27 Nonlinear Identification of an Aero-Engine Component Using Polynomial Nonlinear State Space Model 263 Fig. 27.2 Experimental set-up of the Engine casing suspended on a TRL-6 frame. (a) Test Set-up, (b) Instrumented section of the casing was designed to replicate the traditional horizontal configuration of such component when attached to an aircraft. A total of 32 accelerometers was employed to instrument the structure, all of which are single-axial sensors. For this investigation, only the first subassembly section where the plate is bolted to the engine casing as illustrated Fig. 27.2b was fully instrumented, a large shaker visible in Fig. 27.2a was used to apply excitation inputs to the casing in a vertical direction. The first stage of the experimental investigation involved conducting a set of traditional modal tests on the casing and plate assembly to obtain the natural frequencies and damping ratios at low amplitude of vibration. In addition, stepped-sine test at high excitation levels was carried out on the assembly to check for symptoms of nonlinearity that may be present in the instrumented section of the casing. 27.4 Linear Analysis of the Casing and Plate Assembly The first measurements obtained from the experimental test comprised of several low random data which were acquired based on broadband excitation, the choice of broadband excitation was made based on its conventional use in modal testing. Broadband excitation also provides some early information on the behaviour of the structure and experimental configuration, the low level random test was performed using the Spectral Test module in LMS Test Lab. The test structure was excited near the flange connecting the inter-case (middle case) with the first case as shown in Fig. 27.2a. The structure was excited using burst random excitation ranging between 30 and 500 Hz. The Frequency Response Function (FRFs) obtained from the test were exploited to identify the linear modal parameters of the engine casing. Figure 27.3 shows a selection of the FRFs obtained from the low-level test. Before getting an insight of the nonlinear behaviour of the casing, a linear modal analysis based on the FRF’s obtained from low-level random test was conducted using the PolyMAX method [17]. The applied random excitation has an RMS value of 45 N, Fig. 27.3 shows a selection of the FRFs obtained from the low-level test and Table 27.1 presents the corresponding natural frequencies and damping ratios of the casing with the plate bolted to the first cylinder. 27.5 Non-linear Detection Based on Distortions in Time-Series and FRF Measurements Equally, to check for symptoms of nonlinear behaviour in the casing and plate assembly, stepped sine and sine-sweep tests were conducted on the assembly. The new assembly was also excited at several excitation levels covering a frequency bandwidth of 80–90 Hz. Figure 27.4 shows a selection of the FRFs and phase results obtained from the stepped sine test. In Fig. 27.4 stepped-sine FRFs and phase are presented for plate casing assembly, the test was concentrated around the first
RkJQdWJsaXNoZXIy MTMzNzEzMQ==