Juan-Luis François

BWR fuel reloads design using a Tabu search technique

Abstract We have developed a system to design optimized boiling water reactor fuel reloads. This system is based on the Tabu Search technique along with the heuristic rules of Control Cell Core and Low Leakage. These heuristic rules are a common practice in fuel management to maximize fuel assembly utilization and minimize core vessel damage, respectively. The system uses the 3-D simulator code CM-PRESTO and it has as objective function to maximize the cycle length while satisfying the operational thermal limits and cold shutdown constraints. In the system tabu search ideas such as random dynamic tabu tenure, and frequency-based memory are used. To test this system an optimized boiling water reactor cycle was designed and compared against an actual operating cycle. Numerical experiments show an improved energy cycle compared with the loading patterns generated by engineer expertise and genetic algorithms.

A practical optimization procedure for radial BWR fuel lattice design using tabu search with a multiobjective function

Abstract An optimization procedure based on the tabu search (TS) method was developed for the design of radial enrichment and gadolinia distributions for boiling water reactor (BWR) fuel lattices. The procedure was coded in a computing system in which the optimization code uses the tabu search method to select potential solutions and the HELIOS code to evaluate them. The goal of the procedure is to search for an optimal fuel utilization, looking for a lattice with minimum average enrichment, with minimum deviation of reactivity targets and with a local power peaking factor (PPF) lower than a limit value. Time-dependent-depletion (TDD) effects were considered in the optimization process. The additive utility function method was used to convert the multiobjective optimization problem into a single objective problem. A strategy to reduce the computing time employed by the optimization was developed and is explained in this paper. An example is presented for a 10×10 fuel lattice with 10 different fuel compositions. The main contribution of this study is the development of a practical TDD optimization procedure for BWR fuel lattice design, using TS with a multiobjective function, and a strategy to economize computing time.

AXIAL: a system for boiling water reactor fuel assembly axial optimization using genetic algorithms

Abstract A system named AXIAL is developed based on the genetic algorithms (GA) optimization method, using the 3D steady state simulator code Core-Master-PRESTO (CM-PRESTO) to evaluate the objective function. The feasibility of this methodology is investigated for a typical boiling water reactor (BWR) fuel assembly (FA). The axial location of different fuel compositions is found in order to minimize the FA mean enrichment needed to obtain the cycle length under the safety constraints. Thermal limits are evaluated at the end of cycle using the Haling calculation; the hot excess reactivity and the shutdown margin at the beginning of cycle are also evaluated. The implemented objective function is very flexible and complete, incorporating all the thermal and reactivity limits imposed during fuel design analysis; furthermore, additional constraints can be easily introduced in order to obtain an improved solution. The results show a small improvement in the FA average enrichment obtained with the system related to the reference case that has been studied. The results show that the system converge to an optimal solution, it is observed that the mean fuel enrichment decreases while all the constraints are satisfied. A comparison was also performed using one-point and two-points crossover operator and the results of a sensitivity study for different mutation percentage are also showed.

Boiling Water Reactor Fuel Assembly Axial Design Optimization Using Tabu Search

Abstract In this paper the implementation of the tabu search (TS) optimization method to a boiling water reactor’s (BWR’s) fuel assembly (FA) axial design is described. The objective of this implementation is to test the TS method for the search of optimal FA axial designs. This implementation has been linked to the reactor core simulator CM-PRESTO in order to evaluate each design proposed in a reactor cycle operation. The evaluation of the proposed fuel designs takes into account the most important safety limits included in a BWR in-core analysis based on the Haling principle. Results obtained show that TS is a promising method for solving the axial design problem. However, it merits further study in order to find better adaptation of the TS method for the specific problem.

Validation of a methodology for fuel management analysis of Laguna Verde nuclear power plant

Abstract This paper shows the validation of the fuel management methodology based on the state of the art lattice physics code HELIOS and the CM- PRESTO code, for the fuel management analysis of the Laguna Verde nuclear power plant (LVNPP). The validation of these codes is performed with data from the first five operating cycles of LVNPP Unit 1. HELIOS calculations were performed for three different BWR standard type assemblies and compared with Monte Carlo, RECORD and fuel vendors’ results. The CM- PRESTO results are compared against plant information, such as K-effective at hot and cold conditions, thermal limits calculated by the process computer and instruments readings from the traveling in-core probes (TIP) and the local power range monitor (LPRM). Results show an improvement compared with those obtained with the previous methodology based on RECORD / PRESTO -B.

Development of an automated system for fuel reload patterns design

In this paper an automated system to generate fuel reload patterns for a boiling water reactor (BWR), based on heuristic search methods using engineers expertise is presented. The main components of the system are the knowledge base, the inference engine, the 3D BWR simulator PRESTO-B and the user interface. The knowledge base includes a generation knowledge base and a modification knowledge base, which are concerned with the way human experts generate reload patterns. The system has been developed and applied to the Laguna Verde Nuclear Power Plant, achieving similar patterns to those used in the operation. No optimization algorithm has been incorporated in this system, therefore the generated reload patterns are the best estimate according the knowledge and experience of the nuclear engineers. Future works are being developed in this area using evolutionary optimization techniques as a complement of this system.

