From field vibration data to laboratory simulation

The objective of a test program is to simulate a specified dynamic environment. The question is: How do I go from field-test data to a reasonable laboratory or finite-element simulation? This paper looks at a simple situation of a two degree-of-freedom vehicle that has a broadband excitation source and a two degree-of-freedom test item that can be attached to the vehicle. The paper shows that significantly different field data are obtained depending on field-test conditions, and vastly different simulation results are obtained when using these different field data as laboratory inputs. A relatively simple theoretical model is developed that explains how field data in this case can be modified in the frequency domain to obtain a more realistic laboratory test environment. This presentation shows that a general theory of vibration testing is needed, a theory that provides guidelines for this process of obtaining field data and developing realistic test specifications.