CIVILICA We Respect the Science
(ناشر تخصصی کنفرانسهای کشور / شماره مجوز انتشارات از وزارت فرهنگ و ارشاد اسلامی: ۸۹۷۱)

Discrete Sliding Control of Nonlinear Non affine Systems Using Neural Network and Approximate Models

عنوان مقاله: Discrete Sliding Control of Nonlinear Non affine Systems Using Neural Network and Approximate Models
شناسه ملی مقاله: ICEE14_228
منتشر شده در چهاردهمین کنفرانس مهندسی برق ایران در سال 1385
مشخصات نویسندگان مقاله:

Saeed Mohammad-Hoseini - Iran University of Science and Technology
Mohammad Farrokhi

خلاصه مقاله:
In this paper, an indirect combined controller is proposed for non-affine nonlinear systems based on affine approximate input-output model of the system. Here, the NARMA model, which is an exact representation of the input-output model, is approximated with an affine input-output model. The conventional controllers which are based on this approximated model dose not guarantee good performance of the closed-loop system for large magnitude control signal so, in this paper a robust combined controller is proposed which guarantees a reliable performance for the closed-loop system, even for considerable model approximation errors.

کلمات کلیدی:
Sliding mode, nonlinear systems, neural networks, model approximation

صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/54899/