Development of a scatter search optimization algorithm for boiling water reactor fuel lattice design

Abstract The development of a basic scatter search (SS) algorithm for the optimization of radial enrichment and gadolinia distributions for boiling water reactor (BWR) fuel lattices is presented in this paper. Scatter search is considered an evolutionary algorithm that constructs solutions by combining others. The goal of this methodology is to enable the implementation of solution procedures that can derive new solutions from combined elements. The main mechanism for combining solutions is such that a new solution is created from the strategic combination of other solutions to explore the solutions’ space. Thus, an algorithm based on SS to design a 10 × 10 fuel pin array with two water zones and diagonal symmetry was developed. The lattice performance is evaluated using a global objective function, in which the multiobjective optimization problem is converted into a single-objective problem using weighting factors to attach decision-maker preferences to each objective. The objective function is evaluated using values obtained from the HELIOS code. The results show that the main design variables (average lattice enrichment and power peaking factor) are improved, related to the reference lattice, while the reactivity requirement is satisfied. Results also demonstrate that the SS method is an efficient optimization algorithm when it is applied to the BWR design and optimization problem. Its main features are based on the use of heuristic rules since the beginning of the process, which allows directing the optimization process to the solution, and the use of the diversity mechanism in the combination operator, which allows covering the search space in an efficient way.

Study on the temperature distributions in fuel assemblies of lead-cooled fast reactors

The aim of this paper is to make a comparative study of two concepts of Lead-Cooled Fast Reactor (LFR) fuel assemblies, from a point of view of the thermofluids performance. The sub-channel analysis approach was applied to determine the temperature distribution in the fuel, in the cladding and in the lead-coolant. The mathematical model is fully transient and takes into account the heat transfer in an annular fuel pellet design. The thermofluid is modelled with a mass, energy and momentum balance with thermal expansion effects. The neutronic processes are modelled with point kinetic equations for power generation with feedback fuel temperature and expansion effects. The numerical experiments consider steady-state and transient behaviours. The numerical comparison shows that a hexagonal assembly is an option to compact the size of the LFR core design. This option leads to higher temperature in the fuel and the cladding than in the case of a rectangular assembly design. Results show the LFR with square array is more sensitive to power changes than the hexagonal array at the same nominal power and with the same transient conditions.

An alternative algorithm for the linearization process of transmutation and decay networks

Abstract To modeling the changes occurring in the nuclear reactor’s fuel composition, it is necessary to solve a coupled system of first order differential equations, known as the Bateman equations. Nowadays, there are two main methods to accomplish this task: the linear chain method and the matrix exponential method. The general procedure for the linear chain method consists in breaking a transmutation network into independent depletion chains (also known as “linear chains”) and then solving each one analytically. The common way to build these linear chains is using a Depth-First-Search (DFS) algorithm, which consists in finding every possible path in a network, tracking the decay and transmutation reactions for a set of isotopes until one stable appears or there is no more information to continue. At this point, the algorithm moves backwards searching a branch or an untraveled path, and then the procedure is repeated. In the present work, an alternative new algorithm for building linear chains is developed, which uses a special notation and reduces the problem of finding paths to the problem of ordering a sequence of characters. Unlike the DFS, the algorithm developed has not a backward routine, but it has a “fill” procedure instead. The last property decreases the computation time spent in build linear chains and is useful with cyclic chains. We carry out a comparative analysis including computational schemes based on the running time of the algorithms, versus the length of the linear chains built. We considered two kinds of networks: (1) where the initial element is a heavy isotope that undergoes fission reaction and (2) where the first element is a fission product. In all the practical scenarios the proposed algorithm is faster than the DFS’s, nevertheless when the values of the chain’s length are large enough, the running times converge, being necessary to use a more complex and advanced sorting method.

Study of the Tritium Breeding Blanket in an Hybrid Fusion–Fission Transmutation System

The reactor studied in this work is the hybrid fusion–fission transmutation system (FFTS), which is a fusion–fission hybrid reactor with a central compact fusion neutron source (CFNS). It is based on the Tokamak concept, and it is surrounded by a zone made of transuranic elements obtained from reprocessing and recycling of spent fuel of light water reactors. High-energy neutrons, of fourteen MeV, are generated in the CFNS; they are produced by the deuterium-tritium reaction. In this study, the MCNPX Monte Carlo code was used to build up a model of the FFTS for studying the tritium breeding capability of the system. Tritium is produced from neutron capture in lithium, which is located in blankets specifically designed for this purpose. The tritium breeding ratio (TBR) is defined as the average number of tritium atoms bred per tritium atom burnt in the deuterium-tritium reaction. We must have TBR>1, for a self-sustained fusion economy. In the first step of this work, the location of the lithium blankets was defined. Afterwards, different blanket materials were tested: natural lithium, enriched lithium in 6Li, different lithium alloys with neutron multipliers like lead and beryllium [Li4SiO4, LiTiO3, FLiNaBe, FLiBe, Pb-15.8Li (Li-6 at 90 %)]. Finally, a study was carried out to determine the relationship between the width of the blanket and the tritium breeding. Concerning the blanket locations, we defined four: one in the central column of the FFTS, one in the upper and one in the bottom part of the fusion region of the system, and the last one in the external part of the fission region. This means that the first three blankets use high-energy neutrons from the deuterium-tritium reaction, and the fourth blanket uses neutron leaking from the fission reactions. The principal results show that the best option is the blanket with Pb-15.8Li with lithium enriched at 90 % in lithium-6 with TBR = 1.09. It was found that the blanket at the external part of the fission region has the higher tritium breeding capability. Regarding the blanket width, it was observed that most of the tritium breeding is carried out in the first 5 cm of the blanket, and beyond this width breeding is minimal; therefore, for the blanket it is more important to have a high view factor to neutrons (i.e., a big surface exposed to neutrons) than a deep region. Finally, it is important to mention that the FFTS was critical during 1000 days that were simulated with MCNPX.