Fault Diagnosis of Coal Mining Machinery Based on State Parameters

Abstract

The reliable operation of coal mining machinery acts as an important guarantee for safe productions in underground coal mines. The status monitoring and fault diagnosis of traditional coal mining machinery mainly rely on threshold judgments. However, a single judgment condition and a long fault propagation chain can be found in the method of threshold judgments, which make it difficult to accurately seek the fault type. By using the data analysis of state parameters for coal mining machinery, fault parameters and propagation paths can be analyzed effectively. This paper takes the cutting unit of a certain type of bolter miners as an example, a static and dynamic numerical analysis method of the cutting unit of bolter miners are established by virtue of FTA-Petri net models and BP-Firefly neural networks, which can provide a new perspective for fault diagnosis of coal mining machinery.