Parameter Identification of Train basic resistance Based on Particle Swarm Optimization

Train basic resistance parameters (BRP) are important foundation for train traction calculations. These parameters are time-varying with wheel rail wear and weather conditions, which resulting in the calculation model inconsistent with the real model. Therefore, this paper proposes a train basic resistance parameters identification method (BRPI) based on particle swarm optimization (PSO) algorithm. By sampling train operation data, the fitting speed curve of each set of parameters is obtained under same control sequence. Taking speed fitting as objective function, PSO algorithm is used to identify a set of basic resistance parameters with highest fitness. In order to eliminate the defect of easily falling into local solution, a modified speed update strategy of PSO algorithm is proposed. Case studies are used to verify the effectiveness of the proposed method, results show this method has fast convergence rate and excellent solution speed, and it effectively improve the accuracy of train controller.