A study on automatic adjustment of the HCCI engine controller using machine learning

An automatic adjustment method of the model-based controller an HCCI engine was designed in this study. As modeling errors are inevitable, feedback control is usually introduced to reduce effect errors. However, especially under transient conditions, performance may deteriorate, because a with information previous cycle. The thought be improved by taking into consideration. Therefore, algorithm adjust input based on prediction error developed. learned and predicted ReOS-ELM (Regularized Online Sequential Extreme Learning Machine), which online machine learning low computational load. conducted every adjusted so that output model including its coincided reference. reference tracking proposed method.