Intelligent triggering data classification method for loosening part monitoring system

Abstract

The invention discloses an intelligent triggering data classification method for a loosening part monitoring system. The method comprises steps that original triggering data of the loosening part monitoring system is acquired; data corresponding to each channel is acquired based on the original triggering data; waveform characteristic extraction of the data of each channel is carried out to acquire a characteristic vector of each channel; a single-channel classification model is established, the data classification result of each channel is acquired based on the single-channel classification model and the characteristic vector of each channel, distinguishing, statistics and management of actual loosening part knock signal data and error triggering data are carried out in an intelligent mode, and data analysis efficiency of a loosening part is improved.

Description

technical field [0001] The invention relates to the field of reactor and primary circuit loose parts monitoring, in particular to a method for intelligently classifying trigger data of a loose parts monitoring system. Background technique [0002] The interior of the nuclear reactor pressure vessel includes reactor internals and fuel assemblies. The reactor internals are mainly composed of compression assemblies, basket assemblies, and under-core support assemblies. Most of the components and parts in each assembly are connected by screws and pins. Under the condition of long-term continuous operation of the pile, due to the impact of water flow and flow-induced vibration, some parts may become loose or even fall off, thus forming loose parts in the primary circuit. During construction, refueling or maintenance, there is also the possibility of leaving metal parts (called foreign parts) in the primary circuit system. If these loose parts are not found and disposed of in tim...

AI Technical Summary

Problems solved by technology

[0004] First, the real impact data of loose parts may be submerged in these "false trigger" data, which affects the timeliness and accuracy of analysis of loose parts;

[0005] Second, the workload of system operation and maintenance personnel to analyze the trigger data of the daily loose parts system is heavy and mechanical.

[0007] In the existing technology, the existing loose parts monitoring system has the technical problems of easily causing false triggering, poor analysis timeliness and accuracy, and low trigger data analysis efficiency

Images