PSI - Issue 37

Aleksey Mironov et al. / Procedia Structural Integrity 37 (2022) 241–249 Aleksey Mironov, Pavel Doronkin / Structural Integrity Procedia 00 (2019) 000 – 000

242

2

Introduction The health and usage monitoring systems are widely used for vibration monitoring of power plant units on helicopters as prescribed in the regulations UK Civil Aviation Authority, (2011). In contrast with a power plant, critical helicopter structures like rotors or tail boom are still out of permanent monitoring. Practically all non-destructive techniques (NDT), including ultrasound, eddy current, X-ray referred by Ihn et al. (2001) and Speckmann (2002) make the inspection of structural components available during the maintenance check or the overhaul. In between ground checks the aircraft structures remain unobservable for a long time, while some latent defects, especially in composite parts, may grow and increase risk of damage. So, there is the need to provide permanent monitoring of helicopter structures both on ground and in flight. There are different approaches to this task, for instance, Janeliukstis et al. (2021) considered the load and damage monitoring applying embedded sensors in composite structures. One of the promising approaches to rotating blades monitoring is the modal analysis, allowing to use the relationship between structural mechanical and modal properties. Any fault of a blade, such as ply separation of composite one, results in the redistribution of masses, stiffness or damping of the considered structure. Such redistribution modifies the modal properties of the blade. Modal analysis techniques estimate the modal parameters of structural parts and by such a way allow monitoring of their changes. For instance, Lakhdara et al. (2013) used the change of modal parameters of the composite structure for damages detection by applying the specific methods of vibration analysis. For the helicopter rotor blade Ramirez et al. (2013) proposes a network of autonomous wireless sensors that allow monitoring. Looking for a solution for SHM, the laboratory White et al. (2010) fabricated a rotor blade equipped with accelerometers and innovative optical Fiber Bragg sensors and investigated its load monitoring and damage detection capabilities. In recent years more researchers have applied Operational Modal Analysis (OMA) methods using only the output signals of vibrated structure as more appropriate for operating structures. Earlier Peeters et al. (1995) considered Stochastic Subspace Techniques and Brincker et al. (2000) discussed Modal Identification from Ambient Response using Frequency Domain Decomposition. In aviation, Abdelghani et al. (1999) considered the output-only subspace based modal analysis techniques and reported the experimental results of the Paris MS760 airplane case for the proposed monitoring algorithms. There are also trial applications of OMA for civil structures health monitoring, for instance Rainieri et al. (2007) considered the theoretical and applicative aspects of OMA. Different theoretical and experimental works, like Rizo-Patron et al. (2017) considered the estimation of modal parameters for helicopters using only ambient excitation (instead of controlled actuation). Upon expanding its application, the OMA approach becomes most promising for the monitoring of rotating blades in flight. At the same time, the successful application of OMA to aircraft structures is impeded by several obstacles. External factors such as rotational speed can affect the modal parameters of the blade more than damages. So for example, Manzato et al. (2012) discusses the experimental case of Ariane 5 launcher using Stochastic Subspace Identification algorithms considering the influence of changing mass on modal properties. Also, the measurement system needs optimal solutions to match limitations in mass, sizes and costs being applicable to rotating blades. Light and cheap deformation sensors, as Solovyev et al. (2020) and Mironov et al. (2021a) reported, may provide a more effective solution as the modal curvature may be usable also for damage detection. Resuming, we may see the OMA is the promising approach, however, there are technical, methodical and economic problems for its implementation on helicopters. 1. Issues of OMA application on a blade As the basic method of modal estimation from vibration data this research study applies Singular Value Decomposition (SVD) of power spectra matrix described by Mironov D. et al. (2018). The basic assumption of OMA application is the uniform distribution of dynamic loads in frequency range. The typical procedure of OMA application includes measurement of operating blade vibrations, pre-processing of measured data, estimation of modal parameters and its identification. To estimate modal parameters of the blade, OMA techniques must use the modal model having Degrees Of Freedom (DOF) sufficient to reflect blade’s vibration shapes. For instance, Böswald et al. (2017) from DLR presented some cases of the modal parameters study of taxiing aircraft, where the model includes hundreds of DOF related to accelerometers on the structures.