A Neural Network-Based Model for Wind Farm Output in Probabilistic Studies of Power Systems
محل انتشار: بیست و یکمین کنفرانس مهندسی برق ایران
سال انتشار: 1392
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 1,281
فایل این مقاله در 6 صفحه با فرمت PDF قابل دریافت می باشد
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
این مقاله در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
ICEE21_710
تاریخ نمایه سازی: 27 مرداد 1392
چکیده مقاله:
The penetration of wind energy in power systems has been growing due to its interminable and mild environmental effects. The intrinsic attributes of this environmentally-friendly energy, i.e., the stochastic nature of wind farms generation, however, imposes various technical and financial challenges into power systems. So, developing an accurate wind farm modeling approach aimed at taking into account the wind generation intermittency can relieve many of these challenges. Therefore, this paper takes a step to an efficient wind farm modeling procedure employing an accurate as well as well-known Neural Network (NN)-based tool. The proposed approach is comprised of two main steps. The wind speed is predicted in the first step by the time-series NN method. It then continues in the second step with extracting the predicted wind farm outputs and representing the wind farms generation via a multistate model. To this end, an efficient clustering method, i.e., the Fuzzy c-means Clustering Method (FCM) is employed to find the optimal states. Having applied the proposed approach on the case studies, it will confirm that the proposed analytical multistate model can efficiently consider the wind farms output uncertainties in various studies like the conventional units with less computational burden
کلیدواژه ها:
Fuzzy c-means Clustering Method (FCM) ، multistate model ، time-series Neural Network (NN) ، wind farm output modelling ، wind speed forecasting
نویسندگان
S Riahinia
Control, Sharif University of Technology
M Fotuhi-Firuzabad
Sharif University of Technology