Samuel H. Levine

AUTOLOAD, An Automatic Optimal Pressurized Water Reactor Reload Design System with an Expert Module

An automatic optimal pressurized water reactor (PWR) reload design expert system AUTOLOAD has been developed. It employs two important new techniques. The first is a new loading priority scheme that defines the optimal placement of the fuel in the core that has the maximum end-of-cycle state k[sub eff]. The second is a new power-shape-driven progressive iteration method for automatically determining the burnable poison (BP) loading in the fresh fuel assemblies. The Haling power distribution is used in converting the theoretically optimal solution into the practical design, which meets the design constraints for the given fuel assemblies. AUTOLOAD is a combination of C and FORTRAN languages. It requires only the required cycle length, the maximum peak normalized power, the BP type, the number of fresh fuel assemblies, the assembly burnup, and BP histories of the available fuel assemblies as its input. Knowledge-based modules have been built into the expert system computer code to perform all of the tasks involved in reloading a PWR. AUTOLOAD takes only [approximately] 30 CPU min on an IBM 3090 600s mainframe to accomplish a practical reload design. A maximum of 12.5% fresh fuel enrichment saving is observed compared with the core used by the utility.

A two-step method for developing a control rod program for boiling water reactors

This paper reports on a two-step method that is established for the generation of a long-term control rod program for boiling water reactors (BWRs). The new method assumes a time-variant target power distribution in core depletion. In the new method, the BWR control rod programming is divided into two steps. In step 1, a sequence of optimal, exposure-dependent Haling power distribution profiles is generated, utilizing the spectral shift concept. In step 2, a set of exposure-dependent control rod patterns is developed by using the Haling profiles generated at step 1 as a target. The new method is implemented in a computer program named OCTOPUS. The optimization procedure of OCTOPUS is based on the method of approximation programming, in which the SIMULATE-E code is used to determine the nucleonics characteristics of the reactor core state. In a test in cycle length over a time-invariant, target Haling power distribution case because of a moderate application of spectral shift. No thermal limits of the core were violated. The gain in cycle length could be increased further by broadening the extent of the spetral shift.

Application of neutron radiography for fluid flow visualization

Real-time thermal neutron radiography has been applied to the visualization of fluid flows. Since neutrons can penetrate metal casings, the technique may be useful for the visualization of fluids flowing inside metal enclosures, such as valves, engine or transmission components, etc. The technique described involves shadowgraph imaging of neutron-opaque tracer materials (either solid or fluid particles) as they convert in a stream of neutron-transport ambient fluid. Real-time motion pictures of several simple flows have been recorded, from which velocities, regions of flow separation, rate of mixing, and other information about the flow field can be obtained. The neutron radiography facility at the Penn State Breazeale Nuclear Reactor and the studies performed to determine viable liquids useful in neutron radiography applications are described. Some samples of successful flow visualizations are also presented.

Iterative solution to the optimal poison management problem in pressurized water reactors

A new method for solving the optimal poison management problem for a multiregion pressurized water reactor has been developed. The optimization objective is to maximize the end-of-cycle core excess reactivity for any given beginning-of-cycle fuel loading. The problem is treated as an optimal control problem with the region burnup and control absorber concentrations acting as the state and control variables, respectively. Constraints are placed on the power peaking, soluble boron concentration, and control absorber concentrations. The solution method consists of successive relinearizations of the system equations resulting in a sequence of nonlinear programming problems whose solutions converge to the desired optimal control solution. Application of the method to several test problems based on a simplified three-region reactor suggests a bang-bang optimal control strategy with the peak power location switching between the inner and outer regions of the core and the critical soluble boron concentration as low as possible throughout the cycle.

Heuristic optimization of pressurized water reactor fuel cycle design under general constraints

Optimization techniques in fuel management have directed modern fuel cycle designs to use low-leakage loading patterns. Future optimization calculations involving low-leakage patterns must utilize nucleonic models that are both fast operationally and rigorous. A two-dimensional two-group diffusion theory code is developed and lattice homogenization constants are generated using a modified LEOPARD code to fulfill these criteria. Based on these two codes, a heuristic optimization study is performed that considers the general constraints (e.g., spent-fuel storage limit and mechanical burnup limit) given to a utility fuel cycle designer. The optimum cycle length that minimizes the fuel cost is {approximately} 600 effective full-power days for the conditions assumed.

