Кафедри
Постійне посилання на розділhttps://repository.kpi.kharkov.ua/handle/KhPI-Press/35393
Переглянути
5 результатів
Результати пошуку
Документ Models and computer simulations of mechanical behavior of two-component material for structures data-driven reliability prediction(Національний технічний університет "Харківський політехнічний інститут", 2023) Shapovalova, Mariia I.; Vodka, Oleksii O.This monograph deals with investigation the influence of material microstructure on its mechanical properties. The authors use machine learning and computer vision techniques to assess material microstructure, and computational and experimental methods to determine mechanical characteristics. The theoretical background of material elastic characteristics is investigated, with a focus on processing microstructure images, determining microstructure stress-strain states, and yield criterion. The study also examines the microstructure of dual- component material, generating statistically equivalent artificial microstructures, and evaluating the yield surface. The authors apply data-driven yield surface for structural reliability prediction and evaluate the proposed algorithm for predicting the reliability on example of the Kirsch plate and water pump housing.Публікація Application of machine learning methods for predicting the mechanical behavior of dispersion-strengthened composite material(ФОП Паляниця В. А., 2022) Babudzhan, Ruslan A.; Vodka, Oleksii O.; Shapovalova, Mariia I.The work is devoted for creating a model for approximating the solution by the finite element method of the problem of plane deformation of a dispersion-strengthened composite material. An algorithm for constructing a parametric 2-D composite model is proposed. The processing of the parameters of the microstructure material stress-strain state occurs using a convolutional neural network. A surrogate model is used for calculations speed up and determine the overall approximations quality of such type mechanics problems.Публікація Application of computational intelligence methods for the heterogeneous material stress state evaluation(Національний університет "Одеська політехніка", 2022) Babudzhan, Ruslan A.; Vodka, Oleksii O.; Shapovalova, Mariia I.The use of surrogate models provides great advantages in working with computer-aided design and 3D modeling systems, which opens up new opportunities for designing complex systems. They also allow us to significantly rationalize the use of computing power in automated systems, for which response time and low energy consumption are critical. This work is devoted to the creation of a surrogate model for approximating the finite element solution of the problem of dispersion–strengthened composite plane sample deformation. An algorithm for constructing a parametric two–dimensional model of a composite is proposed. The calculation model is created using the ANSYS Mechanical computer-aided design and analysis program using the APDL scripting model builder. The parameters of the stress-strain state of the material microstructure are processed using a convolutional neural network. A neural network based on the U–Net architecture of the encoder-decoder type has been created to predict the distribution of equivalent stresses in the material according to the sample geometry and load values. A direct sequence of layers is taken from the specified architecture. To increase the speed and stability of training, the type of part of the convolutional layers has been changed. The architecture of the network consists of serially connected blocks, each of which combines layers such as convolution, normalization, activation, subsampling, and a latent space that connects the encoder and decoder and adds load data. To combine the load vector, such a neural network architecture as a concatenator is created, which additionally includes the Dense, Reshape and Concatenate layers. The model loss function is defined as the root mean square error over all points of the source matrix, which calculates the difference between the actual value of the target variable and the value generated by the surrogate model. Optimization of the loss function is performed using the first–order gradient local optimization method ADAM. The study of the model learning process is illustrated by plots of loss functions and additional metrics. There is a tendency for the indicators to coincide between the training and validation sets, which indicates the generalizing capability of the model. Analyzing the output of the model and the value of the metrics, a conclusion is made about the sufficient quality of the model. However, the values of the network weights after training are still not optimal in terms of minimizing the loss function. And also, to accurately reproduce the solution of the finite element method (FEM), the proposed model is quite simple and requires clarification. The speed comparison of obtaining results by the FEM and using the architecture of the neural network is proposed. The surrogate model is significantly ahead of the FEM and is used to speed up calculations and determine the overall quality of the approximation of problems of mechanics of this type.Публікація A Probability Approach to the Prediction of the High-Cycle Fatigue Lifetime Considering Aging Degradation of the Material(Точка, 2013) Larin, Oleksiy; Vodka, Oleksii O.The paper deals with the development of a new approach for high-cycle fatigue lifetime prediction, which is made in the stochastic framework and allows to take into account the natural degradation of the material properties. Mathematical expectation, correlation function and variance of the continuum damage function have been obtained.Публікація Stochastic Dynamics of the Specialized Vehicle with Nonlinear Suspension(NTU "KhPI", 2016) Larin, Oleksiy O.; Vodka, Oleksii O.; Kaidalov, Ruslan O.; Bashtovoi, Volodymyr M.This work deals with the theoretical modelling of the vertical dynamics of the vehicle which has an additional level of suspension for a cargo platform with the nonlinear stiffness. The paper presents the design scheme of the additional level of cushioning having a quasi-zero stiffness in the equilibrium position. The mathematical model of the dynamic behavior of specialized vehicles is developed as a nonlinear discrete system. The results of numeric calculations of the vehicle dynamic response on the stochastic load is represented based on the developed model. Vertical vibrations of the cargo platform caused by the kinematics random influence applied to the axels of the vehicle are analyzed. The load is applied to the axels of the vehicle with a time delay. The results of the comparative analysis are displayed for the frequencies and amplitudes of the vehicle vertical vibrations within two different suspensions: in the linear and nonlinear statements.