Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций

Конференция: II International Scientific Conference "Advanced Technologies in Aerospace, Mechanical and Automation Engineering"; Krasnoyarsk; Krasnoyarsk

Год издания: 2020

Идентификатор DOI: 10.1088/1757-899X/734/1/012100

Аннотация: Dynamic system identification is commonly reduced to an optimization problem which is complex and multimodal. To find a global optimum of this problem, evolutionary algorithms are often applied. However, as it was shown in many studies, conventional evolution-based algorithms do not demonstrate the acceptable performance for this cПоказать полностьюlass of problems, therefore, some effective modifications have been proposed so far. In our study, we combine two approaches which were previously used in linear dynamic system identification and allowed their accurate identification. More specifically, we present a cooperative evolutionary algorithm with a restart metaheuristic and apply it for the parameter identification of a nonlinear cascaded system. The experimental results prove the effectiveness of the proposed evolution-based identification compared to other known solutions of this problem.

Журнал: IOP Conference Series: Materials Science and Engineering

Выпуск журнала: 734

Номера страниц: 12100

Издатель: Institute of Physics and IOP Publishing Limited

• Brester C (Reshetnev Siberian State University of Science and Technology)
• Ryzhikov I (Reshetnev Siberian State University of Science and Technology)
• Stanovov V (Reshetnev Siberian State University of Science and Technology)
• Semenkin E (Reshetnev Siberian State University of Science and Technology)
• Kolehmainen M (University of Eastern Finland)

• Ядро РИНЦ (eLIBRARY.RU)