A State Monitoring Method of Roadheader Cutting Arm Based on BP Neural Network

Abstract

State monitoring is usually carried out for fault diagnosis of machinery that works in severe environments or situations. This paper applies the BP neural network novelty in the state monitoring of the cutting arm on the roadheader in the purpose of improving efficiency and accuracy of corresponding fault diagnosis the training method is Fast BP Method. To evaluate the fore-mentioned proposal, the real state monitoring data of a working roadheader were collected, the real data is analyzed by MALTAB, multiple sets of eigenvectors were extracted and a signature database was established by analyzing the real data, followed by fault diagnosis based on BP neural network and the simulation results are analyzed at the end of the text. The diagnosis results show that our proposal gives high efficiency and accuracy, which presents its feasibility and practicability evidently.