PWR loading pattern optimization using Harmony Search algorithm

• Published in: Annals of nuclear energy Vol. 53; pp. 288 - 298
• Main Authors: Poursalehi, N., Zolfaghari, A., Minuchehr, A.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2013
• Subjects: Fitness function; Harmony Search algorithm; Loading pattern optimization; Nodal expansion method; Harmony Search algorithm; Loading pattern optimization; Nodal expansion method; Fitness function
• ISSN: 0306-4549; 1873-2100
• DOI: 10.1016/j.anucene.2012.06.037

► Numerical results reveal that the HS method is reliable. ► The great advantage of HS is significant gain in computational cost. ► On the average, the final band width of search fitness values is narrow. ► Our experiments show that the search approaches the optimal value fast. In this paper a core reloading technique using Harmony Search, HS, is presented in the context of finding an optimal configuration of fuel assemblies, FA, in pressurized water reactors. To implement and evaluate the proposed technique a Harmony Search along Nodal Expansion Code for 2-D geometry, HSNEC2D, is developed to obtain nearly optimal arrangement of fuel assemblies in PWR cores. This code consists of two sections including Harmony Search algorithm and Nodal Expansion modules using fourth degree flux expansion which solves two dimensional-multi group diffusion equations with one node per fuel assembly. Two optimization test problems are investigated to demonstrate the HS algorithm capability in converging to near optimal loading pattern in the fuel management field and other subjects. Results, convergence rate and reliability of the method are quite promising and show the HS algorithm performs very well and is comparable to other competitive algorithms such as Genetic Algorithm and Particle Swarm Intelligence. Furthermore, implementation of nodal expansion technique along HS causes considerable reduction of computational time to process and analysis optimization in the core fuel management problems.