Wyniki wyszukiwania - BazTech

1

Convolutional neural networks in the detection of astronomical objects from the Messier catalog

Beluch Witold, Śliwa Paweł

Computer Assisted Methods in Engineering and Science

|

2023

|

Vol. 30, no. 4

461--479

EN

This paper explores the application of convolutional neural networks in the field of amateur astronomy. The authors have employed the available astronomical datasets to develop a detector for identifying astronomical objects from the Messier catalog. A concept framework for creating such a detector for astronomical objects using artificial intelligence tools in the form of a detector based on convolutional neural networks is presented. Augmentation and pre-processing procedures have been used to extend the feature distribution in the training set. Examples confirming the effectiveness of the proposed detector of astronomical objects are presented.

2

Multiobjective global optimization of mechanical systems with cracks

Beluch Witold, Dlugosz Adam

Journal of Theoretical and Applied Mechanics

|

2020

|

Vol. 58 nr 2

553--564

EN

The paper is devoted to the multiobjective shape optimization of cracked structures. The two main goals are: reduction of the negative crack influence of identified cracks and optimal design of structural elements to reduce the risk of crack occurrence and growth. NURBS (Non-Uniform Rational B-Splines) curves are used to model the structure boundaries. Global optimization methods in the form of evolutionary algorithms are employed. As different optimization criteria are considered simultaneously, the efficient multiobjective optimization method are applied. An in-house multiobjective evolutionary algorithm is proposed as an efficient optimization tool. The dual boundary element method is used to solve the boundaryvalue problem.

3

Algorytm mrówkowy do dyskretnej optymalizacji wybranych układów mechanicznych

Beluch Witold, Nowak Damian

Modelowanie Inżynierskie

|

2019

|

T. 39, nr 70

5--9

PL

W artykule przedstawiono algorytm mrówkowy opracowany w celu rozwiązywania zadań globalnej optymalizacji dyskretnej układów prętowych w postaci kratownic. Zastosowane podejście opiera się na algorytmie mrówkowym z funkcją kary, w którym zastosowano ewolucyjną optymalizację strukturalną w celu uwzględnienia optymalizacji topologicznej konstrukcji. Przedstawiony przykład numeryczny ilustruje skuteczność zaproponowanej metody optymalizacji.

EN

The paper presents an Ant Colony Optimization (ACO) algorithm designed to solve global discrete optimization problems of bar systems in the form of trusses. Proposed approach is based on the ACO algorithm with a penalty function in which evolutionary structural optimization has been applied to deal with the topological optimization of trusses. Presented numerical example illustrates the effectiveness of the proposed discrete optimization method.

4

Response surfaces in the numerical homogenization of non-linear porous materials

Beluch Witold, Hatłas Marcin

Engineering Transactions

|

2019

|

Vol. 67, iss. 2

213--226

EN

The paper deals with the numerical homogenization of structures made of non-linear porous material. Material non-linearity causes a significant increase in computational costs for numerical homogenization procedure. An application of the response surface methodology allows a significant reduction of the computational effort providing good approximation precision. Finite element method commercial software is employed to solve the boundary-value problem in both scales. Due to the significant reduction in computing time, the proposed attitude may be applied for different optimization and identification tasks for inhomogeneous, non-linear media, especially with the use of global optimization methods.