Full-scale aircraft structural fatigue tests are extremely complex, not only in terms of the test definition and implementation, but also from the control systems viewpoint. The load spectra that these tests are subjected to are also becoming more complex as manufacturers and certifying authorities strive for more realistic loading conditions. At the same time, there is pressure to complete the test more quickly or at least in the same time as previous tests that were simpler and contained fewer cycles. This can only be achieved by optimizing the test configuration and control system parameters or in some cases implementing new control strategies. This is difficult on a full-scale test because of all the complex interactions on these multi-channel tests. Methods that may work well on smaller tests can sometimes result in poorer performance. Thus there is a need for a greater understanding of the system and some predictive capability. This can be achieved with computer modeling.
This paper suggests an implementation scheme in terms of the introduction and acceptance of the concepts of modeling to a full-scale structural test laboratory and provides examples of the use of modeling at various stages in its implementation. Examples are presented from the very basic level of education, using models to understand sources of problems on simple tests, to evaluating new control strategies and predicting the impulse response of a test system (structure/actuators/control system). Finally, the possibility of predicting system behavior with sufficient accuracy to optimize the test for given test and control system hardware is discussed.