Mutipower System Energy Management Stratagy Research For Extended-range Electric Bus

At this stage battery technology has limited the further development of pure electricvehicles, so the extended range electric vehicle as a long driving range, relatively lowmanufacturing costs of new energy vehicle, becomes a promising transitional type tomeet current demand.This thesis studies mainly the Multi-power sources energy management strategy aboutExtended Range Electric Bus (E_REB). At first E_REB power system parameter ismatched and it’s models are built, followed by a series of studies on E_REB multiple powersource system, from the traditional vehicle control algorithm to the global optimization ofenergy management, then to the local optimization of real-time strategy, the main contentsof this paper are as follows:1) According to E_REB’s structural characteristics and the design target, the paperproposed E_REB’s vehicle power system parameter matching method, On the basis of this,the parameters for the power source system, driving system and transmission systemare matched, and then completed the design of the vehicle’s overall program.Forward simulation model of extended range electric vehicle is established, and provide asimulation platform for E_REB multi energy control strategy design and verification.2) This thesis applies a hierarchical, modular design idea, from the point of energy andsignal flow, the vehicle control system is conducted a hierarchical division. By dividing thecontrol function into blocks, we decompose the complex control system into relativelysimple subsystems. The top layer is the vehicle control module，the second layer is rangeextended device control and battery control module, the bottom layer is the drive controlsubsystem, brake energy recovery subsystems etc.. By this control system structure, theeffective management of system function can be realized.3) The control strategy of E_REB based on fuzzy rules is studied. Proposes designidea of a double layer fuzzy controller with three input variable to CD controller, itspurpose is to balance and optimize the two indexes of E_REB driving range and fueleconomy, to reasonable consume the battery energy in the mileage range. And theeffectiveness of the strategy is verified for the vehicle control through the simulation, whichis better than the traditional charge depletion and charge sustains strategy.4) The energy management strategy of E_REB based on the global optimization is researched on. A deterministic dynamic programming (DDP) algorithm based ondetermining model and a stochastic dynamic programming (SDP) algorithm based onstochastic model are proposed. Based on the specified SOC, and mileage range conditions,the global optimization results are calculated. the optimized results can provide optimumreference for the control strategies developing of transient or local optimization.5) On the basis of stochastic dynamic programming algorithm, a model predictivecontrol algorithms is designed and developed. The thesis applies Markov stochasticprediction model to predict the power demand of driving cycle, to convert the SDPalgorithm into micro scale optimization algorithms within the scope of sight, then carryon rolling optimization, local solving online, to get globally suboptimal strategy that can berealized in real time.