Application of diagnostic techniques in improving vibration and acoustic prediction

Numerical techniques in vibration and noise control have traditionally been implemented in a prediction-only role, primarily using the finite and boundary element methods. However, the key to this approach is to accurately represent not only the physical system but also the excitation (structural and airborne), which is challenging in most cases. Recent work has shown that diagnostic and holography techniques can be used when limited information is available regarding the system. For example, contribution analysis can be performed to assess which parts of the system are the potentially dominant noise sources. Additionally, the inverse boundary element method can be used to accurately predict both structural and aeroacoustic sources. For structural problems where excitations are not known, transfer path analysis can be performed to complement a forward finite element analysis. Several examples using a diesel engine, an engine cover, and a water-cooled chiller are presented.