A Novel Parameter Estimation Fuzzy Network for Target Approximation
Publish place: 9th Annual Conference of Computer Society of Iran
Publish Year: 1382
نوع سند: مقاله کنفرانسی
زبان: English
View: 2,128
This Paper With 8 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
ACCSI09_108
تاریخ نمایه سازی: 4 بهمن 1386
Abstract:
Here we introduce a novel Neuro-Fuzzy system for approximation and prediction of complex plants. The proposed system uses parameter estimation and scaling techniques; so called PENTA which stands for Parameter Estimation Network for Target Approximation. PENTA consists of two major components: a scaling component and a parameter estimation unit. The first component provides the gains required by the second component which is a set of primitive functions used for scaling. So the overall output of network is a synthetic function that is linear composition of these parameters and primitive functions. These two components have been expanded through five layers of a Neuro-Fuzzy network. Experimental results show that in comparison with current Neuro-Fuzzy systems like ANFIS, this novel system can have a better adaptation to complex plants. Also results show the unique capability of this system, in consecutive prediction of chaotic systems
Keywords:
Authors
Kourosh Neshatian
Iran Telecom Research Center North Kargar St., Tehran, Iran
Mohammad Teshnehlab
K. N. Toosi University Sharyatee St., Tehran, Iran