| Train running process costs much more energy than other parts of rail transit. Most of the recent researches on train optimal operation were based on the fact that the problem is served as a single objective optimization problem, which minimize the energy consumption with the constraint of safety and time. This paper maintains that the operation of train is a multiple objective optimization problem. The problem needs to satisfy safety, economic, punctual, comfortable and precision stop simultaneously. Such optimization problems sometimes do not exist an absolutely optimal solution, but there exists several Pareto optimal solutions.First, this paper makes an analysis on train's kinematics. Then with considering the combination of driving strategy and regenerative braking, the paper build a multi-objective optimization model, which serve energy consumption, punctuality and comfort as the goal with the block speed limit constraint. When the train running with a stated strategy in the line which has several different speed limits, the speed of train may exceed the speed limit while turn to the next block. So, this paper analysis the possible situations when the trains turn into the next speed limit block with making a mode switch strategy of train operation. And the optimal model has been modular programmed base upon Matlab.The optimal mode was covert to a single-objective problem by trade off tree objectives of the model, then SGA was applied in solve it. And multi-objective GA was applied to solve the multi-objective optimal model.Finally, simulations based upon two different lines were made to confirm and analysis the performance of two algorithms. The results present that both algorithm have good performance on energy consumption, punctuality and comfort, and can adapt different line conditions. The SGA method provides a faster convergence. The MOGA has a low dependence of parameter; According to the demand of policymaker it can choose a optimum solution more flexible.
|
PDF Full Text Request