Radial Basis Neural Network Based Islanding Detection in Distributed Generation

سال انتشار: 1393
نوع سند: مقاله ژورنالی
زبان: انگلیسی
مشاهده: 525

فایل این مقاله در 10 صفحه با فرمت PDF قابل دریافت می باشد

این مقاله در بخشهای موضوعی زیر دسته بندی شده است:

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

JR_IJE-27-7_007

تاریخ نمایه سازی: 12 آبان 1393

چکیده مقاله:

This paper investigates a new integrated diagnostic system for islanding detection by means of a neuralnetwork approach for distributed generation. Islanding is an important concern for grid connecteddistributed resources due to personnel and equipment safety. Several methods based on passive andactive detection scheme have been proposed. While passive schemes have a large non detection zone(NDZ); concern has been raised on active method due to its degrading power quality effect. Reliablydetecting this condition is regarded as an ongoing challenges in existing methods are not totalysatisfactory. The main emphasis of the proposed scheme is to reduce the NDZ to as close as possibleand to keep the output power quality unchanged. In addition, this technique can also overcome theproblem of setting the detection thresholds inherent in the existing techniques. In this study, wepropose to use a radial basis neural network for islanding detection.The proposed algorithm iscompared with the widely used rate of change of frequency relays (ROCOF) and was found to workeffectively in situations where ROCOF fails. This approach utilizes rate of change of frequency at thetarget distributed generation location and was fed to the radial basis neural network for intelligentislanding detection. Hence a better reliability is provided. This approach utilizes the artificial neuralnetwork (ANN) as a machine learning technology for processing and analyzing the large data setsprovided from network simulations using MATLAB software. To validate the feasibility of thisapproach, the method has been validated through several conditions and different loadings, switchingoperations, and network conditions. Simulation studies showed that the RBNN-based algorithm detectsislanding situation more accurately than other islanding detection algorithms.

کلیدواژه ها:

Keywords:Distributed GenerationIslanding DetectionNon Detection ZoneRate of Change of FrequencyRadial Basis Neural Network

نویسندگان

m Tarafdar

Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran

n Ghadimi

bArdabil Branch, Islamic Azad University, Ardabil, Iran