http://localhost:8081/jspui/handle/123456789/121 2026-10-19T11:33:40Z http://localhost:8081/jspui/handle/123456789/21568 Title: VIBRATION ANALYSIS OF KIRCHOFF’S PLATE USING ISOGEOMETRIC ANALYSIS Authors: Abhi Abstract: Vibration analysis of a structure is very important. It can be under the influence of no external force which is known as free vibration, or it can be under the influence of constant force acting on the body which is called as forced vibration. Vibration analysis can be done by using the methods of finite element analysis. Iso-geometric analysis (IGA) is an advanced computing technique in the field of numerical methods. It reduces the computational time as lesser number of elements are required and the analysis is done by using such functions which represent the exact geometry. NURBS functions are used as compared to Lagrangian functions used in FEM. There are many theories for analysis of plates. Kirchhoff’s plate theory is a classical theory used in the analysis of thin plates. The thin plate has many applications in the field of acoustic instruments, electronics industry, aircraft wings. The modal analysis and the consequent mode shape study of the thin plates is very important for their use. Iso-geometric analysis method has been found to be quite effective for the vibration analysis and thus we apply the same to the vibration analysis of thin plate. In order to check the accuracy of IGA results, vibration problems of fixed bar and fixed beam have been studied using IGA. Free vibration analysis of Kirchhoff plate of uniform thickness is done and the IGA model is validated with the exact solution. Based on this model, vibration study of thin isotropic plate with varying thickness is done taking into account four types of variation in profile. This study of varying thickness profiles is the new work that has been done in this study. Linear profile as well as parabolic profiles have been taken both in the decreasing and the increasing form. The variation in the dimensionless frequency number is noted for all the cases in the simply supported case as well as in the fully clamped state. The results obtained are in excellent agreement with the analytical solution and are more accurate than the results obtained by finite element method (FEM). It is found that the plate with variable thickness is stiffer at the edges which are clamped, which in turn increases the natural frequency but it keeps on decreasing as we move along the direction of decreasing thickness. 2023-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/21567 Title: PROGNOSTIC ANALYSIS OF ROLLING BEARINGS USING ONE DIMENSIONAL CONVOLUTIONAL NEURAL NETWORK Authors: Bhukya, Haritha Abstract: The accurate prediction of the remaining useful life (RUL) of bearings plays a vital role in ensuring the reliability and efficiency of industrial machinery. The main aim of this study is prognostic analysis of rolling bearing for RUL prediction. In this study, signal processing methods are used to denoise the signal and features are extracted using both time-domain and frequency-domain analysis. Prognostic sensitive features are selected using feature ranking metrics namely, monotonicity and trendability. The respective contribution of feature ranking metric is taken into account using equal weighting factor. Further, feature fusion is carried out using PCA (principal component Analysis) technique. The reduced dimensional features are then utilized to obtain unidirectional health indicator (HI). The bearing lifetime response using HI is then modeled using Exponential degradation model which is then utilized to make new predictions. The exponential degradation model suffers with two major limitations namely, subjective selection of the first fault predicting time (FFPT) and the reduced accuracy due to random errors in the stochastic process. To overcome these issues, the Wasserstein distance (WD) method is implemented to compute FFPT. The WD method measures the difference between the probability distributions of normal condition features and faulty condition features thereby enabling accurate estimation of FFPT for accurate RUL estimations. Subsequently, One-dimensional convolutional neural network (1D-CNN) is implemented for online condition monitoring of bearing data. The 1D-CNN is employed to extract informative features from the sequential sensor data collected from the bearings. By employing multiple convolutional layers with varying filter sizes, the 1D-CNN effectively captures spatial dependencies within the sensor readings, enabling it to learn complex patterns related to bearing degradation. Besides this, the long short-term memory (LSTM) network is utilized to model the long term temporal dependencies for sequential time-series for the extracted features. The LSTM network leverages its memory cells and gating mechanisms to retain relevant information and capture the sequential dependencies, further enhancing the accuracy of the RUL predictions. To optimize the performance of the LSTM model, Bayesian optimization is implemented to optimally select the both model and training parameters for better accuracy of RUL predictions. Furthermore, the performance of the deep learning models is quantified using performance assessment metrics such as root mean square error (RMSE), and mean absolute error (MAE). In addition to this, extensive comparison study of the different RUL prediction models is carried out to demonstrate pattern learning ability of various machine lending models. 2023-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/21566 Title: HYDRAULIC NOISE PREDICTION AND CONTROL METHODS Authors: Kumar, Medikonda Syam Abstract: Noise, Vibration and Harshness have always been a primary focus of research area in a customer centric product design. As these are the adverse effects of a working machine which mainly causes various inadmissible effects such as deterioration of working lifetime of the product, creating uncomfortable work environment or even worse might create prolonged health hazards. Fluid Borne noise (FBN), even though insignificant in terms of its magnitude, is believed to be a source of vibration and noise occurrence in various hydraulic machines and hydraulic systems. Recent trend is that hydraulics are used in agro-machinery due their high power density and better controllability. One of the disadvantages of such systems is believed to be excessive noise generated during operations at certain conditions. This work is intended to identify or predict such conditions. Since first layer of transmission of such noise is the fluid layer, focus is on FBN, pertaining to certain physical and thermal properties of the system considered. Chapter 2 gives an overview of noise and FBN position in noise significance and mathematical definitions considered in order to model the hydraulic systems in acoustic domain. Such definitions mainly include the wave propagation properties inside the passive hydraulic elements. Chapter 3 consists of the hybrid approach using computational fluid dynamics (CFD) techniques with Turbulent (k-omega) Shear Stress Transport model and computational acoustics (CA) with Broadband Noise Sources model in ANSYS Fluent (2021 R1) commercial software Chapter 4 consists of mathematical formulations to model the passive elements of the circuit such as rigid pipe, hose, accumulator, etc. in acoustic model Chapter 5 provides with results of numerical simulations of centrifugal pumps with contours of pressure, velocity and acoustic power level of the surfaces and zones of pump along with plots of efficiencies and heads varying with blade number, blade outlet angles and blade width. Also, the variation of pressure ripples and flow ripples are plotted against harmonics of the pump. The objective being to identify the best blade numbers and blade designs that give the best hydraulic efficiency for the same flow conditions. 2023-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/21565 Title: Isogeometric Analysis on Bi-Variate Bézier Triangulation Authors: Kyaw, Phyo Hein Abstract: The main objectives of this thesis are twofold: firstly, to develop an automated process for surface parameterization using a set of bounding surfaces, and secondly, to enable the calculation of individual meshes within the domain. The proposed methodology ensures the preservation of precise geometric details while achieving an optimal design. By employing a triangular discretization technique, even intricate geometric domains can be accurately modeled. To create the mesh, an automatic mesh generator is employed, utilizing a Quadtree construction approach. 2023-06-01T00:00:00Z