Optimizing Linearity of the Throttle Position Sensor Based on RBF-CMGA

Abstract

In order to discuss linearity of throttle position sensor. Based on the synergetic theory of the radial basal function neural network (RBFNN) and the contractive mapping genetic arithmetic (CMGA), the throttle position sensor structural parameters, such as, the length and radius of the primary coil, prejudicial distance and radius of prejudicial disc, primary exciting current were optimally forecasted by RBF-CMGA. Several analogue test of diesel engine are made on the basis of throttle position sensor control system of adding loads and decreasing loads. The optimal parameter has been calculated on the well-balanced state of throttle position sensor, and it was very good in the aspect of both dynamic state that is 0.25×10-3 sounds and linearity is 0.4% in the throttle position sensor control system of diesel engine, which can meet the needs of well-balanced state of diesel engine, It is feasible that the throttle position sensor linearity and structural parameters are optimized by RBF-CMGA.