CN-122026394-A - Power system frequency modulation parameter aggregation method considering new energy

CN122026394ACN 122026394 ACN122026394 ACN 122026394ACN-122026394-A

Abstract

The invention discloses a method for aggregating frequency modulation parameters of a power system taking new energy into account, which belongs to the technical field of operation and control of the power system and comprises the steps of obtaining operation state data of each new energy station in the power system and real-time frequency data of a power grid, carrying out fusion processing to generate a multi-dimensional operation data set, extracting characteristic indexes representing frequency modulation response capacity of each new energy station to construct dynamic response characteristic vectors of each new energy station, dividing each new energy station into different frequency modulation resource clusters, analyzing frequency modulation response complementary characteristics among different frequency modulation resource clusters, establishing a cooperative gain function, carrying out mapping calculation on original frequency modulation parameters of each new energy station in the frequency modulation resource clusters, and generating cluster equivalent frequency modulation parameters. By adopting a technical path based on multi-dimensional feature extraction and dynamic clustering and introducing a cooperative gain function and a closed loop feedback mechanism, equivalent modeling of new energy frequency modulation resources can be realized, so that the frequency modulation control capability of a power grid on high-proportion new energy is improved.

Inventors

Wang Shengrun
FANG JUNWU
ZHAO LICHUN
YANG CHUNLI
ZHANG WEI
FU ZHENG
YANG YUCHI
GE YU
ZHANG JIAN

Assignees

国家电投集团东北电力开发有限公司

Dates

Publication Date: 20260512
Application Date: 20251215

Claims (10)

