CN-122020342-A - Nuclear power plant abnormality detection method and device, electronic equipment and storage medium

CN122020342ACN 122020342 ACN122020342 ACN 122020342ACN-122020342-A

Abstract

The invention relates to the technical field of intelligent operation and maintenance of nuclear power plants, and discloses a method, a device, electronic equipment and a storage medium for detecting abnormality of a nuclear power plant, wherein the method comprises the steps of obtaining first operation data of a nuclear power unit; the method comprises the steps of inputting first operation data into a stacked self-encoder to obtain first data features of the first operation data and first reconstruction errors of the first operation data, carrying out cluster analysis on the first data features based on a Gaussian mixture model to obtain minimum distances between the first data features and cluster centers of a plurality of target clusters obtained by clustering the Gaussian mixture model, wherein the plurality of target clusters are obtained by clustering second data features respectively obtained by the stacked self-encoder for a plurality of second operation data by the Gaussian mixture model, and determining abnormal detection results of a nuclear power unit based on comparison relations between the first reconstruction errors and reconstruction error thresholds and comparison relations between the minimum distances and distance thresholds. The method and the device can solve the problem of low abnormality detection accuracy of the nuclear power plant.

Inventors

Yuan Jinxiao
WANG SHAOHUA
ZHANG ZHILIANG
SONG JIANJUN
FAN SUI
CHEN JIE
Cao Xingdi
LIU WENJIE
WANG JIAN

Assignees

中国核电工程有限公司

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (10)

1. A method for detecting anomalies in a nuclear power plant, the method comprising: acquiring first operation data of a nuclear power unit; Inputting the first operation data into a stacked self-encoder to obtain a first data characteristic of the first operation data and a first reconstruction error of the first operation data; Performing cluster analysis on the first data features based on a Gaussian mixture model to obtain minimum distances between the first data features and cluster centers of a plurality of target clusters obtained by clustering the Gaussian mixture model, wherein the plurality of target clusters are obtained by clustering second data features respectively obtained by stacking the self-encoder for a plurality of second operation data by the Gaussian mixture model; and determining an abnormality detection result of the nuclear power unit based on the comparison relation between the first reconstruction error and a reconstruction error threshold value and the comparison relation between the minimum distance and a distance threshold value.
2. The method of claim 1, wherein the reconstruction error threshold is a maximum of second reconstruction errors respectively derived by the stacked self-encoder for the plurality of second operational data.
3. The method of claim 1, wherein the distance threshold is a maximum of target distances between each of the second data features and a cluster center of the plurality of target clusters, respectively.
4. The method of claim 1, wherein the determining the anomaly detection result for the nuclear power unit based on the comparison of the first reconstruction error to a reconstruction error threshold and the comparison of the minimum distance to a distance threshold comprises: and if the first reconstruction error is larger than the reconstruction error threshold and the minimum distance is larger than the distance threshold, judging that the abnormality detection result of the nuclear power unit is that the nuclear power unit is abnormal.
5. The method according to claim 1, wherein the method further comprises: acquiring the plurality of second operation data; respectively inputting the plurality of second operation data into the stacked self-encoder to obtain second data characteristics respectively corresponding to the plurality of second operation data; and carrying out cluster analysis on second data features respectively corresponding to the plurality of second operation data based on the Gaussian mixture model to obtain the plurality of target cluster clusters.
6. The method of claim 5, wherein performing cluster analysis on second data features corresponding to the plurality of second operation data respectively based on the gaussian mixture model to obtain the plurality of target clusters comprises: acquiring a plurality of candidate cluster numbers; Evaluating the Gaussian mixture model according to the number of candidate clusters based on third data features corresponding to a plurality of third operation data respectively to obtain at least one clustering index value of the Gaussian mixture model under the number of candidate clusters; determining a target cluster number from the plurality of candidate cluster numbers based on the at least one cluster index value; And carrying out cluster analysis on second data features respectively corresponding to the plurality of second operation data based on the target cluster number and the Gaussian mixture model to obtain a plurality of target cluster groups.
7. The method of claim 6, wherein the at least one cluster index value comprises a Bayesian information criterion value, a red pool information criterion value, and a contour coefficient, wherein the determining a target cluster number from the plurality of candidate cluster numbers based on the at least one cluster index value comprises: Carrying out weighted fusion on the Bayesian information criterion value, the red pool information criterion value and the contour coefficient under the same candidate cluster number to obtain a cluster evaluation result of the Gaussian mixture model under each candidate cluster number; And determining the target cluster number from the candidate cluster numbers based on the cluster evaluation result.
8. A nuclear power plant abnormality detection apparatus, characterized by comprising: The data acquisition module is used for acquiring first operation data of the nuclear power unit; The feature extraction module is used for inputting the first operation data into the stacked self-encoder to obtain a first data feature of the first operation data and a first reconstruction error of the first operation data; The cluster analysis module is used for carrying out cluster analysis on the first data features based on a Gaussian mixture model to obtain the minimum distance between the first data features and the cluster centers of a plurality of target cluster clusters obtained by clustering the Gaussian mixture model, wherein the plurality of target cluster clusters are obtained by clustering second data features respectively obtained by stacking the self-encoder for a plurality of second operation data by the Gaussian mixture model; The abnormality detection module is used for determining an abnormality detection result of the nuclear power unit based on the comparison relation between the first reconstruction error and a reconstruction error threshold value and the comparison relation between the minimum distance and a distance threshold value.
9. An electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the nuclear power plant anomaly detection method of any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon computer instructions for causing a computer to execute the nuclear power plant abnormality detection method according to any one of claims 1 to 7.

