Equipment life and health monitoring method and system based on original fault data

Abstract

The invention discloses an equipment life and health monitoring method and system based on original fault data. The method includes the steps: preprocessing the original fault data; building a two-parameter Weibull distribution model, determining a failure distribution function value by the aid of an approximate medium rank formula method, fitting out a regression equation of the Weibull distribution model by a least-square parameter estimation method at fault occurrence time based on the calculated failure distribution function value, and determining scale parameters and shape parameters; substituting the scale parameters and the shape parameters into a failure distribution function in two-parameter Weibull distribution, and determining a failure distribution function model; determining a reliability function; equipment reliability calculation: substituting time intervals between given time and last fault repair time into a reliability function, and calculating equipment reliability; dividing the health interval of equipment according to the obtained equipment reliability; giving suggested equipment maintenance starting and ending time according to the health interval of the equipment.

Description

technical field [0001] The invention relates to the application field of manufacturing equipment monitoring, in particular to a method and system for monitoring equipment life and health based on original fault data. Background technique [0002] With the continuous innovation of high technology, the highly integrated, intelligent and efficient problem analysis and processing capabilities of modern equipment are increasingly enhanced. In contrast, the fault diagnosis, maintenance and reliability of equipment have attracted more and more attention. . [0003] There are many kinds of equipment in the manufacturing industry, and the quantity is huge. The failure or instability of a certain equipment or key equipment will lead to the interruption of the entire production line or the decline of product quality. The purpose of equipment life prediction and health monitoring in the manufacturing industry is to accurately predict equipment failures, formulate efficient maintenance ...

AI Technical Summary

Problems solved by technology

[0005] Chinese invention patent: The failure rate prediction method combined with BP neural network and two-parameter Weibull distribution. The collection of fault data is based on the monitoring data of equipment operation status. The collection of fault data is complex and difficult. Once the fault data fails, it will cause the failure The failure of the method, the applicability of the method model is poor

[0006] Chinese invention patent: The method of determining the optimal decommissioning time of power grid equipment based on the three-parameter Weibull distribution uses the maximum correlation coefficient optimization method to solve the position parameters. The solution process is complicated, which increases the complexity of the algorithm and reduces the operating speed.

[0007] Chinese invention patent: A reliability-based transformer equipment life assessment method is mainly aimed at the life prediction of large-scale transformer equipment in the field of power transmission and transformation. The data algorithm is mainly based on manual experience, and it is applicable to equipment in other fields, especially production and manufacturing process equipment. lower

[0008] The above-mentioned example algorithm has high randomness and high requirements for input data, the algorithm is complex, and the operation speed is slow. It does not classify whether it needs to build a model according to the actual use of the manufacturing equipment, and lacks scientific quantification for health measurement. The boundaries of the equipment failures are not prioritized, and failure maintenance plans are not considered in production planning and scheduling

Images