining the intraday net load curves of each typical mode, the operating status of various types of controllable units in the corresponding power system is obtained, including: The short-term flexibility resources of thermal power plants are constrained by their upward and downward ramp rates and maximum and minimum operating output. On a long-term scale, in addition to the combined cycle gas turbine units in operation, flexibility is also achieved by starting and stopping the gas turbine units and adjusting the output of the coal-fired units that have undergone flexibility modifications. Hydropower units serve as peak-shaving power sources, providing corresponding flexibility resources at various time scales throughout the day based on their operating status. Power electronic energy storage devices are flexible power sources that can be quickly adjusted. Among them, supercapacitors and compressed air energy storage devices are suitable for flexible adjustment on time scales of <15 minutes and 15~60 minutes, respectively. The size of controllable load regulation resources is determined by the size of the electricity load of the demand-side response implementation agency and its own maximum output variation limit.

6. The power system flexibility assessment method based on entropy weight method improved clustering and multivariate variational mode decomposition as described in claim 1, characterized in that, A matching analysis of flexibility resources and flexibility requirements at the same time scale is conducted to calculate the probability of flexibility deficiency, expected flexibility abundance, and expected flexibility insufficiency at each time scale as flexibility indicators of the power system. These are then weighted to form a comprehensive system flexibility assessment index, including: The probability of lack of flexibility, the expected abundance of flexibility, and the expected insufficiency of flexibility are selected as flexibility indicators for power systems. Based on the flexibility index, the flexibility indexes at each scale are weighted and summed to form the scale-weighted flexibility index. The weight of each scale-weighted flexibility index is determined by the average flexibility demand per unit time, so as to reflect the impact of fluctuations on the overall flexibility of the power system.

7. The power system flexibility assessment method based on entropy weight method improved clustering and multivariate variational mode decomposition as described in claim 6, characterized in that, The probability of lack of flexibility, the expected abundance of flexibility, and the expected deficiency of flexibility are selected as flexibility indicators for the power system, specifically: Statistical analysis was conducted on the flexibility-deficient ramp-up segment where flexibility resources were lower than flexibility requirements to obtain the probability of flexibility deficiency. Based on the flexibility abundance, resources and demand of a certain ramp segment, as well as the total installed capacity of the power system, set the flexibility abundance expectation; By statistically analyzing the flexibility deficit in the insufficient flexibility ramp-up section, the expected value of the insufficient flexibility sample is calculated to reflect the severity of the flexibility gap in the power system.

8. A power system flexibility assessment system based on entropy weight method-improved clustering and multivariate variational mode decomposition, characterized in that, include: The clustering module obtains the intraday net load curve and performs similarity clustering on the net load curve using a clustering algorithm based on the entropy weight method, simplifying the sample data into representative typical patterns. The flexibility requirement determination module decomposes the intraday net load curves of different clusters based on the multivariate variational mode decomposition algorithm, dividing them into net load component curves under multiple fluctuation frequency bands. By identifying the waveform, it divides the upward ramp set and the downward ramp set, and obtains the upward and downward flexibility requirements at each time scale. The flexibility resource determination module combines the units according to the intraday net load curves of each typical mode, obtains the corresponding operating status of each type of controllable unit in the power system, and calculates the upward and downward flexibility resources under different fluctuation time scales based on the operating status, output regulation characteristics and fluctuation period of each type of controllable unit. The assessment module performs matching analysis on flexibility resources and flexibility requirements at the same time scale, calculates the probability of flexibility deficiency, expected flexibility abundance, and expected flexibility insufficiency at each time scale as flexibility indicators of the power system, and weights them to form a comprehensive indicator for power system flexibility assessment.

9. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any of the methods described in claims 1 to 7.

10. A computing device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing any of the methods described in claims 1 to 7.

Images