Description

Nuclear power plant abnormality detection method and device, electronic equipment and storage medium Technical Field The invention relates to the technical field of intelligent operation and maintenance of nuclear power plants, in particular to a method and device for detecting abnormality of a nuclear power plant, electronic equipment and a storage medium. Background Nuclear power plants operate in highly complex and safety critical environments, even subtle anomalies may evolve into catastrophic failures. Therefore, accurate and timely anomaly detection is important for maintaining operation integrity, protecting personnel safety and preventing environmental hazards. However, the conventional monitoring method generally needs to wait for the abnormal value to reach a larger fixed threshold value to trigger an alarm, so that it is difficult to accurately detect the fine abnormality in the complex high-dimensional data generated by the nuclear power plant system in time, and at this time, the accident often happens already, which results in low detection accuracy of the abnormality of the nuclear power plant. Disclosure of Invention The invention provides a nuclear power plant abnormality detection method, a device, electronic equipment and a storage medium, which are used for solving the problem of low accuracy of nuclear power plant abnormality detection. In a first aspect, the present invention provides a method for detecting an abnormality in a nuclear power plant, including: acquiring first operation data of a nuclear power unit; Inputting the first operation data into a stacked self-encoder to obtain a first data characteristic of the first operation data and a first reconstruction error of the first operation data; Performing cluster analysis on the first data features based on a Gaussian mixture model to obtain minimum distances between the first data features and cluster centers of a plurality of target clusters obtained by clustering the Gaussian mixture model, wherein the plurality of target clusters are obtained by clustering second data features respectively obtained by stacking the self-encoder for a plurality of second operation data by the Gaussian mixture model; and determining an abnormality detection result of the nuclear power unit based on the comparison relation between the first reconstruction error and a reconstruction error threshold value and the comparison relation between the minimum distance and a distance threshold value. In a second aspect, the present invention provides an abnormality detection apparatus for a nuclear power plant, comprising: The data acquisition module is used for acquiring first operation data of the nuclear power unit; The feature extraction module is used for inputting the first operation data into the stacked self-encoder to obtain a first data feature of the first operation data and a first reconstruction error of the first operation data; The cluster analysis module is used for carrying out cluster analysis on the first data features based on a Gaussian mixture model to obtain the minimum distance between the first data features and the cluster centers of a plurality of target cluster clusters obtained by clustering the Gaussian mixture model, wherein the plurality of target cluster clusters are obtained by clustering second data features respectively obtained by stacking the self-encoder for a plurality of second operation data by the Gaussian mixture model; The abnormality detection module is used for determining an abnormality detection result of the nuclear power unit based on the comparison relation between the first reconstruction error and a reconstruction error threshold value and the comparison relation between the minimum distance and a distance threshold value. In a third aspect, the present invention provides an electronic device, including a memory and a processor, where the memory and the processor are communicatively connected to each other, and the memory stores computer instructions, and the processor executes the computer instructions, so as to execute the method for detecting an abnormality of a nuclear power plant according to the first aspect or any embodiment corresponding to the first aspect. In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the nuclear power plant abnormality detection method of the first aspect or any one of its corresponding embodiments. According to the nuclear power plant abnormality detection method, the first operation data of the nuclear power unit is obtained, so that comprehensive and accurate basic data are provided for subsequent abnormality detection. The first operation data is input into the stacked self-encoder, the stacked self-encoder is utilized for feature extraction, complex nonlinear relations in the line operation data can be effectively captured, further, first data features of