Yazar "Kati, Hilal Doganay" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Artificial neural network-based parameter identification of a beam-3D tip attachment system
    (Taylor & Francis Inc, 2025) Gokdag, Hakan; Kati, Hilal Doganay
    In this study, an artificial neural network (ANN)-based approach is proposed for estimating the tip mass attachment and structural damping parameters in a beam-3D mass system undergoing combined bending and torsional vibrations. The study begins with a detailed explanation of the calculation of the frequency response functions (FRFs) for this specific system. Subsequently, a difference vector is defined, based on the discrepancy between experimental and numerical FRF curves. This vector is dependent on ten parameters: the mass of the tip attachment, its mass moments of inertia, and the coordinates of the mass center, and the structural damping ratios of the beam. An orthogonal design method is then employed to create a design space for these parameters, and the elements of the difference vector are calculated for each point within this space. The points in the design space and the computed difference vectors are utilized as input and output data for training the ANN. The optimization process conducted with the obtained ANN model allows for realistic estimation of the tip mass parameters and damping values. The analysis reveals that as the design space widens, the parameter estimation process becomes increasingly challenging. This, in turn, necessitates a larger number of points in the design space and more neurons in the hidden layers of the trained network. In cases where the design space is small to medium in size, the parameter estimation errors are observed to be <5%. However, for wider design spaces, the estimation errors tend to increase.
  • Küçük Resim Yok
    Öğe
    Identification of the tip mass parameters in a beam-tip mass system using response surface methodology
    (Walter De Gruyter Gmbh, 2024) Gokdag, Hakan; Kati, Hilal Doganay
    In this study, a response surface based approach is introduced to determine the physical parameters of the tip mass of a beam - tip mass system, such as mass, mass moment of inertia and coordinates of the centre of gravity with respect to the beam end point. To this end, first, a difference function was formulated based on the differences between the peak frequencies and peak amplitudes of the experimental and analytical frequency response functions. Later, observation points were established in the design space using orthogonal arrays, and a response surface was developed using the difference function values at these points. Next, the tip mass parameters were determined by minimizing the response surface with genetic algorithm and particle swarm optimization as well as fmincon, a gradient-based solver of the Matlab program. For comparison purposes, those parameters were obtained by also direct minimization of the difference function with the same algorithms. It was concluded that the tip mass parameters were successfully determined within reasonable error limits by the response surface method with less computational burden. Finally, the effect of design space width on the response surface quality is demonstrated numerically.
  • Küçük Resim Yok
    Öğe
    Numerical Analysis of Crack Path Effects on the Vibration Behaviour of Aluminium Alloy Beams and Its Identification via Artificial Neural Networks
    (Mdpi, 2025) Kati, Hilal Doganay; Buhari, Jamilu; Francese, Arturo; He, Feiyang; Khan, Muhammad
    Understanding and predicting the behaviour of fatigue cracks are essential for ensuring safety, optimising maintenance strategies, and extending the lifespan of critical components in industries such as aerospace, automotive, civil engineering and energy. Traditional methods using vibration-based dynamic responses have provided effective tools for crack detection but often fail to predict crack propagation paths accurately. This study focuses on identifying crack propagation paths in an aluminium alloy 2024-T42 cantilever beam using dynamic response through numerical simulations and artificial neural networks (ANNs). A unified damping ratio of the specimens was measured using an ICP (R) accelerometer vibration sensor for the numerical simulation. Through systematic investigation of 46 crack paths of varying depths and orientations, it was observed that the crack propagation path significantly influenced the beam's natural frequencies and resonance amplitudes. The results indicated a decreasing frequency trend and an increasing amplitude trend as the propagation angle changed from vertical to inclined. A similar trend was observed when the crack path changed from a predominantly vertical orientation to a more complex path with varying angles. Using ANNs, a model was developed to predict natural frequencies and amplitudes from the given crack paths, achieving a high accuracy with a mean absolute percentage error of 1.564%.