Multi-Objective Optimization For The Operation Of Multi-Locomotive Traction Heavy-Haul Trains

As an important development direction of railway freight transportation in China,heavy-haul transportation is receiving more and more attention from experts and scholars.With the expansion of heavy-haul train formation,the length and weight increase accordingly,and the train operation has become complicated and difficult.At present,the heavy-haul train Driver Advisory System(DAS)and Automatic Train Operation(ATO)system have become a trend of development,and the relevant train optimization methods need further study.The optimal operation of heavy-haul trains is to optimize the control methods and speed curves of heavy-haul trains under various scenarios,considering gentle slopes and steep downhills,taking into account the optimization objectives of energy saving,punctuality,safe and smooth operation of trains.In this thesis,the optimal operation methods of heavy-haul trains are carried out by combining theoretical analysis with simulation verification.The main contents are as follows:(1)According to the analysis of train dynamics principles,the single-particle model and multi-particle model of heavy-haul trains are established,and the multi-particle model is simplified linearly.The simplified multi-particle model is simulated by MATLAB,and the feasibility of the model is verified.(2)Based on the MPC-PI(Model Predictive Control-Proportional Integral)cascade control algorithm,the multi-objective optimization of train energy consumption,coupler force and speed tracking deviation is realized.Based on the data of Daqin heavy-haul line,the MPC-PI algorithm is adopted to simulate the train operation under different weights of the optimization objective function.The simulation results verify the effectiveness of the proposed control algorithm.(3)Based on the improved ant colony algorithm,the optimal operation curves of heavy-haul trains are obtained,considering the three heuristic factors:energy consumption,running time and acceleration.Different optimal operation curves are generated by setting different running times.Based on the data of Daqin heavy-haul line,the simulations of the original and the improved ant colony algorithms are compared and analyzed,and the results show that the improved ant colony algorithm has better convergence than that of the original one.(4)Considering the long and steep downhill section,the braking characteristics of heavy-haul trains are analyzed,and two braking strategies are proposed for the process of cyclic braking.One is based on double air brake pipes.The method of cyclic braking under double brake pipes is designed and implemented,which effectively improves the operation smoothness of the train.Using the data of long and steep downhills of Daqin heavy-haul line,the simulation comparison between the existing single air brake pipe and the designed double air brake pipes is carried out.The results show that the braking strategy based on double air brake pipes effectively improves the smoothness of train operation.The other cyclic braking method is based on fuzzy control under the existing single air brake pipe system.According to the actual train operation data of Shenshuo heavy-haul line,the optimal control rules corresponding to different running times are formulated,and the dynamic adjustment of cyclic braking strategy is realized.The simulations based on the data of a long and steep downhill of Shenshuo line show that the method can realize the optimal control of heavy-haul train cyclic braking,and the simulation data is better than the manual manipulation data.