Development of an asymmetric multiple-position neutron source (AMPNS) method to monitor the criticality of a degraded reactor core

Abstract Tne analytical/experimental method has been developed to to monitor the subcritical reactivity and unfold the k ∞ distribution of a degraded reactor core. The method uses several fixed neutron detectors and a 252 Cfneutron source placed sequentially in multiple positions in the core. Therefore, it is called the asymmetric multiple-position neutron source (AMPNS) method. The AMPNS method employs the nucleonic codes to analyze in two dimensions the neutron multiplication of a 252 Cf neutron source. An optimization program, GPM, was utilized to unfold the k ∞ distribution of the degraded core, in which the desired performance measure minimizes the error between the calculated and the measured count rates of the degraded reactor core. The analytical/experimental approach is validated by performing experiments using the Penn. State Breazeale TRIGA reactor (PSBR). A significant result of this study has been to provide a means to plan the source and detector placements and assign core cells to the damaged TMI-2 core as well as to monitor the criticality during the recovery period.

Optimized Depletion of Lumped Burnable Poisons in Pressurized Water Reactors

Sets of basic poison depletion curves have been computed and used with a linear and a nonlinear program to match as closely as possible a required optimized fuel assembly depletion profile. The technique is applied to the fuel assembly design of the Three Mile Island-Unit One reactor wherein either solid poison or annular poison rods are used to contain burnable poisons (BPs). These BPs are either in the form of fine particles or spheroids uniformally distributed in the BP rods. The optimization technique selects the best basic depletion curves to provide an optimum fit to the designed fuel assembly depletion profile. The basic poison depletion curves are nonorthogonal and not complete. Thus, the best fit is not exact because the desired depletion profile requires sudden changes that cannot be attained by the depletion of the BPs.

Nonequilibrium bipolar charging of aerosol particles: Theory and experiment

Abstract A theoretical and experimental study of particle charge acquisition in a region containing unequal current densities of ions of opposite polarity (a nonequilibrium bipolar region) is presented. A continuum model of ion transport is combined with charging expressions to describe the nonequilibrium bipolar charging of aerosol particles. Experiments employing an 11-MCi 90Sr-90Y β source and electric and magnetic fields are performed to verify the calculational method. Trajectories of 50–100-μm-diameter glass beads falling through the experimental apparatus are recorded on film by a photographic technique utilizing an open aperture camera and stroboscopic light source. Predicted increases in the particle charging rate upon application of a magnetic field strength of approximately 1000 G are clearly demonstrated by experiments. Measured charging rates are in fair agreement with theory, especially in the presence of the magnetic field. This research provides a quantitative basis upon which particle charging and motion in bipolar ionic regions may be calculated.

Genetic algorithm application for burnable poison placement in pwrs with optimized UO2/Gd2O3 fuel pin configurations

In this paper, an efficient genetic algorithm has placed burnable poisons (BPs) into all of the fresh fuel positions in the core employing the optimized BP configurations and techniques developed in two previous papers. Of importance was the previous development of a Kinf filter, which greatly reduced the computational time. The Kinf filter eliminated many of the invalid genotypes/phenotypes before making a precise core depletion analysis. An extensive BP library was generated by the CASMO-4/TABLES-3 codes. The process was automated with a user-friendly program developed for this purpose. The BPs were vendor UO2/Gd2O3 fuel assembly designs used in a reference Three Mile Island Unit 1 core. The optimized UO2/Gd2O3 fuel pin configurations have small residual binding at end of cycle (EOC), and BP loading optimization results with 97.2 ppm soluble boron at EOC while it was 94.4 ppm with the available vendor designs. The result was that optimized UO2/Gd2O3 fuel pin configurations were developed with unique self-shielding properties and residual binding that also provided a 6.89% reduction in the total required Gd amount, providing extra savings in fuel cost.

Genetic algorithm to optimize the UO2/Gd2O3 fuel pin designs in a pressurized water reactor

An efficient and practical genetic algorithm (GA) was developed to optimize the UO2/Gd2O3 fuel pin burnable poison (BP) configurations for fresh fuel assembly (FA) designs loaded in a pressurized water reactor core. The objective of the optimization was to minimize the residual binding due to residual Gd isotopes in the fuel at the end of cycle (EOC). The GA process for creating new BP designs in a coded form called genotypes is generated randomly resulting in a large number of invalid designs. Each new BP design or genotype created by the new GA must be decoded into its corresponding phenotype so that it can be evaluated with a coupled fuel lattice and core depletion calculation. It is essential that most of the invalid designs be eliminated before performing the precise coupled fuel lattice calculation because of the long CPU time that it takes for this calculation. The elimination was accomplished in the new GA by incorporating a beginning-of-cycle (BOC) Kinf filter. The BOC Kinf filter eliminated most of the invalid new genotypes by assigning a high negative penalty to all genotypes that have a BOC Kinf greater than some limit (1.065) for the reference TMI-1 FA. This filter eliminates the need for performing coupled lattice and core depletion calculations for these genotypes. It accelerated the solution process and allowed evaluation of all new genotypes within one day. In this way, the GA minimized the residual binding using an objective function, which maximized the EOC soluble boron (SB) concentration. In essence, the EOC SB or its equivalent EOC keff was maximized.