State Index Evaluation and Optimization Method Based on Signal Detection and ROC Analysis

Abstract

The invention discloses a state index evaluation and optimization method based on signal detection and ROC analysis. The method comprises a step of determining a target moving component and a failuremode and preparing monitoring signals of the moving component in a healthy state and the failure mode, a step of combing state indexes and extracting the state indexes from a healthy signal and a failure signal, and establishing a state index set, a step of estimating probability density functions of state indexes in a healthy state and a failure state, a step of calculating a state index discrimination index based on the signal detection theory and carrying out state index estimation, a step of solving an ROC curve, a step of constructing an optimization target function based on the ROC curve, and a step of solving an equation, obtaining an optimization target function minimum point, and thus obtaining a state index optimal threshold. The method provided by the invention has the advantages of a simple principle, simple operation and high stability and reliability and is applicable to the state index evaluation and optimization of various rotating machinery state monitoring and fault diagnosis.

Description

technical field [0001] The invention belongs to the field of monitoring and fault diagnosis, relates to the problem of state index evaluation and threshold optimization, and in particular relates to a state index evaluation and optimization method based on signal detection and ROC analysis. Background technique [0002] Rotating mechanical components (including gears, bearings, shaft parts, etc.) are widely used in the transmission system of large equipment. The health status of rotating mechanical components has a great impact on the operating performance of the equipment. Causes equipment downtime, which has a great impact on production. In response to this problem, people have vigorously developed a condition-based maintenance (CBM) strategy, which uses various sensors to monitor mechanical equipment, and can sense the health status of the equipment in time, so that the equipment failure information can be learned at the first time. In the CBM system, the state index is ...

AI Technical Summary

Problems solved by technology

[0003] However, the increase of the state index also brings some problems: First, which state index should be selected in the CBM system

In order to achieve effective anomaly detection, engineers need to make a trade-off between the failure hit rate and the false alarm rate. False alarms will cause unnecessary shutdowns and cause unprofitable maintenance costs; increasing the status index threshold can reduce the false alarm rate, but At the same time, it will cause a decrease in the failure hit rate, resulting in the failure to detect some failures in the operation process in time, increasing the operation risk of the system

Images