Font Size: a A A

Control Strategy For Hybrid Electric Tracked Vehicle Based On Stochastic Dynamic Programming

Posted on:2016-05-07Degree:MasterType:Thesis
Country:ChinaCandidate:S S XiongFull Text:PDF
GTID:2272330452965132Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
With the development of Electro-mechanical transmission and electric transmissiontechnology, Electro-mechanical transmission tracked vehicles developed countries competeto become hot. Taking aElectro-mechanical transmission tracked vehicle for the study,carried out to expand its power distribution for control technology, the stochastic dynamicprogramming method conducted in-depth theoretical study designed stochastic dynamicprogramming power control strategy.Based on the performance requirements of Electro-mechanical transmission trackedvehicles to determine the power distribution control strategy optimization goal: theeconomy, power and electrical performance, multi-objective optimization problem. Trackedvehicle conditions for statistical analysis, it can be divided into low, medium, high speedconditions. And in accordance with the actual needs of different conditions, the use of AHPto determine the weight of each weight coefficient optimization goals, the multi-objectiveoptimization problem into a single objective optimization problem under the samedimension.Electro-mechanical transmission system dynamics is analyzed.Based on the MATLABsoftware platforms, such as Simdriveline and Simpowersystem vehicle driveline modelingand electrical system modeling toolkit, a physical model of the vehicle is builded; Theauthor adopts SDP approach by modeling driver’s power request as aMarkov chain. Andobtain electricity and driving power, respectively, by the maximum likelihood estimate ofthe transfer matrix method, the dual Markov models establishment of a driver’s drivingdemand for power and electricity demand power.With stochastic dynamic programming as the theoretical basis for the powerdistribution control strategy for global optimization.The states were vehicle speed,Enginetorque, battery’s state of charge () and the generator speed; the control variable wereEngine speed and motor torque, the randomwas power demand;and in the economy, powerand generation capacity for integrated optimization objectives.Stochastic dynamicprogramming model for power distribution controlwas established. The SDP problem canbe solved bynew policy iterationalgorithm,witch was based on value iteration and policyiterationalgorithm, and obtained optimal control strategy. The use of Embedded MATLAB functions can be programmed to establish astochastic dynamic programming model of the power distribution controller, built afeed-forward Electro-mechanical transmission tracked vehicle simulation model,stochastic dynamic programming control strategy simulation. Comparison of existingcontrol strategies based on logical rules, shows that for random multi-objectiveoptimization problem, using stochastic dynamic programming power distribution controlstrategy can improve the overall performance of the vehicle.
Keywords/Search Tags:Tracked vehicles, Electro-mechanical transmission, power distribution, stochastic dynamic programming, multi-objective optimization
PDF Full Text Request
Related items