Goal of the paper is to present and comment the results of applying the Invasive Weed Optimization (IWO) Algorithm for solving the two-phase inverse Stefan problem. Idea of the IWO algorithm is based on the dynamic growth and reproduction of the weeds colony which makes it a part of the artificial intelligence algorithms inspired by the behaviors from real world. Inverse Stefan problem is the ill-posed boundary value problem describing thermal processes with the change of phase and it consists in reconstruction of some missing elements by using some additional information.
PL
Celem pracy jest przedstawienie procedury służącej do rozwiązania dwufazowego odwrotnego zagadnienia Stefana przy użyciu algorytmu inwazji chwastów IWO (Invasive Weed Optimization algorithm), naśladującego zachowanie populacji chwastów. Rozważane zadanie polegać będzie na odtworzeniu współczynnika wnikania ciepła tak, aby funkcje opisujące położenie granicy rozdziału faz oraz rozkłady temperatury spełniały model matematyczny zadania. Jedna z zasadniczych części procedury polegać będzie na minimalizacji odpowiednio skonstruowanego funkcjonału za pomocą algorytmu IWO.
In this paper, an algorithm will be presented that enables solving the two-phase inverse Stefan problem, where the additional information consists of temperature measurements in selected points of the solid phase. The problem consists in the reconstruction of the function describing the heat transfer coefficient, so that the temperature in the given points of the solid phase would differ as little as possible from the predefined values. The featured examples of calculations show a very good approximation of the exact solution and stability of the algorithm.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Purpose: This paper presents the method of the calculation of the cooling condition in the two-dimensional solidification processes. Design/methodology/approach: The considered problem consists in the reconstruction of the function that describes the heat transfer coefficient on the boundary, when the temperature measurements in selected points of the solid phase are well-known. In calculations the alternating phase truncation method, the genetic algorithm and the Tikhonov regularization were used. Findings: The calculations show a very good approximation of the exact solution and the stability of the procedure. Research limitations/implications: On the bases of the results that every start-up of the genetic algorithm leads to similar results, which are reflected by very low values of the standard deviation. Originality/value: The calculations point to the stability of the proposed method in view of the input data errors, number of control points, substantiating the usability of such approach.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.