Probability-Constrained Optimal Control of Combustion Engines

This paper summarizes the framework of applying probability-constrained optimal control for combustion engines. The probabilistic constrained control problem is formulated in engine, after which, both the data driven-based methodology and explicit model-based are briefly reviewed to address the problem. Two study cases are given as the explanation for the applications of probability-constrained optimal control in engine system. The first study case is knock probability constrained thermal efficiency optimization. An iterative learning-based control strategy is proposed to solve the problem by combining extremum seeking method and likelihood-based method. On the other hand, scenario approach is applied to solve torque tracking control with combustion phase probabilistic threshold.