A balanced layout problem of cylinders in a cylindrical container of the minimal radius

We study a balanced layout problem of a collection of homogeneous circular cylinders onto the given bearing plates of a cylindrical container of minimal radius  taking into account behavior constraints. We consider a reduced model of a spacecraft as the mechanical system. The latter is formed by mea...

Повний опис

Збережено в:
Бібліографічні деталі
Дата:2015
Автори: Панкратов, А. В., Романова, Т. Е., Коваленко, А. А.
Формат: Стаття
Мова:Russian
Опубліковано: Journal of Mechanical Engineering 2015
Теми:
Онлайн доступ:https://journals.uran.ua/jme/article/view/40261
Теги: Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
Назва журналу:Journal of Mechanical Engineering

Репозитарії

Journal of Mechanical Engineering
Опис
Резюме:We study a balanced layout problem of a collection of homogeneous circular cylinders onto the given bearing plates of a cylindrical container of minimal radius  taking into account behavior constraints. We consider a reduced model of a spacecraft as the mechanical system. The latter is formed by means of a cylindrical container with placed objects (technical equipment) onto the given bearing plates. Behavior constraints include dynamic equilibrium, moments of inertia, stability constraints. A mathematical model of the problem is constructed in the form of nonlinear programming problem, using phi-functions. We develop the efficient algorithm, involving the multistart method, an algorithm for constructing a set of feasible starting points and IPOPT to solve nonlinear programming problems. In order to simplify a nontrivial procedure of searching for a feasible starting point we apply a special algorithm, which is based on homothetic transformations of circles. The proposed solution method allows us: to search for local optimal solutions for the balanced layout problem of cylinders in a cylindrical container of the minimal radius, improve a convergence of the local optimization and reduce the computational time. We present a number of known benchmark instances to demonstrate the high efficiency of our approach.