On Using Quasi-Newton Algorithms of the Broyden Class for Model-to-Test Correlation

Abstract

The correlation of a model with test results is a common task in engineering. Often genetic algorithms or adaptive particle swarm algorithms are used for this task. In this paper another approach is presented using two quasi Newton algorithms of the class defined by Broyden. A study is performed with thermal space industry models showing the performance of this approach. By comparing it to the results of other studies it is shown that this approach reduces the number of iterations by several orders of magnitude. This reduces the calculation time for a typical thermal model correlation from weeks or months to hours or days.