Deep-based conditional probability density function forecasting of residential loads

Mousa Afrasiabi, Mohammad Mohammadi, Mohammad Rastegar, Lina Stankovic, Shahabodin Afrasiabi, Mohammad Khazaei

Research output: Contribution to journalArticlepeer-review

116 Citations (Scopus)
168 Downloads (Pure)

Abstract

This paper proposes a direct model for conditional probability density forecasting of residential loads, based on a deep mixture network. Probabilistic residential load forecasting can provide comprehensive information about future uncertain-ties in demand. An end-to-end composite model comprising convolution neural networks (CNNs) and gated recurrent unit (GRU) is designed for probabilistic residential load forecasting. Then, the designed deep model is merged into a mixture density network (MDN) to directly predict probability density functions (PDFs). In addition, several techniques, including adversarial training, are presented to formulate a new loss function in the direct probabilistic residential load forecasting (PRLF) model. Several state-of-the-art deep and shallow forecasting models are also presented in order to compare the results. Furthermore, the effectiveness of the proposed deep mixture model in characterizing predicted PDFs is demonstrated through comparison with kernel density estimation, Monte Carlo dropout, a combined probabilistic load forecasting method and the proposed MDN without adversarial training
Original languageEnglish
Article number8988175
Pages (from-to)3746-3757
Number of pages12
JournalIEEE Transactions on Smart Grid
Volume11
Issue number4
Early online date10 Feb 2020
DOIs
Publication statusPublished - 31 Jul 2020

Keywords

  • residential load forecasting
  • conditional probabilistic load forecasting
  • deep mixture network

Fingerprint

Dive into the research topics of 'Deep-based conditional probability density function forecasting of residential loads'. Together they form a unique fingerprint.

Cite this