Fractional Order ANFIS Sliding Mode Controller for Two-Time Scale Dynamics in PWR

In this research work, a novel model of Pressurized Water Reactor (PWR) dynamics is developed with special emphasis on nuclear fuel burn-up or fuel depletion dynamics. PWR dynamics is identified and decomposed into fast and slow dynamic modes for the first time in this research work. The stiff two-time scale reactor dynamics problem is addressed, and a new sophisticated fractional order two-time scale sliding mode controller is designed. The PWR dynamics is uncertain due to three distinct operating conditions of nuclear reactor core as Beginning of Core (BOC), Middle of Core (MOC) and End of Core (EOC) synthesizing a variable structure model. The model uncertainties are estimated using Adaptive Neuro-Fuzzy Inference System (ANFIS) while different reactivity components are addressed as active disturbance or measurement noise. The novel robust control design problem is a big challenge in this research. The proposed controller is designed, tested and validated against benchmark data and found excellent in performance.