Neural Network Model of Assessing the Technical Condition of the Power Unit of a Hybrid Vehicle

Abstract

The paper discusses a neural network model of assessing the technical condition of a hybrid vehicle based on energy indices. The model enables to realize the service life capabilities of the power unit, eliminates premature repairs, reduces the number of downtimes during repair and maintenance, and enables to determine the optimal service life of a hybrid vehicle. In the paper is presented the developed method of adaptation of the technical condition management strategy of the hybrid power unit to change the external conditions of the vehicle operation on the basis of neural network and neuro-fuzzy approximation of control laws when implementing the concept of reinforcement training. This approach eliminates the lack of a priori information about the mode of motion in these operating conditions, as well as errors of mathematical models due to the full use of current information.