Parameters Estimation of Automotive Ignition Coils Based on Support Vector Regression

Ignition coils is an important part of automotive ignition system, which directly influences the vehicle’s power performance, energy consumption and environmental protection property. In this paper, a MIMO model based on support vector regression (SVR) is proposed aiming at estimating technological parameters of automotive ignition coils. With the performance parameters, such as spark current and ignition energy as inputs, the model can reach the goal to estimate the technological parameters such as windings and wire diameters of primary coils and secondary coils by choosing appropriate SVR parameters, kernel functions and training algorithm. Furthermore, experimental samples of ignition coils are specially designed, manufactured and measured to obtain training data. Simulated results prove the rationality and accuracy of the model, which shows that the model can provide guidance for the design of ignition coils.