1. The power system frequency modulation parameter aggregation method considering new energy is characterized by comprising the following steps: Acquiring running state data of each new energy station in the power system and real-time frequency data of a power grid, and carrying out fusion processing on the running state data and the real-time frequency data to generate a multi-dimensional running data set; Extracting characteristic indexes representing the frequency modulation response capability of each new energy station from the multidimensional operation data set, and constructing dynamic response characteristic vectors of each new energy station based on the characteristic indexes; Dividing each new energy station into different frequency modulation resource clusters based on the dynamic response feature vector; Analyzing complementary characteristics of frequency modulation response among all new energy stations in each frequency modulation resource cluster, and establishing a cooperative gain function; And performing mapping calculation on the original frequency modulation parameters of each new energy station in the frequency modulation resource cluster by using the cooperative gain function, and generating a cluster equivalent frequency modulation parameter corresponding to the frequency modulation resource cluster.
2. The method of power system frequency modulation parameter aggregation accounting for new energy as claimed in claim 1, wherein said generating a multi-dimensional operational dataset comprises: Acquiring output fluctuation parameters, available capacity parameters and real-time frequency deviation parameters of a power grid of each new energy station in the power system; performing time sequence alignment processing on the output fluctuation parameter, the available capacity parameter and the real-time frequency deviation parameter to generate a synchronous operation data sequence; and carrying out multidimensional normalization processing on the synchronous operation data sequence to generate the multidimensional operation data set.
3. The method for aggregating frequency modulation parameters of a power system according to claim 1, wherein extracting feature indexes representing frequency modulation response capabilities of each new energy station from the multidimensional operation data set, and constructing dynamic response feature vectors of each new energy station based on the feature indexes comprises: Extracting characteristic indexes representing frequency modulation response capacity of each new energy station from the multi-dimensional operation data set, wherein the characteristic indexes comprise an instantaneous power adjustable margin, a power change rate limit value and historical frequency response sensitivity; and normalizing the instantaneous power adjustable margin, the power change rate limit value and the historical frequency response sensitivity in the characteristic index, and then combining to form a dynamic response characteristic vector.
4. The method for aggregating frequency modulation parameters of a power system accounting for new energy according to claim 3, further comprising: Acquiring historical frequency modulation response data and corresponding historical frequency data of each new energy station; calculating the frequency modulation action delay time and the frequency modulation power tracking error of each new energy station according to the historical frequency modulation response data and the historical frequency data; Generating response characteristic evaluation vectors of all new energy stations according to the frequency modulation action delay time and the frequency modulation power tracking error; Judging whether the response characteristic evaluation vector meets a preset quality threshold condition or not; when the response characteristic evaluation vector does not meet the quality threshold condition, eliminating the corresponding new energy station; When the response characteristic evaluation vector meets the quality threshold condition, the instantaneous power adjustable margin, the power change rate limit value and the historical frequency response sensitivity in the characteristic index are normalized and then combined, and the combination is weighted and fused with the response characteristic evaluation vector to generate a dynamic response characteristic vector.
5. The method for aggregating frequency modulation parameters of a power system according to claim 3, wherein the dividing each new energy station into different frequency modulation resource clusters based on the dynamic response feature vector comprises: Calculating the density distribution of each new energy station in the feature space by using the dynamic response feature vector; And dividing a plurality of density areas based on the density distribution, and classifying the new energy stations positioned in the same density area into the same frequency modulation resource cluster.
6. The method for aggregating frequency modulation parameters of a power system accounting for new energy as defined in claim 5, wherein said establishing a collaborative gain function comprises: Calculating the response characteristic complementarity between any two new energy stations according to the dynamic response characteristic vectors of the new energy stations in each frequency modulation resource cluster; and constructing a cooperative gain function reflecting the overall coordination level of all new energy stations in the cluster based on the response characteristic complementarity.
7. The method for aggregating frequency modulation parameters of a power system based on new energy as defined in claim 6, wherein constructing a collaborative gain function reflecting an overall coordination level of all new energy sites in the cluster based on the response characteristic complementarity comprises: Calculating the weight coefficient of each new energy station in the cluster for cluster frequency modulation contribution by using the response characteristic complementarity; Constructing an optimization model with preset cluster equivalent frequency modulation capacity as a target by taking the weight coefficient as a variable; and solving the optimization model to obtain a function expression of the collaborative gain function.
8. The method for aggregating frequency modulation parameters of a power system according to claim 6, wherein the mapping calculation is performed on the original frequency modulation parameters of each new energy station in the frequency modulation resource cluster by using the collaborative gain function, and generating cluster equivalent frequency modulation parameters corresponding to the frequency modulation resource cluster includes: acquiring original frequency modulation parameters of each new energy station in the frequency modulation resource cluster; And inputting the original frequency modulation parameters into the collaborative gain function, and calculating to obtain cluster equivalent frequency modulation parameters, wherein the cluster equivalent frequency modulation parameters comprise equivalent frequency modulation dead zones, equivalent difference modulation coefficients and equivalent power limit values.
9. The method for aggregating frequency modulation parameters of a power system that accounts for new energy according to claim 8, further comprising: Monitoring the actual frequency modulation response performance of the frequency modulation resource cluster in real time; Calculating a frequency modulation deviation index of the actual frequency modulation response performance and the expected performance; When the frequency modulation deviation index is smaller than a preset deviation threshold value, dynamically correcting the cooperative gain function; And when the frequency modulation deviation index is larger than or equal to the deviation threshold value, re-triggering the construction of the dynamic response characteristic vector and the division of the frequency modulation resource cluster.
10. The utility model provides an electric power system frequency modulation parameter aggregation system who considers new forms of energy which characterized in that, the system includes: the data acquisition module is used for acquiring the running state data of each new energy station in the power system and the real-time frequency data of the power grid, and carrying out fusion processing on the running state data and the real-time frequency data to generate a multi-dimensional running data set; the feature extraction module is used for extracting feature indexes representing the frequency modulation response capability of each new energy station from the multi-dimensional operation data set and constructing dynamic response feature vectors of each new energy station based on the feature indexes; the cluster dividing module is used for dividing each new energy station into different frequency modulation resource clusters based on the dynamic response feature vector; The cooperative modeling module is used for analyzing the complementary characteristics of the frequency modulation response among the new energy stations in each frequency modulation resource cluster and establishing a cooperative gain function; And the equivalent parameter generation module is used for carrying out mapping calculation on the original frequency modulation parameters of each new energy station in the frequency modulation resource cluster by utilizing the cooperative gain function to generate cluster equivalent frequency modulation parameters corresponding to the frequency modulation resource cluster.

