Optimization Research On Recommended Speed Curve Of Urban Rail Train Based On Improved Artificial Bee Colony Algorithm

At present,urban rail transit is still in a period of rapid development,but with it comes increased energy consumption and transportation costs.Therefore,in order to implement to the basic national policies of environmental protection and energy conservation,it is necessary to achieve the energy-saving and emission reduction of urban rail transit system.The train traction energy consumption is the main part of the energy consumption generated by urban rail transit system.Therefore,the energy saving and emission reduction of urban rail transit system can be realized by reducing the train traction energy consumption.Automatic Train Operation(ATO)is an important embodiment of urban rail transit intelligence and automation.ATO realizes automatic operation between stations by automatically adjusting train speed.ATO control structure includes an optimization layer and a control layer,the optimization layer calculates the optimal recommended speed curve according to a certain algorithm,and the control layer tracks the recommended speed curve according to a certain control strategy to control the train for traction or braking.And this tracking is closely related to the train traction energy consumption.Therefore,optimizing the recommended speed curve is an important breakthrough to realize train energy conservation and emission reduction.Aiming at the shortcomings of most existing swarm intelligence algorithms for the optimization of the recommended speed curve,this thesis presents two improved artificial bee colony algorithms with stronger searching ability and better convergence.In the process research,the actual situation is fully considered,the train tracking operation control strategy and speed curve related performance indicators are taken into account,the line data is also discretized,thus an optimization model is established.Finally,the effectiveness of the proposed method is proved by the simulation analysis of some station data of Beijing Metro Yizhuang Line.The main research contents include the following points:(1)Firstly,the relevant theories of ATO system are introduced,including its working principle,basic functions,two-layer structure,etc.According to the relevant kinematic principle,the force and operating condition of the train in the process of running are analyzed in detail,the operating condition conversion principle is given,and the train kinematic model is established according to Newton’s second theorem.At the same time,the relevant performance indications of train speed curve are introduced,which lays a theoretical foundation for the subsequent research.(2)Secondly,the basic model,specific process,advantages and disadvantages of conventional ABC(Artificial Bee Colony,ABC)algorithm are introduced.Aiming at the shortcomings of conventional ABC algorithms,two improved ABC algorithms are introduced from the initial model,leading bee model,following bee model and other basic models.One is the PABC(Pbest-guided ABC)algorithm,which uses opposition learning to generate the initial population in the initial stage and local optimal solution to guide the mechanism in the following bee stage.The other is Opposition-learning Adaptive Quick Artificial Bee Colony(OAQABC)Algorithm.Sphere function and Griewank function are used as test functions to verify the performance of PABC algorithm and OAQABC algorithm respectively.(3)Then,the optimization problem of train speed curve is transformed into the problem of solving the optimal path in speed-distance two-dimensional space.On this basis,starting from the actual situation,some strategies related to the optimization of the recommended speed curve,such as line discretization and equivalent slope,are designed.Based on the above theoretical analysis and traction energy consumption calculation theory,an optimization model is also established.The optimization strategies and optimization model are combined with PABC algorithm and OAQABC algorithm respectively,and the specific flow of solving the optimal recommendation speed curve by the two algorithms are given.(4)Finally,MATLAB is used as a simulation tool,and the actual data of Yizhuang line are used to simulate and verify the optimization effect of the two improved ABC algorithms on the optimization model,the effectiveness of the optimization methods and optimization strategies proposed in this paper is analyzed and illustrated.