Fault Diagnosis of Coal Electrical Shearer Based on Quantum Neural

Abstract

An improved quantum neural network model and its learning algorithm are proposed for fault diagnosis of the coal electrical haulage shearer in order to on line monitor working states of the large mining rotating machines. Based on traditional BP neural network, the three-layer quantum neural network is constructed to combine quantum calculation and neural network for the error correction learning algorithm. According to the information processing mode of the biology neuron and the quantum computing theory, the improved quantum neural network model has the ability of identifying uncertainty in fault data classifications and approximating the nonlinear function for different fault types to monitor the electrical motor voltage, current, temperature, shearer location, boom inclination, haulage speed and direction in the coal electrical cutting machines. The theory analysis and simulation experiment results show that the control performances and the safety reliability of the coal shearer are obviously improved, while the quantum neural network model is applied to the nonlinear feature fault diagnosis of the coal electrical haulage shearer.