Description

Power system frequency modulation parameter aggregation method considering new energy Technical Field The invention relates to the technical field of operation and control of power systems, in particular to a power system frequency modulation parameter aggregation method taking new energy into account. Background In an electric power system, maintaining frequency stability is one of core tasks for guaranteeing safe operation, and conventionally, the frequency modulation capability of synchronous power supplies such as a hydroelectric generating set, a thermal power generating set and the like is mainly relied on. With the continuous improvement of the power generation proportion of new energy sources such as wind power, photovoltaic and the like, the frequency modulation response characteristics of the power supplies are obviously different from those of the traditional synchronous units due to the intermittence and fluctuation of the output of the power supplies and the characteristic of grid connection of power electronic equipment. How to effectively aggregate the parameters of a plurality of new energy stations with different characteristics to form one or a plurality of equivalent frequency modulation resource models, and to analyze and decide the frequency control by a power supply network dispatching center, is a technical problem to be solved in the current power system operation field. In the prior art, a simple weighted average method is generally adopted for the aggregation of distributed resources, for example, according to the rated capacity or the current available capacity proportion of each station, the frequency modulation parameters of the stations are subjected to arithmetic average or weighted average, so as to obtain a total equivalent parameter. Another idea is to group roughly according to the type of power source, for example, all wind farms into one group, all photovoltaic power plants into another group, and then to aggregate the parameters within the group. These approaches simplify the complexity of the scheduling model to some extent. However, the simple weighted average method completely ignores the huge differences of different new energy stations in real-time frequency modulation capability, dynamic response speed and historical response quality, so that the aggregated equivalent parameters deviate from the overall response characteristics of the actual clusters seriously, and the actual frequency modulation capability cannot be reflected accurately. According to the rough grouping mode of the power supply type, the response characteristic differentiation possibly generated by different geographic positions, equipment types and control strategies among sites of the same type is ignored, and the fine effective aggregation cannot be realized. More importantly, the conventional method generally lacks consideration of response characteristic complementarity and synergistic effect among units in a cluster, and fails to establish a mechanism for performing model self-adaptive correction according to actual operation effects, so that an aggregation model is rapidly disabled when the operation state of new energy is changed. Disclosure of Invention In order to solve the problems, the invention provides a power system frequency modulation parameter aggregation method considering new energy, which adopts a technical path based on multi-dimensional feature extraction and dynamic clustering, introduces a cooperative gain function and a closed loop feedback mechanism, and can realize equivalent modeling of new energy frequency modulation resources, thereby improving the frequency modulation control capability of a power grid on high-proportion new energy. The above object can be achieved by the following scheme: A method for aggregating frequency modulation parameters of an electric power system for considering new energy includes the steps of obtaining running state data of each new energy station in the electric power system and real-time frequency data of an electric network, carrying out fusion processing on the running state data and the real-time frequency data to generate a multi-dimensional running data set, extracting characteristic indexes representing frequency modulation response capacity of each new energy station from the multi-dimensional running data set, constructing dynamic response characteristic vectors of each new energy station based on the characteristic indexes, dividing each new energy station into different frequency modulation resource clusters based on the dynamic response characteristic vectors, analyzing frequency modulation response complementary characteristics among each new energy station in each frequency modulation resource cluster, establishing a cooperative gain function, and carrying out mapping calculation on original frequency modulation parameters of each new energy station in each frequency modulation resource cluster by utilizing the cooperative