Equipment failure prediction method under cloud manufacturing

Abstract

The invention discloses a method for predicting equipment failure under cloud manufacturing. The operation steps of the method are as follows: 1. Establishing a fault code comparison table according to the equipment composition structure and working principle; 2. Establishing a fault data model; 3. Mining association rules. Adopt Socket (through a two-way communication connection to realize data exchange, one end of this connection) transmission mode, use intermediate equipment to integrate fault data and abnormal parameters of different manufacturers and different equipment into the database, and use HDFS (Hadoop Distributed File System) Distributed storage to solve the problem of big data processing and storage; use Java programming (an object-oriented programming language) to realize the improved Apriori algorithm (frequent itemset algorithm for mining association rules) of Map / Reduce (parallel computing model), and add timing The concept mines association rules for equipment failure characteristics, predicts equipment failure trends, and achieves the purpose of equipment failure prediction. The present invention takes a cutting machine as an example to provide a specific implementation.

Description

technical field [0001] The invention discloses a method for predicting equipment failure under cloud manufacturing, belonging to the field of equipment failure prediction. Background technique [0002] With the rapid development of modern technology and industrial technology, especially information technology, the complexity, integration and intelligence of manufacturing equipment are constantly increasing. With the development of equipment, the cost of its development, production, especially maintenance and guarantee is getting higher and higher. At the same time, due to the increase of components and influencing factors, the probability of failure and function failure is gradually increasing. Therefore, how to early warning of equipment failure and how to control the components or factors that may fail in advance, that is, failure prediction, has gradually become the focus of researchers. Focus. Regarding equipment failure prediction, predecessors have carried out a lot ...

AI Technical Summary

Problems solved by technology

There are also many studies on the parallelization of association rules. For example, the China Mobile Research Institute has developed a parallel data mining software BC-PDM using the cloud computing platform Hadoop, which realizes reliable storage and efficient mining of massive data. There is no research on fault feature analysis, running parallel thought to mine association rules between faults for fault prediction

Images