A fuzzy logic map-based knock control for spark ignition engines

Knock control represents one of the most critical aspects to reach optimal thermal efficiency in spark ignition engines, and its research is crucially important because it determines thermal efficiency, engine durability, and power density, as well as noise and emission performance. In this paper, a spark advance control based on a map learning technique is combined with a knock estimator to maximize the engine efficiency while keeping the knock probability below a desired limit. The proposed controller is experimentally validated on a production spark ignition gasoline engine test bench, and compared with a conventional spark advance controller in both, steady and transient conditions. From experimental results, a benefit in terms of thermal efficiency, control stability and engine security are achieved. The results show that the proposed method is capable of regulating the knock probability to a target percentage with low spark advance and thermal efficiency dispersion than the conventional controller.