Monthly electric load forecasting using transfer learning for smart cities

Seung Min Jung, Sungwoo Park, Seung Won Jung, Eenjun Hwang

Research output: Contribution to journalArticle

Abstract

Monthly electric load forecasting is essential to efficiently operate urban power grids. Although diverse forecasting models based on artificial intelligence techniques have been proposed with good performance, they require sufficient datasets for training. In the case of monthly forecasting, because just one data point is generated per month, it is not easy to collect sufficient data to construct models. This lack of data can be alleviated using transfer learning techniques. In this paper, we propose a novel monthly electric load forecasting scheme for a city or district based on transfer learning using similar data from other cities or districts. To do this, we collected the monthly electric load data from 25 districts in Seoul for five categories and various external data, such as calendar, population, and weather data. Then, based on the available data of the target city or district, we selected similar data from the collected datasets by calculating the Pearson correlation coefficient and constructed a forecasting model using the selected data. Lastly, we fine-tuned the model using the target data. To demonstrate the effectiveness of our model, we conducted an extensive comparison with other popular machine-learning techniques through various experiments. We report some of the results.

Original languageEnglish
Article number6364
JournalSustainability (Switzerland)
Volume12
Issue number16
DOIs
Publication statusPublished - 2020 Aug

Keywords

  • Deep neural network
  • Mid-term load forecasting
  • Monthly electric load forecasting
  • Pearson correlation coefficient
  • Smart city
  • Transfer learning

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Renewable Energy, Sustainability and the Environment
  • Management, Monitoring, Policy and Law

Fingerprint Dive into the research topics of 'Monthly electric load forecasting using transfer learning for smart cities'. Together they form a unique fingerprint.

  • Cite this