Multiobjective Simulated Annealing Method Implementation For Pwr Fuel Loading Pattern Optimization Using Corebn Code

MULTIOBJECTIVE SIMULATED ANNEALING METHOD IMPLEMENTATION FOR PWR FUEL LOADING PATTERN OPTIMIZATION USING COREBN CODE

Christina Novila Soewono, Alexander Agung, Sihana
Jurusan Teknik Fisika Fakultas Teknik - Universitas Gadjah Mada
ABSTRACT
MULTIOBJECTIVE SIMULATED ANNEALING METHOD IMPLEMENTATION FOR PWR FUEL LOADING PATTERN OPTIMIZATION USING COREBN CODE. Optimizing loading/reloading pattern design is one of nuclear fuel management activities in order to reduce fuel cycle costs while satisfying safety constraints and operational targets. Multiplication factor at the end of cycle and maximum power peaking factors are the parameters to define the optimal LP design. This optimization initial fuel loading pattern study is based on multiobjective simulated annealing algorithm which is coupled to COREBN code for core burn up calculation. Optimization is implemented on ¼ core model (52 fuel assemblies) which represent the whole core. The result will then be compared to standard model in order to observe the improvement.

Keywords: optimization, loading pattern, multiplication factor at end of cycle, power peaking factors, multiobjective simulated annealing, COREBN
Proceeding Seminar Nasional Ke-15 "TEKNOLOGI DAN KESELAMATAN PLTN SERTA FASILITAS NUKLIR", Surakarta, 17 Oktober 2009

Read more...

Implementation Of Genetic Algorithm Method For PWR Fuel Loading Pattern Optimization Using COREBN Code

IMPLEMENTATION OF GENETIC ALGORITHM METHOD FOR PWR FUEL LOADING PATTERN OPTIMIZATION USING COREBN CODE

Petrus, Alexander Agung, Sihana
Jurusan Teknik Fisika, Fakultas Teknik, Universitas Gadjah Mada
ABSTRACT
IMPLEMENTATION OF GENETIC ALGORITHM METHOD FOR PWR FUEL LOADING PATTERN OPTIMIZATION USING COREBN CODE. Since the large number of possible combination for the fuel assembly loading in the core at the beginning of reactor operation, the core configuration optimized to find an optimal core configuration that will achieve maximum k_eff at end of cycle and minimum power peaking factor (PPF). This optimization has 2 Genetic Algorithm methods, the first method uses single objective and the second method uses multi objective. The optimization uses ¼ symmetry reactor core model (52 fuel assemblies position), with 3 types of fuel assemblies consists 13 assemblies of 1,5%, 15 assemblies of 2,5% and 24 assemblies of 3% U-235 enrichment without burnable poisson rod. Neutronic calculation of fuel assembly using PIJBurn code and core calculation using COREBN code. From the single objective optimization is obtained the optimum configuration with 8,9% (60 days) cycle length extension and 23,31% decrease in PPF compared to standard model. For multi objective optimization obtained a set pareto front containing 47 non-dominated solutions. By using standard deviation of the crowding distances method, a single final solutions is obtained. The solution gives 10,45% (70 days) cycle length extension and 27,7 % decrease in PPF compared to standard model. Both of optimization method success to obtain optimum solution and fulfill the safety standard.

Keywords: fuel assembly, k_eff, PPF, Genetic Algorithm, cycle length.
Proceeding Seminar Nasional Ke-15 "TEKNOLOGI DAN KESELAMATAN PLTN SERTA FASILITAS NUKLIR", Surakarta, 17 Oktober 2009

Read more...