Using Uncertainty Bounds in the Design of an Embedded Real-Time Type-2 Neuro-Fuzzy Speed Controller for Marine Diesel Engines

Marine diesel engines operate in highly dynamic and uncertain environments, hence they require robust and accurate speed controllers that can handle the encountered uncertainties. Type-2 Fuzzy Logic Controllers (FLCs) can handle such uncertainties; however they have a computational overhead associated with the iterative type-reduction process which can diminish the FLC real-time performance. Furthermore, manually designing a type-2 FLC is a difficult task particularly as the number of membership function parameters and rules increase. In this paper, we will introduce an embedded Real-Time Type-2 Neuro-Fuzzy Controller (RT2NFC) which overcomes the iterative type-reduction overhead and learns the parameters of interval type-2 FLC for marine engines. We have performed numerous experiments on a real diesel engine testing platform in which we compared our RT2NFC to a T2NFC based on the iterative type reduction procedure. Both T2NFCs were embedded on an industrial microcontroller platform where they handled the uncertainties to produce accurate and robust speed controllers that outperformed the currently used commercial engine controller. The RT2NFC gave approximately the same control response as the T2NFC, whilst the RT2NFC avoided the type-reduction overhead thus giving a faster real-time response.