Pratik uygulamalar için operasyonel iletim yolu analizi'nin değerlendirilmesi

Although the design process and manufacturing technology of complex mechanical systems, e.g. automotive and large home appliances, have been improved, the need for vibration and noise reduction has become even more important. In this sense, Transfer Path Analysis (TPA) – also defined as Source Path Contribution (SPC) Analysis – can be used for the identification of significant vibration source(s) and related vibro-acoustic transmission path(s) in a system to solve the Noise-Vibration-Harshness (NVH) problems and to improve the vibro-acoustic design. Conventional TPA methods are based on overall signal synthesizing, where a contribution is computed from a multiplication of source strength and the Transfer Function (TF) from the source to the receiver. As these methods need identification of TFs between input and output variables as a separate process, they are more time consuming relative to the alternative methods. Furthermore, as the source strengths, also stated as operational input loads, cannot be determined directly but require the use of one of the indirect methods (e.g. Matrix Inversion Method (MIM), Dynamic Stiffness Method (DSM), etc.), conventional TPA methods results in additional uncertainties and less reliable results. Therefore, an alternative approach to classical or traditional TPA methods has been proposed in the mid 2000's. This new method, named as Operational Transfer Path Analysis (OTPA), is based upon signal decomposition technique. Within this approach, the TFs are directly estimated directly from measured operational data without the need for performing separate task for this purpose. Besides, the transmissibility function (the ratio of two responses) is generally used for the operational TFs instead of force to response TFs. In summary, OTPA is more practical than conventional TPA methods because of requiring operational measurements only and reducing total measurement time and effort. The main purposes of this thesis are (i) to implement the OTPA procedure using the MATLAB code, (ii) to verify implemented OTPA procedure using a simple flat plate test case and (iii) to apply the OTPA to a washing machine. To achieve the objectives, a literature survey including the classical TPA and OTPA methods is performed first and presented in the first chapter. Following this, in the second chapter, the theoretical background of these two methods is explained in detail. To quantify various source(s) and/or transfer path(s) contributions, the background theory of classical TPA consists of two main stages: identification of Frequency Response Functions (FRFs) and determination of operational loads. Determination of operational loads is examined under two subsections as indirect methods for structural and acoustical operational loads, respectively. For the OTPA method, the theory is firstly introduced as a Least-Square Algorithm (LSA) problem for the estimation of TFs (or FRFs). Then, the enhanced version of the OTPA theory is presented, which makes use of Singular Value Decomposition (SVD) and Principal Component Analysis (PCA) techniques to minimize the measurement noise influences and to avoid numerical problems during the TF estimation. After introducing the literature survey and theory in detail, the OTPA method is employed for different case studies in this thesis. In chapter 3, this method is numerically and experimentally performed on a simple flat plate to carry out the basic steps of the OTPA method by comparing the measured quantities with the computed counterparts. Dynamic analyses are carried out on the structure to get necessary data for comparison purposes. In addition, the subjects of linear averaging and phase-assigned spectrums are described and used in the content of OTPA application on a simple plate. Finally, the OTPA algorithm using LSA and SVD techniques is implemented into a MATLAB code to process the measurement results to estimate the TFs and compute overall responses. In chapter 4, the OTPA method is applied to a front-loading drum-type washing machine to determine and rank the critical vibro-acoustic transmission paths. The source-transfer paths-receiver model, sensor positions and operational conditions for the analysis are explained in detail. FRF measurements are made and associated modal analysis is carried out on the mechanical system for the identification of appropriate response location during the OPTA analysis. Then, considering two different experimental cases for two different operational measurement sets, the application of OPTA to a washing machine is demonstrated and the results of the contribution analyses are discussed. In the last chapter, the general conclusions are briefly summarized and some recommendations for future studies are presented.