| In recent years,with the rapid development of China's economy,the urbanization process is accelerated,the rapid expansion of urban scale,the rapid increase in the number of urban population,existing urban traffic is unable to meet the demands of urban development.Rail transit plays a important role in relieving urban traffic pressure and improving urban traffic capacity.Under this underground,Automatic train operation(ATO)system has been extensively studied.ATO system can improve the efficiency of rail transit and ensure the safety of train operation,Highly efficient ATO system to ease the pressure of urban traffic has a very important significance.This article discusses the function of the ATO system,the train driving strategy,analyzes the relevant parameters of train operation,and a multi-objective optimization model of urban rail train is established.Through the optimization of automatic train driving strategy,under the premise of the train operation safety,to achieve the better effect of punctuality,parking accuracy,comfort and energy saving.Train operation process is a complex nonlinear process,and the traditional control method is difficult to obtain better optimization results.Therefore,this article uses intelligent algorithm optimization,study on the simulation model and intelligent algorithm of the train driving system,to achieve the purpose of speed control with a safe,accurate parking,punctual,comforting and energy-saving orientation.The main contents of this article are the following:(1)Through the analysis of the function of ATO system,summarize the train driving principle and optimization strategy of ATO system,analyze performance index which the ATO system needs to optimize and train dynamics model.On this basis,the mathematical model of train operation index is established.(2)A hybrid algorithm based on particle swarm optimization(PSO)algorithm and improved cuckoo search(CS)algorithm is proposed,which is called PSO-ICS algorithm.The multi-objective optimization mathematical model of train operation is established,and the multi-objective optimization problem is transformed into a single objective optimization problem by weighted summation.PSO algorithm,CS algorithm,PSO-CS algorithm and PSO-ICS algorithm are used to optimize the automatic train operation curve,respectively.By comparing the simulation results,it can be found that the PSO-ICS algorithm has the fastest convergence speed and the best performance in the four methods.(3)Optimization of train operation curve based on unified optimization algorithm of multi-swarm hierarchical structure.The PSO algorithm and the CS algorithm are combined by multi-swarm hierarchical structure,use of information exchange between swarms,use an external file to save the Pareto optimal solution set and guide the entire population evolution.The number of small swarms is discussed and the time complexity of the algorithm is analyzed.The experimental results show that the unified optimization algorithm of multi-swarm hierarchical structure has better convergence and diversity.(4)The chaotic strategy and crowded entropy are used to improve the unified optimization algorithm of multi-swarm hierarchical structure.And the superiority of the improved algorithm is verified by simulation experiment.The Pareto optimal solution set obtained by the improved algorithm is cut and repaired.Remove some non-inferior solution,so that all solutions as far as possible evenly distributed in the solution space.(5)Design the fuzzy controller,the train running optimization curve as the input signal.Through the comparison of the input and output signals,it can be seen that the fuzzy controller can achieve good control effect,and capable of accurately controlling the automatic train operation. |
PDF Full Text Request