Research On Adaptive Improved ECMS Control Strategy Based On Condition Recognition

In recent years,the haze is deeply stinging people's nerves.The issues of energy and environmentare urgent.At present,people generally believe that the technology of vehicle energy-saving is one of the most effective solutions in all means.Because of the long driven distance and do not rely on infrastructure,hybrid electric vehiclesis welcomed by public.The hybrid system has been occupying three seatsin the "Ward Top Ten engines" from the year of 2016."The 13 th Five-year Plan" clearly statethat,by the year of 2020,all the production of fuel vehicles must be "micro Mixed ",we can see the application of hybrid technology is becomingpopular.Based on the "National Natural Science Foundation of China"(NSFC),this paper proposes a minimum adaptive control method which based on the conditional improvement of the instantaneous equivalent fuel consumption,aiming at the optimal fuel economy and the real-time application of the on-line.This paper analyzes the mathematical models of the engine,motor and power battery in detail.After this,make a secondary development based on the ADVISOR simulation software,and uses the method of theoretical modeling and experimental modeling to construct the simulation platform.Which provides the basic theory for the follow-upresearch andverification.Secondly,the global optimization control strategy based on dynamic programming(DP)theory is designed to simulate the vehicle fuel economy under the condition which the SOC is maintained.Then we analyzethe vehicle extreme value of fuel consumption,and to obtain the working conditions and parameters of the dynamic components.Follow by these,The basic theory of instantaneous equivalent fuel consumption minimum strategy is introduced,and the equivalent factors of ECMS are optimized by using the working parameters of dynamic components under global optimization.Then the optimal ECMS control strategy is proposed,and the simulation results are compared with the traditional rule control strategy and the traditional ECMS control strategy under the NEDC condition.The results show that:compare with the conventional control strategy,I-ECMS' fuel consumption is improved by 10.83%,then compare the traditional ECMS,the fuel consumption is improved by 3.21% and the SOC is improved by 1.48%.As the optimal result of I-ECMS strategy is only for a single condition,in order to improve the adaptability of the strategy under complex conditions,it is necessary to adjust the equivalent factor in real time according to the different operating conditions of the vehicle.So working condition on-line adaptive recognizer is designed by the working conditions classification and pattern recognition: Firstly,sixteen kinds of standard cycle conditions are used to form the sample condition database,and the different characteristic parameters are selected.These parameters are used to classify the sample conditions into six typical cases,and calculate the each optimal equivalent factors which under the typical conditions.Then design the neural network identification to do pattern recognition.After the fully training,the accuracy rate of the recognizer is 98.8%.Finally,the classification and pattern recognition are introduced into the I-ECMS strategy,to grasp the actual driving conditions of the vehicle operating parameters and put the parameters in the mode identifier,so the current traffic conditions are identified as a typical representative of a class.And then use the equivalence factor corresponding to this kind of working condition as the optimization parameter to input the I-ECMS control strategy,the adaptive ECMS control strategy(A-I-ECMS)based on condition recognition is established.The results of joint simulation show that compared with I-ECMS,the fuel consumption of A-I-ECMS strategy is improved by 4.18% and reduce 43.26% of the SOC's fluctuation.It is proved that the A-I-ECMS control strategy has a good control effect.