Optimization of AP1000 power control system setpoints using genetic algorithm

The aim of this work is to obtain the optimum controller parameters in advanced mechanical shim (MSHIM) control system of AP1000. The MSHIM control system is designed to allow the AP1000 nuclear steam supply system (NSSS) to perform power change operations automatically. For the MSHIM control system, the lead-time constant and lead/lag ratio in the average coolant temperature (Tavg) controller, and rate/lag constant and low gain of nonlinear gain unit in the power mismatch controller are considered decision variables through the sensitivity analysis when minimizing the overshoot in the nuclear power and maximum absolute deviation between average coolant temperature and its target value. The multi-objective problem with four decision variables is translated into two subproblems. The lead-time constant and lead/lag ratio in the Tavg controller are firstly optimized by non-dominated sorting genetic algorithm (NSGA-II) with the power mismatch controller defeated. Then rate/lag constant and low gain in the power mismatch controller is optimized by NSGA-II with the optimized parameters in the Tavg controller. Dynamic simulations are performed based on a fast-running NSSS Control & Analysis Platform (NCAP) in each iteration of the optimization process to calculate the objective functions. The 10 percent step load increase transient from 90% to 100% full power is selected as simulation case for the optimization. The calculation results demonstrate that much better reactor control capabilities can be provided by MSHIM control system when employing the optimized control parameters.