| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 785657 | International Journal of Pressure Vessels and Piping | 2007 | 9 Pages |
Abstract
A general theory for the construction of best-fit correlation equations for multi-dimensioned sets of numerical data is presented. This new theory is based on the mathematics of n-dimensional surfaces and goodness-of-fit statistics. It is shown that orthogonal best-fit analytical trend lines for each of the independent parameters of the data can be used to construct an overall best-fit correlation equation that satisfies both physical boundary conditions and best-of-fit statistical measurements. Application of the theory is illustrated by fitting a three-parameter set of numerical finite-element maximum-stress data obtained earlier by Dr. Moffat for pressure vessel nozzles and/or piping system branch connections.
Keywords
Related Topics
Physical Sciences and Engineering
Engineering
Mechanical Engineering
Authors
S.E. Moore, D.G. Moffat,