RECOGNITION OF BOLT QUALITY BASE ON ELMAN NEURAL NETWORK BY ANT COLONY OPTIMIZATION ALGORITHM

The quality and working condition of the bolt anchoring system determine the safety performance of the whole construction project to a large extent. In order to ensure the effect of the bolt in the support system engineering and prevent the occurrence of major disasters, testing of the bolt quality becomes particularly important. The paper presents a way of identifying the bolt quality that use ant colony algorithm to optimize Elman neural network. The weights and thresholds of Elman neural network are optimized by using the ant colony algorithm and the Elman neural network model to recognition bolt quality is established. The results show that the Elman neural network by ant colony algorithm has better recognition effect and higher prediction precision than the Elman neural network and the Elman neural network by genetic algorithm.