Predictive Control Using A Distributed Model For Electric Multiple Unit

China generally uses the distributed power electric multiple unit (EMU), and itcomposed by several trailers which have no power plsnts and motor vehicles which havepower plsnts. With the growing of moving speed for high speed train, the dynamicenvironments of EMU take place the essential changes, and the interaction between the airand EMU has become very intense, so the aerodynamics of EMU are severe deterioration.According to the complex nonlinear characteristics of EMU, The traditional modeling andcontrol methods are difficult to get satisfactory control effect in a certain extent. Therefore, Inview of the structure features for distributed power EMU, and based on the traction/brakecharacteristic curve and the actual operation data, this paper employ data-driven modelingmethod to establish the distributed state-space model on the longitudinal of EMU. Then,based on the distributed state-space model, we design the model predictive controller for it.Main contents of the paper are summarized as following:1. According to the running process of EMU, we Analysis the dynamic model and thebasic mechanical in a horizontal direction of EMU. We also provide the computationalformula and the factors to affecting the basic mechanical. Then we analyze the structurecharacteristics of EMU and point out the deficiency for the centralized model of EMU. It laida theoretical basis for the following distributed model:2. According to the structure features for distributed power EMU, Firstly, a distributeddescription for the EMU operation process is presented. Based on the traction/brakecharacteristic curve and the actual operation data, this paper employs subspace identificationmethod to establish the distributed state-space model for each power unit. In order to verifythe effectiveness of the model, this paper propose the predictive control method which isbased on the distributed model of EMU, and give the synchronization tracking controlalgorithm for traction/braking force and running speed of each power unit.3. In order to verify the validity of the proposed methods, this study targeted is the CRHEMU type-380AL which is running on the Beijing-Shanghai high-speed railway. Thesimulation results show that the proposed methods can make the running speed EMU to highprecision track the given trajectory, and can real-time response the dynamic changes of EMU.It can satisfy the security, energy efficiency, comfort, on-time operation and other controlobjectives.