Study On Model And Control Strategy Of Energy Management System For SHEV

With the oil shortage and increasing pollution, environmental protection and efficient is becoming more and more important to us. Now, a lot of car manufacturers and researchers are turning their attention to vehicles with new energy, which are higher efficiency and cleaner. The hybrid electric vehicles have both the advantages of electric vehicles and traditional vehicles, such as: good economy, low emissions and long driving range. Therefore, hybrid electric vehicles become the research focus of new energy vehicles. Energy management is of fundamental importance in hybrid electric vehicles, for exploiting the advantages deriving from the availability of a rechargeable energy buffer. With the increasing interest in hybrid electric vehicles and their future commercial availability, energy management is becoming even more important, since energy management system can control engine, battery and motor reasonablely according to the driver's driving intention, optimize the output of each power component, to achieve a reasonable distribution of energy in hybrid electric vehicles, and then improve economy and emission.Currently, in the development of energy management system for the series hybrid electric vehicle, control strategies are mostly based on experience, by setting the threshold to control energy distribution, but can't achieve optimal performance of hybrid electric vehicle. Therefore, in this dissertation a foreword model is firstly built for a series hybrid electric vehicle, then, based on the model, distribution of energy is optimized using dynamic programming (DP) and equivalent fuel consumption minimization strategy (ECMS), so that the optimal vehicle performances are achieved. Then the best performance in theory and optimal control methods are found for a series hybrid electric vehicle, so avoid dependence on the control experience, and provide a theoretical basis for the development of other control strategies. Finally a more effective vehicle controller can be designed. This major work done and conclusions in this dissertation are as follows:In order to analyze the performances of the series hybrid electric vehicle which is developed in a project, and develop energy management strategies, so that the energy distribution to optimized to achieve good power and economy, a foreword model is built for the series hybrid electric vehicle in this dissertation, mainly using a modeling method which combines experimental and theoretical modeling methods. For the main components of hybrid electric vehicle (engine, battery and motor), the corresponding feature curves are respectively obtained from the experiments, then the main variables can be calculated using look-up tables. In addition, some models are simplified reasonably, such as: the wheel model, vehicle dynamics model, driving axle model, vehicle accessories model, driver model and vehicle controller model, then the vehicle model is completed for the development of performance and control strategy of the vehicle to provide a complete platform in the future. For the need of building models, in this dissertation, the main power components of the series hybrid electric vehicle are tested using a test bench, the performance curves and parameters for each model are obtained through the tests. Then by comparing the test results and simulation results, good agreements can be found between them, and this show the correctness of the established model, and the models can be applied for the development of the vehicle control strategy.By analyzing the structural features and operating characteristics of the series hybrid electric vehicle, two optimization algorithms are selected to optimize the energy distribution and performances of the vehicle in this dissertation, and one is dynamic programming optimization algorithm, which is a global optimization method, and can obtain the optimal solution of the optimized object; the other is equivalent fuel consumption minimization optimization algorithm, which is a local optimization algorithm, and can obtain the instantaneous optimal solution. Based on the characteristics of the both optimization algorithms, reasonable methods are found to use them to optimize the performances of series hybrid electric vehicle in this dissertation, and the applied processes of both optimization methods are analyzed. In the series hybrid electric vehicle, there are two energy sources (engine and battery), how to distribute energy outputs for both sources and obtaining optimal vehicle performances is the focus of optimization algorithm and control strategy development. In view of this, the state of energy (SOE), which can reflect the level of battery power status, is presented in this dissertation as an input parameter of control strategy, which is more convenient for energy distribution.In order to do deeper research for the performance of optimization of the dynamic programming and equivalent fuel consumption minimization algorithm, based on the model of the series hybrid electric vehicle, simulation tests are done in this dissertation. The simulation results show that dynamic programming optimization method can find the global optimal control methods to achieve optimal vehicle performances if the states of the vehicle in the vehicle driving cycles can be known in advance. Therefore, dynamic programming can provide the global optimal solution in a numerical way and represents a benchmark for the other strategies.Unlike the dynamic programming optimization method, ECMS does not require knowledge of the entire driving cycle in advance, and can optimize the vehicle performances at each time, so is implementable on-line. However, large amount of calculation is needed, so it is difficult to achieve in controller level. The simulation results show that, although the optimal method of equivalent fuel consumption minimization algorithm is a local optimization method, but because of it letting vehicle achieve the optimal control at every moment, so the ultimate vehicle performance is close to the dynamic planning optimization results, and better than the traditional rule-based hybrid electric vehicle control method. In addition, the simulation results show that for different vehicle driving cycles, the corresponding charge and discharge equivalent factors of equivalent fuel consumption minimization algorithm are different. But as long as vehicle driving situation is similar to the vehicle driving cycle in which equivalent factors are get, then there will be little change in vehicle performances. Therefore, in practical applications the corresponding charge and discharge equivalent factors can be obtained by optimizing different vehicle driving cycles (such as rural, urban and highway) in advance, and predict driving situation by a forecasting model, and use corresponding equivalent factors, so that the vehicle performances can reach close to the optimal results all the time.Based on PowerPC566 singlechip a vehicle control unit is developed for the series hybrid electric vehicle, in which fuzzy logic control method is applied to achieve energy distribution. In the design process of fuzzy control strategy the vehicle's power requirement, battery state of energy (SOE) and the generator output power is determined as the input and output parameters of the fuzzy controller. The domains and their corresponding membership functions of the parameters are determined, and based on previous control experience and the results of two optimization methods fuzzy rules are made and fuzzy controller is designed, then, fuzzy control table is got and stored in the controller's memory, after that, the fuzzy controller can control the vehicle by using look-up table control. By comparing the results of fuzzy controller in the test and optimization algorithms in the simulation, adjust the fuzzy rules in fuzzy controller to make the control effect of fuzzy controller gradually close to the optimal. Finally, the series hybrid electric vehicle is tested on a vehicle test bench, and the control effect of the fuzzy controller has proved its ability to effectively control the series hybrid electric vehicle, and can greatly improve the economy, at the same time the battery can be always maintained at its highly efficient operation region.