Particle Swarm Optimization-Based Direct Inverse Control for Controlling the Power Level of the Indonesian Multipurpose Reactor

Abstract

A neural network-direct inverse control (NN-DIC) has been simulated to automatically control the power level of nuclear reactors. This method has been tested on an Indonesian pool type multipurpose reactor, namely, Reaktor Serba Guna-GA Siwabessy (RSG-GAS). The result confirmed that this method still cannot minimize errors and shorten the learning process time. A new method is therefore needed which will improve the performance of the DIC. The objective of this study is to develop a particle swarm optimization-based direct inverse control (PSO-DIC) to overcome the weaknesses of the NN-DIC. In the proposed PSO-DIC, the PSO algorithm is integrated into the DIC technique to train the weights of the DIC controller. This integration is able to accelerate the learning process. To improve the performance of the system identification, a backpropagation (BP) algorithm is introduced into the PSO algorithm. To show the feasibility and effectiveness of this proposed PSO-DIC technique, a case study on power level control of RSG-GAS is performed. The simulation results confirm that the PSO-DIC has better performance than NN-DIC. The new developed PSO-DIC has smaller steady-state error and less overshoot and oscillation.

Particle Swarm Optimization Based Probabilistic Neural Network for Classification of Severe Accident of Nuclear Reactor

“…In nuclear area, PSO has been used to investigate to the nuclear reactor reload optimization problem [15], PWR power distribution flattening, and critical heat flux prediction [16]. Combination of backpropagation neural networks and PSOhas been investigated in development of nuclear reactor control [17].…”