Optimal Control Of High-speed Train Tracking Operation Under Different Block Systems

High-speed railway is the main artery for economic development, and it plays a key rolein the regional economic take-off. The services provided by high-speed train (HST) shouldsimultaneously satisfy the multi-objective requirements of security, punctuality, energy savingand ride comfort. However, since the HST is a complex system running in a dynamicenvironment, it is difficult for the manual operation method which depends a given V-S curve(velocity versus position curve) to fulfill the multi-objective requirements. Besides, the signalmodels widely used in railway transportation are fixed block signal system and quasi-movingblock signal system, while the dense tracking operation of HST under moving block system(MB) has been the development tendency of high-speed railway transportation. Aim at theproblem above, the research on the multi-objective optimization of HST, which is based onthe general block signal system (including the fixe and quasi-moving signal systems) andmoving block signal system, are carried out. So as to realize the secure, punctual,energy-saving and ride comfort operation of HST under different block signal system.The running process of HST is nonlinear as well as its running state is closely relatedwith the conditions of signal model, line characteristic and so on. Therefore, according to theactual characteristic of these conditions, the dynamic model, tracking model, linecharacteristic model and mluti-objective model of HST are established, which provide reliabledata for the multi-objective operation optimization of HST. The main content of this paper isorganized as follows:1. In order to describe the force situation of HST, the dynamic model of HST isestablished, which is based on the force analysis of HST’s running process; Cosidering theactual characteristic of the tracking process and operation conditions of HST, we built thetracking model and line characteristic model of HST, so as to provide a quantitative basic forthe optimization; According to the multi-objective requirements, we also propose a model forthe multi-objective operation optimization of HST.2. With the advantages of efficiency and applicability, the multi-objective particle swarmoptimization algorithm (MOPSO) is employeed to conduct the multi-objective optimization ofHST. Meanwhile, preference information is added to the algorithm to further improve theefficiency of MOPSO, so as to realize the optimization as well as obtain the optimal operationstrategy for HST.3. To validate the effectiveness of the proposed model and method, experiments onmulti-objective operation optimization of HST are carried out with the field data ofCRH380AL (the high-speed train of type-380AL). For the situation that speed restriction mutation appears during the running process of HST, experiments on the three running statesof serious delay, little delay and no delay are performed; For the dens tracking operation ofHST, the preferences of operation sensitivity and energy conservation are added to theMOPSO, so as to verify the efficiency and convergence of the proposed method. In order tovalidate the efficiency of the modeling method in this paper, according to the data sample ofCRH380AL EMUs in the actual operation process, establishing the models of EMUs operationprocess, then simulation and analysis for the parking accuracy, safety and energy consumption.Experimental results demonstrate that the models established in this paper could describethe actual characteristic of the operating process of HST accurately. Besides, it is alsovalidated that the optimal operation strategy obtained by the proposed method could fullfil thesecure, punctual, energy-saving and ride comfort operation of HST.