Optimization of anti-interference ability of the bridge parameter identification algorithm

For the current modal parameter identification method, as the accuracy is not high and the antiinterference ability is not strong in the bridge parameter identification, this paper proposes an improved BP neural network model based on artificial bee colony. Firstly, the particle swarm based on artificial bee colony algorithm is introduced to initialize each parameter of the algorithm, then the Boltzmann selection strategy is adopted to select the optimal solution, and then the improved artificial bee colony algorithm is applied to BP neural network algorithm in order to improve the accuracy of BP neural network algorithm in the bridge parameter identification and anti-disturbance ability. Simulation results show that the proposed improved BP neural network model based on artificial bee colony has higher training accuracy and better anti-interference ability compared to other algorithms. © 2014 CAFET-INNOVA TECHNICAL SOCIETY. All rights reserved.