Identification of mid-frequency loads utilizing response of FE subsystem based on hybrid FE-SEA theory

“Mid-frequency” loads are frequently encountered for complex built-up structures in various engineering fields such as aerospace and vehicle and so on. Accurate identification on “mid-frequency” loads is undoubtedly crucial for ensuring structural reliability and operational performance in these fields. However, few studies have been reported on the identification of “mid-frequency” loads so far. To address this issue, we derived the relationship between the FE subsystem node responses and the SEA subsystem energy responses in the framework of hybrid FE–SEA theory. Based on this relationship, the SEA subsystem energy responses can be obtained from the FE subsystem responses and subsequently used for “mid-frequency” load identification. This method enables flexible placement of sensors in the FE subsystem and significantly reduces the number of sensors required to acquire the SEA subsystem energy responses, which constitutes a key contribution of this work and greatly alleviates the difficulty of obtaining the SEA subsystem energy response in practical applications. The simulated and experimental studies on “mid-frequency” load identification of a complex built-up structure are conducted. The results show that the proposed method can identify more accurately the “mid-frequency” loads than the previous “low-frequency” and “high-frequency” load identification methods. Besides, the locations of selected FE subsystem nodes have little influence on ultimate “mid-frequency” load identification results and their number is much smaller than those used to directly measure the energy responses of SEA subsystems. These merits make the present approach practical in “mid-frequency” load identification of complex built-up structure in operation, where the locations and number of sensors used to accomplish load identification task are usually greatly restricted. Moreover, an experimental study of “mid-frequency” load identification in the case of multi-source excitation has been carried out and the results demonstrate that the presented method is effective in identifying mid-frequency multi-source loads.