Search Results

The purpose of this paper is to use a multilayer perceptron neural network model to identify and control a non-linear electronic throttle system. The neural network model, which represents the dynamic behaviour of the non-linear throttle servo system, was first identified at different operating conditions. The neural network controller model was then designed (or trained) with the throttle identifier network model, so that the tracking control position of the throttle system follows a reference model. The neural network controller training is computationally expensive and requires the use of the dynamic backpropagation algorithm, which is significantly time consuming during on-line implementation. For this reason, the throttle identifier network model is used to assist in training the neural controller in off-line mode. The neural network controller was trained with the same inputs that are fed to the actual throttle system to produce the same output.

This paper describes the approaches, possibilities and benefits of using artificial Neural Networks for Engine Management Systems with the focus on new calibration methods and control strategies. Utilizing this approach in the calibration and function algorithm development process makes it possible to fulfill shorter development time frames for series production intended projects. Additionally, this technology helps to develop new control and diagnostic strategies for new Engine Management Systems to meet the upcoming stricter emission and OBD regulations.

This paper presents another application [1] of using Artificial Neural Networks (ANN) in adaptive tracking control of an electronic throttle system. The ANN learns to model the experimental direct inverse dynamic of the throttle servo system using a multilayer perceptron neural network structure with the dynamic back-propagation algorithm. An off-line training process was used based on an historical set of experimental measurements that covered all operating conditions. This provided sufficient information on the dynamics of the open-loop inverse nonlinear plant model. The identified ANN Direct Inverse Model (ANNDIM) was used as a feed-forward controller combined with an adaptive feed-back gains (PID) controller scheduled [2] at different operating conditions to provide the robustness in tracking control to un-modeled dynamics of the throttle servo system.