Multiobjective Optimization of a Railway Vehicle Dampers Using Genetic Algorithm

Ride comfort, safety, wear and vehicle speed are the most important factors in evaluating the efficiency of railway transportation. In order to decrease the track access charges it is often desirable to run the vehicle at maximum allowed speed, while keeping an admissible amount of wear in system. This usually deteriorates the ride comfort and safety level during the operation. Therefore, an optimization problem to find a tradeoff value for vehicle speed and design parameters is inevitable. Since, ride comfort, safety and wear values are sensitive to primary and secondary suspensions’ damping parameters it is desirable to find the optimum values of such design variables. In this regard, the multiobjective optimization of railway vehicle dampers is considered to increase the cost-efficiency of railway operation. One car vehicle model with 26 degrees of freedom (DOF) along with a set of initial states, design parameters and operational conditions is explored here. All bodies are assumed to be rigid. Vehicle carbody and bogie frames supposed to have the full set of DOF in space. While, only lateral and yaw motions are considered for each wheelset. It is also assumed that wheelset roll angle is a function of the lateral displacement. Primary and secondary suspensions compromised of parallel linear springs and dampers in longitudinal, vertical and lateral directions which connect wheelsets to bogie frames, and bogie frames to carbody, respectively. Lagrange formalism is employed to obtain the system’s equations of motion. The nonlinear heuristic theory is chosen to relate creepages and the corresponding creep contact forces. The dynamic response of the system is obtained for different operational scenarios including ideal and imperfect tangent and curved tracks. Series-based functions are chosen to approximate the harmonic lateral track irregularities. Accelerations at carbody level, shift forces and wear number are used to evaluate the ride comfort, safety and wear, respectively. MATLAB genetic algorithm optimization routine is applied to perform the optimization. The Pareto sets and Pareto fronts obtained from this study provide the vectors of optimal design parameters corresponding to maximum admissible vehicle speed and guarantee the best tradeoff values for safety and comfort with threshold on wear for each operational scenario. Analysis of the obtained results gives insight into multiobjective optimized dynamic response of a railway vehicle and useful hints for designing adaptive bogie systems with the possibility to switch between optimal damping parameters value and provide the best operational efficiency.