PARTICLE SWARM OPTIMIZATION BASED PROBABILISTIC NEURAL NETWORK FOR CLASSIFICATION OF SEVERE ACCIDENT OF NUCLEAR REACTOR

Yoyok Dwi Setyo Pambudi (National Nuclear Energy Agency of Indonesia)

Jurnal Teknologi Reaktor Nuklir Tri Dasa Mega
Vol 23, No 3 (2021): October (2021)

Due to its danger and complexity, the identification and prediction of major severe accident scenarios from an initiating event of a nuclear power plant remains a challenging task. This paper aims to classify severe accident at the Advanced Power Reactor (APR) 1400, which includes the loss of coolant accidents (LOCA), total loss of feedwater (TLOFW), station blackout (SBO), and steam generator tube rupture (SGTR) using a standard  probabilistic neural network (PNN)  and Particle Swarm Optimization Based Probabilistic Neural Network (PSO PNN). The algorithm has been implemented in MATLAB.  The experiment results showed that supervised PNN PSO could classify severe accident of nuclear power plant better than the standar PNN.