Modular fuzzy control for nonlinear multipoint based core model of TMI PWR under variable fuel cycle conditions

Abstract

This paper provides design of modular fuzzy controller based on reactor power for multipoint kinetic model of TMI PWR, under transients and variable operational fuel cycle conditions. The validated four-point kinetic model is developed considering six groups of delayed neutron precursors, reactivity feedbacks and diffusion among selected nodes. Fuel burn-up effects and power variations are modeled using the variable thermal coefficients. The ability of adaptive multi-module fuzzy controller for reactor power is evaluated and compared at start and end of burn-up cycle. The power dependant nonlinearity of reactor core and time dependant variations of reactivity characteristics are handled with modular approach of TS fuzzy control. The performance monitoring and control improvements are achieved by online parametric identification of process model. The four modules are distributed on the power base line using the triangular mapping. Design of fuzzy controller is based on the optimized gain requirement at different power levels, error and error change. Control output is mapped as biased linear function of selected input variables using TS type fuzzy logic inference. Safety analysis is performed with Lyapunov function criteria for multipoint model. Transient analysis is performed for power tracking and accident handling ability of proposed control design and compared with the PID controller. The power variation among the adjacent nodes is reduced up to 5% under load following transients with the proposed model. Modular fuzzy approach found proficient in handling the changing operational conditions and variable dynamics of the system.