skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: ON PREDICTION AND MODEL VALIDATION

Abstract

Quantification of prediction uncertainty is an important consideration when using mathematical models of physical systems. This paper proposes a way to incorporate ''validation data'' in a methodology for quantifying uncertainty of the mathematical predictions. The report outlines a theoretical framework.

Authors:
; ;
Publication Date:
Research Org.:
Los Alamos National Lab., NM (US)
Sponsoring Org.:
US Department of Energy (US)
OSTI Identifier:
774507
Report Number(s):
LA-UR-01-759
TRN: AH200121%%82
DOE Contract Number:
W-7405-ENG-36
Resource Type:
Conference
Resource Relation:
Conference: Conference title not supplied, Conference location not supplied, Conference dates not supplied; Other Information: PBD: 1 Feb 2001
Country of Publication:
United States
Language:
English
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; FORECASTING; MATHEMATICAL MODELS; VALIDATION; PREDICTION EQUATIONS

Citation Formats

M. MCKAY, R. BECKMAN, and K. CAMPBELL. ON PREDICTION AND MODEL VALIDATION. United States: N. p., 2001. Web.
M. MCKAY, R. BECKMAN, & K. CAMPBELL. ON PREDICTION AND MODEL VALIDATION. United States.
M. MCKAY, R. BECKMAN, and K. CAMPBELL. Thu . "ON PREDICTION AND MODEL VALIDATION". United States. doi:. https://www.osti.gov/servlets/purl/774507.
@article{osti_774507,
title = {ON PREDICTION AND MODEL VALIDATION},
author = {M. MCKAY and R. BECKMAN and K. CAMPBELL},
abstractNote = {Quantification of prediction uncertainty is an important consideration when using mathematical models of physical systems. This paper proposes a way to incorporate ''validation data'' in a methodology for quantifying uncertainty of the mathematical predictions. The report outlines a theoretical framework.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Feb 01 00:00:00 EST 2001},
month = {Thu Feb 01 00:00:00 EST 2001}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Stochastic differential equations (SDE) can be used to describe the time-varying nature of wireless channels. This paper validates a long-term fading channel model for estimation and prediction from solely using measured received signal strength measurements. Such channel models can be used for optimizing wireless networks deployed for industrial automation, public access, and communication. This paper uses two different sets of received signal measurement data to estimate an predict the signal strength based on past measurements. The realworld performance of the estimation and prediction algorithm is demonstrated.
  • A pseudo-multi-zone phenomenological model has been created with the ultimate goal of supporting efforts to enable broader commercialization of low temperature combustion modes in diesel engines. The benefits of low temperature combustion are the simultaneous reduction in soot and nitric oxide emissions and increased engine efficiency if combustion is properly controlled. Determining what qualifies as low temperature combustion for any given engine can be difficult without expensive emissions analysis equipment. This determination can be made off-line using computer models or through factory calibration procedures. This process could potentially be simplified if a real-time prediction model could be implemented to runmore » for any engine platform this is the motivation for this study. The major benefit of this model is the ability for it to predict the combustion trajectory, i.e. local temperature and equivalence ratio in the burning zones. The model successfully captures all the expected trends based on the experimental data and even highlights an opportunity for simply using the average reaction temperature and equivalence ratio as an indicator of emissions levels alone - without solving formation sub-models. This general type of modeling effort is not new, but a major effort was made to minimize the calculation duration to enable implementation as an input to real-time next-cycle engine controller Instead of simply using the predicted engine out soot and NOx levels, control decisions could be made based on the trajectory. This has the potential to save large amounts of calibration time because with minor tuning (the model has only one automatically determined constant) it is hoped that the control algorithm would be generally applicable.« less
  • Unlike fresh fuel assumptions typically employed in the criticality safety analysis of spent fuel configurations, burnup credit applications rely on depletion and decay calculations to predict the isotopic composition of spent fuel. These isotopics are used in subsequent criticality calculations to assess the reduced worth of the spent fuel. To validate the codes and data used in depletion approaches, experimental measurements are compared with numerical predictions for relevant spent fuel samples. This paper describes a set of experimentally characterized pressurized-water-reactor (PWR) fuel samples and provides a comparison to results of SCALE-4 depletion calculations. An approach to determine biases and uncertaintiesmore » between calculated and measured isotopic concentrations is discussed, together with a method to statistically combine these terms to obtain a conservative estimate of spent fuel isotopic concentrations.« less