An intelligent parsing method for manufacturing big data health information

Abstract

The invention discloses an intelligent analysis method for manufacturing big data health information. Aiming at fragmentization of historical manufacturing big data, through resampling of multisourcesignals, frequency joint analysis, variable granularity analysis and dimension conversion are used to establish the quality evaluation index, data purification algorithm and deconvolution network of big data in equipment monitoring, so as to realize the purification of big data and the adaptive extraction of complex fault features, in order to realize the intelligent analysis and characterizationtechnology of manufacturing big data health information. Combined with the Deep Learning Theory, a causal relationship between a plurality of health states of a manufacturing process and a monitoringsignal is explored, in order to overcome the congenital defects such as low diagnostic rate and poor robustness of the traditional shallow identification model, the depth identification model of the on-orbit manufacturing process health status is established to describe the health status of the mechanical and electrical equipment in the manufacturing process comprehensively and realize the intelligent identification of multi-mark faults, so as to realize the depth learning and intelligent identification technology of the manufacturing process health status.

Description

technical field [0001] The invention relates to the field of industrial manufacturing big data processing, in particular to an intelligent analysis method for manufacturing big data health information. Background technique [0002] As the market competition continues to intensify, how to improve the competitiveness of industrial electromechanical equipment is the key to the survival of enterprises. The way to improve traditional equipment is to improve product performance and reduce costs. However, as time goes by, the limitations of improving performance and reducing costs become more and more obvious, and innovation is weak. Therefore, how to maintain more mature performance and longer-lasting vitality of existing electromechanical equipment is the direction of future development. Therefore, monitoring the health information of electromechanical equipment is particularly important, and how to intelligently analyze and represent health information requires machine learning...

AI Technical Summary

Problems solved by technology

The improvement method of traditional equipment is to improve product performance and reduce cost, but as time goes by, the limitations of improving performance and reducing cost become more and more obvious, and innovation is weak