The Fault Diagnosis of Garbage Crusher Based on Ant Colony Algorithm and Neural Network

Abstract

The garbage crusher is one of the important parts in recoverable coal production line. To diagnose its faults during the working process, Back Propagation algorithm is used. However, it has some shortcomings, such as low precision solution, slow searching speed and easy convergence to the local minimum points. To overcome this problem, a novel method which integrates Back Propagation neural network (BP NN) and Ant Colony Algorithm(ACA) is proposed in this paper. ACA has the advantages such as positive feedback, distributed computation and using a constructive greedy heuristic. In this paper, ACA is used to train the weights and the thresholds of BP NN, so the searching speed and the precision can be improved. An case study is given. The result shows that the proposed method improves the training efficiency and the fault classification accuracy.