A method of predicting optimum exciter placement for normal mode testing us
ing an a priori mathematical dynamic model of the test structure is present
ed and compared to other approaches. A genetic algorithm is used to perform
the search for an optimum where the number of possible combinations is too
large for an exhaustive search. The proposed methodology is also shown to
provide an estimate of the effective number of modes in a given frequency r
ange. The technique is demonstrated using a mathematical model of a benchma
rk rectangular perspex plate and the results validated experimentally. The
technique is also applied to a model of a representative aircraft structure
. Finally, an extension of this methodology, suitable for mid-test exciter
optimization, is described.