Research On The Control Scheme And The Speed Regulation Method Of The Economic Cruise Based On The Neural Network

With the annual increasing amounts of China’s highway mileage and automobile demands, the energy shortages have become serious. The present research on the automotive adaptive cruise control technology mainly focuses on enhancing driving comfort and driving safety. But there is none of the cruise control system whose controlling target is the fuel economy. So this paper puts forward a new style of cruise control research based on the fuel economy, this cruise mode can reduce the fuel consumption of the vehicle as much as possible, besides it can also better achieve the energy-saving and emission reduction effect.The paper first studies the theory model of the economic cruise control; it includes the vehicle longitudinal dynamic model, the fuel economic model and the vehicle safety distance model. Secondly, the paper studies the control scheme of the vehicle economic cruise and designs the corresponding hardware device, it completes the type selection of the main control unit, the sensors and the execution structure. Thirdly, this paper contrasts and analyzes several methods of calculating the engine fuel consumption rate which conclude the interpolation, the multivariate nonlinear regression and mathematical statistics and the regress equation of the least squares3times matrix in allusion to the engine universal characteristic data. And put forward to a new method of identifying the engine universal characteristic based on the neural network, and applies two different kinds of neural networks which include the RBF network and the two layer neural network based on the LM algorithm to research the model of the engine universal characteristic, contrasts and analyzes the identification results, establishes the model of the engine universal characteristic based on the two layer neural network who applies the LM algorithm. And then studies the self learning function of the engine universal performance neural network model, analyzes the self learning method of the neural network model and the storage and extraction methods of the engine universal characteristic parameter. Finally analyzes the speed control method of the economic cruise control strategy in application of the MATLAB simulation, includes the vehicle constant acceleration simulation based on the road shape, the vehicle variable acceleration simulation based on the road shape and the constant speed control simulation based on the road shape, and verifies the feasibility of the economic cruise control theory through the simulation results.