Computational methods for management of hybrid vehicles

As the electronic engine technologies advances, the engine management tool has drawled attention for being one of the main efficiency improvement methodologies and for being applied on Diesel and Otto cycle engines. With the creation of hybrid vehicles there comes the need to manage several engines simultaneously in order to optimize the energy consumption and to reduce the waste emission, among other improvements. To accomplish this objectives are necessary to adapt this system to the driver's needs and to improve its controls. In order to doing so, we propose an intelligent approach for controlling this managing system using of artificial and computational intelligence techniques such as Bayesian Nets, Neural networks and Genetic Algorithms. The intent of using these self-improving learning techniques is to improve the system during the time it is being used, adapting it to have better performance in situations such as obtaining maximum torque, optimizing maximum velocity or reducing fuel consumption.