Search NASA⌕ Search

Engineering topics

Hines, Wes

Publications and source records attributed to Hines, Wes.

Stochastic gradient descent for optimization for nuclear systems

The use of gradient descent methods for optimizing k-eigenvalue nuclear systems has been shown to be useful in the past, but the use of k-eigenvalue gradients have proved computationally challenging due to their stochastic nature. ADAM is a gradient descent method that accounts for gradients with a stochastic nature. This analysis uses challenge problems constructed to verify if ADAM is a suitable tool to optimize k-eigenvalue nuclear systems. ADAM is able to successfully optimize nuclear systems using the gradients of k-eigenvalue problems despite their stochastic nature and uncertainty. Furthermore, it is clearly demonstrated that low-compute time, high-variance estimates of the gradient lead to better performance in the optimization challenge problems tested here.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Artificial Intelligence Design of Nuclear Systems

Advanced manufacturing applications in the nuclear industry have the potential to revolutionize nuclear design. While current engineering design objectives may be consistent with objectives from the past, such as those intended to minimize the fuel cost or to limit the maximum fuel temperature for a given power level, the possible geometries which can be considered as candidate designs have changed. Optimizations applied to current designs in nuclear engineering are limited by traditional geometric shapes that are easy to manufacture with traditional processes: slabs as fuel plates, cylinders as fuel rods, spheres as fuel pebbles, axis-aligned coolant channels, and more. Advanced manufacturing allows for an approach to the nuclear design problem with arbitrary geometries that can enable previously unrealized optimizations in performance and cost. This work explores the application of arbitrary geometry to the nuclear engineering design process.

42 ENGINEERING↗

Identifying Suitable Degradation Parameters for Individual-Based Prognostics

The ultimate goal of most prognostic systems is accurate prediction of the remaining useful life of individual systems or components based on their use and performance. Traditionally, individual-based prognostic methods use a measure of degradation to make lifetime estimates. Degradation measures may include sensed measurements, such as temperature or vibration level, or inferred measurements, such as model residuals or physics-based model predictions. Often, it is beneficial to combine several measures of degradation into a single parameter. Parameter features such as trendability, monotonicity, and prognosability can be used to compare candidate prognostic parameters to determine which is most useful for individual-based prognosis. By quantifying these features for a given parameter, the metrics can be used with any traditional optimization technique to identify an appropriate parameter. This parameter may be used with a parametric extrapolation model to make prognostic estimates for an individual unit. The proposed methods are illustrated with an application to simulated turbofan engine data.

Diagnostics↗

Application of Monitoring and Prognostics to Small Modular Reactors

Prognostics is one component of a full health monitoring system, which generally includes plant condition monitoring, fault detection, fault diagnostics, and estimation of remaining useful life. Empirical methods for prognostics have been widely studied, and the efficacy of these models is well accepted. However, these methods typically require large amounts of run-to-failure data for accurate model development. This has been one of the main roadblocks for developing prognostic models for high reliability or safety critical systems: equipment rarely malfunctions; when it does, it is often times repaired before failure occurs. This paper presents the work completed in developing and validating an accurate prognostic system for several potential IRIS system fault modes using high-fidelity simulated data before any plant operation takes place. Fault modes under investigation include heat exchanger fouling, sensor calibration drifts, and pump degradation. Prognostic models are developed which can be applied to the IRIS plant from day one of operation to facilitate the final step in a full health monitoring system. These estimates of remaining useful life can be used to inform both operations and maintenance planning to reduce plant downtime and provide the highest level of power production from the beginning of operation.

On-Line Monitoring↗

FY-2010 Process Monitoring Technology Final Report

During FY 2010, work under the Spectroscopy-Based Process Monitoring task included ordering and receiving four fluid flow meters and four flow visible-near infrared spectrometer cells to be instrumented within the centrifugal contactor system at Pacific Northwest National Laboratory (PNNL). Initial demonstrations of real-time spectroscopic measurements on cold-stream simulants were conducted using plutonium (Pu)/uranium (U) (PUREX) solvent extraction process conditions. The specific test case examined the extraction of neodymium nitrate (Nd(NO3)3) from an aqueous nitric acid (HNO3) feed into a tri-n-butyl phosphate (TBP)/ n-dodecane solvent. Demonstration testing of this system included diverting a sample from the aqueous feed meanwhile monitoring the process in every phase using the on-line spectroscopic process monitoring system. The purpose of this demonstration was to test whether spectroscopic monitoring is capable of determining the mass balance of metal nitrate species involved in a cross-current solvent extraction scheme while also diverting a sample from the system. The diversion scenario involved diverting a portion of the feed from a counter-current extraction system while a continuous extraction experiment was underway. A successful test would demonstrate the ability of the process monitoring system to detect and quantify the diversion of material from the system during a real-time continuous solvent extraction experiment. The system was designed to mimic a PUREX-type extraction process with a bank of four centrifugal contactors. The aqueous feed contained Nd(NO3)3 in HNO3, and the organic phase was composed of TBP/n-dodecane. The amount of sample observed to be diverted by on-line spectroscopic process monitoring was measured to be 3 mmol (3 x 10-3 mol) Nd3+. This value was in excellent agreement with the 2.9 mmol Nd3+ value based on the known mass of sample taken (i.e., diverted) directly from the system feed solution.

process monitoring, spectroscopy-based monitoring,↗