| Iidentification of High Dimension System by Using New Local Model Network Structure |
| کد مقاله : 1150-CFIS (R2) |
| نویسندگان |
|
سید محمد عماد اولیایی1، محمد تشنه لب *2 1آزمایشگاه سیستمهای هوشمند، دانشکده برق و کامپیوتر ، دانشگاه صنعتی خواجه نصیر الدین طوسی، تهران ، ایران 2دانشگاه خواجه نصیرالدین طوسی |
| چکیده مقاله |
| High dimensional systems challenge the system identification tools that is called curse of dimension. In this paper, the LMN structure has been developed, and a new incremental algorithm has been proposed to stand against this phenomenon. The new algorithm is based on Genetic algorithm and LOLIMOT algorithm that is called GLOLIMOT. The proposed idea covers the disadvantages of training in fuzzy systems and interpretations in neural networks at the same time, also reduces and optimizes the search space. Finally, the proposed idea is tested on single-shaft industrial gas turbine prototype model, which has high complexity and high dimension. The results indicate improvement in performance of the proposed structure and algorithm. |
| کلیدواژه ها |
| LMN, LOLIMOT, Genetic Algorithm, GLOLIMOT, System Identification, Gas Turbine |
| وضعیت: پذیرفته شده برای ارائه شفاهی |