Nuclear Energy for New Europe 2009

(This is a HTML version of the abstract for information only. It can differ from the original, submitted by the author(s). Special characters, formulas and figures are not properly reproduced here. Please, contact the author(s) or refer to the printed Book of Abstracts for the correct version.)

No: 904

Conference: Nuclear Energy for New Europe 2009

Title: Use of Heuristics for optimizing filling up of Canisters with Spent Fuel

Theme: Radioactive Waste Management

Author(s): Gašper Žerovnik, Matjaž Ravnik, Luka Snoj, Marjan Kromar, Irena Mele

Contact : Gašper Žerovnik

E-mail: gasper.zerovnik@ijs.si

Address: Institut "Jožef Stefan"
1001 Ljubljana

Country: Slovenia

One of the possible policies for spent nuclear fuel (SNF) management is direct disposal. Currently, in Slovenia this option is seriously considered for pressurized water reactor (PWR) NPP Krško decommissioning and waste disposal program. In cooperation with the Agency for Radwaste Management (ARAO), we considered Swedish concept of deep repository in hard rock. In the reference repository design for PWR SNF up to 4 spent fuel assemblies (SFA) are encapsulated in copper canisters which are then emplaced in vertical boreholes deep underground. According to the study of heat propagation in and around the repository [1], the heat output of each canister should not exceed the maximum allowed limit which depends on the specific thermal conditions of each site and characteristics of canister itself. For the purpose of this study a fixed maximum allowed decay heat per canister was adopted. Even with this simplification, the combinatorial problem of emplacing more than 1500 SFA in groups of 4 SFA is still hard and there is no hope of finding an optimal solution. We developed an efficient heuristic approach [2] to optimize the solution. The applied heuristics are mostly based on deterministic methods, later we included some use of random numbers. We analyzed the efficiency of different heuristics on certain datasets. Also, some insight on operation of the algorithms has been provided.
We demonstrated the use of our heuristic approach to optimize the filling of SFA in copper canisters for final deep SNF repository, according to maximal allowed thermal power per canister, and limit of 4 spent fuel assemblies per canister. Within the framework of the study [3], we calculated the minimum required number of canisters as a function of time after the NPP shutdown, and determined the optimal deposition time for final disposal for NPP Krško SNF. Also, uncertainty analysis has been performed for this case.
The idea behind the optimization method is universal and thus can be implemented for any SNF, canister, and repository design, or even other applications outside nuclear engineering. The problem is open and subject to further improvement.

[1] L. Ageskog, and P. Jansson, Heat propagation in and around the deep repository, TR-99-02, VBB Anläggning AB, Swedish Nuclear Fuel and Waste Management Co, January 1999

No:	904
Conference:	Nuclear Energy for New Europe 2009
Title:	Use of Heuristics for optimizing filling up of Canisters with Spent Fuel
Theme:	Radioactive Waste Management
Author(s):	Gašper Žerovnik, Matjaž Ravnik, Luka Snoj, Marjan Kromar, Irena Mele
Contact :	Gašper Žerovnik
E-mail:	gasper.zerovnik@ijs.si
Address:	Institut "Jožef Stefan" 1001 Ljubljana
Country:	Slovenia

One of the possible policies for spent nuclear fuel (SNF) management is direct disposal. Currently, in Slovenia this option is seriously considered for pressurized water reactor (PWR) NPP Krško decommissioning and waste disposal program. In cooperation with the Agency for Radwaste Management (ARAO), we considered Swedish concept of deep repository in hard rock. In the reference repository design for PWR SNF up to 4 spent fuel assemblies (SFA) are encapsulated in copper canisters which are then emplaced in vertical boreholes deep underground. According to the study of heat propagation in and around the repository [1], the heat output of each canister should not exceed the maximum allowed limit which depends on the specific thermal conditions of each site and characteristics of canister itself. For the purpose of this study a fixed maximum allowed decay heat per canister was adopted. Even with this simplification, the combinatorial problem of emplacing more than 1500 SFA in groups of 4 SFA is still hard and there is no hope of finding an optimal solution. We developed an efficient heuristic approach [2] to optimize the solution. The applied heuristics are mostly based on deterministic methods, later we included some use of random numbers. We analyzed the efficiency of different heuristics on certain datasets. Also, some insight on operation of the algorithms has been provided. We demonstrated the use of our heuristic approach to optimize the filling of SFA in copper canisters for final deep SNF repository, according to maximal allowed thermal power per canister, and limit of 4 spent fuel assemblies per canister. Within the framework of the study [3], we calculated the minimum required number of canisters as a function of time after the NPP shutdown, and determined the optimal deposition time for final disposal for NPP Krško SNF. Also, uncertainty analysis has been performed for this case. The idea behind the optimization method is universal and thus can be implemented for any SNF, canister, and repository design, or even other applications outside nuclear engineering. The problem is open and subject to further improvement. [1] L. Ageskog, and P. Jansson, Heat propagation in and around the deep repository, TR-99-02, VBB Anläggning AB, Swedish Nuclear Fuel and Waste Management Co, January 1999