Fuzzy grey forecasting model optimized by moth-flame optimization algorithm for short time electricity consumption

TANYOLAÇ BİLGİÇ C., BİLGİÇ B., Çebi F.

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, cilt.42, sa.1, ss.129-138, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 42 Sayı: 1
  • Basım Tarihi: 2022
  • Doi Numarası: 10.3233/jifs-219181
  • Dergi Adı: JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Communication Abstracts, Compendex, Computer & Applied Sciences, INSPEC, Metadex, zbMATH, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.129-138
  • Anahtar Kelimeler: Grey forecasting, MFO-TFGM (1,1), parameter optimization, moth-flame optimization, TFGM (1,1), ENERGY-CONSUMPTION, CHINA, REGRESSION, ARIMA, MFO
  • İstanbul Teknik Üniversitesi Adresli: Evet

Özet

It is significant that the forecasting models give the closest result to the true value. Forecasting models are widespread in the literature. The grey model gives successful results with limited data. The existing Triangular Fuzzy Grey Model (TFGM (1,1)) in the literature is very useful in that it gives the maximum, minimum and average value directly in the data. A novel combined forecasting model named, Moth Flame Optimization Algorithm optimization of Triangular Fuzzy Grey Model, MFO-TFGM (1,1), is presented in this study. The existing TFGM (1,1) model parameters are optimized by a new nature- inspired heuristic algorithm named Moth-Flame Optimization algorithm which is inspired by the moths flying path. Unlike the studies in the literature, in order to improve the forecasting accuracy, six parameters (lambda(L), lambda(M), lambda(R), alpha, beta and -gamma) were optimized After the steps of the model is presented, a forecasting implementation has been made with the proposed model. Turkey's hourly electricity consumption data is utilized to show the success of the prediction model. Prediction results of proposed model is compared with TFGM (1,1). MFO-TFGM (1,1) performs higher forecasting accuracy.