International Conference on Computer and Knowledge Engineering

Home / 12th International Conference on Computer and Knowledge Engineering

A parallel CNN-BiGRU network for short-term load forecasting in demand-side management

Authors :

Arghavan Irankhah¹ Sahar Rezazadeh Saatlou² Mohammad Hossein Yaghmaee³ Sara Ershadi-Nasab⁴ Mohammad Alishahi⁵

1- Department of Computer Engineering Ferdowsi University of Mashhad Mashhad, Iran 2- Department of Computer Engineering Ferdowsi University of Mashhad Mashhad, Iran 3- Department of Computer Engineering Ferdowsi University of Mashhad Mashhad, Iran 4- Department of Computer Engineering Ferdowsi University of Mashhad Mashhad, Iran 5- Research center of smart distribution networks

Keywords :

Deep Learning

Abstract :

Nowadays power companies are trying to monitor energy consumption to provide demand response. Energy management and scheduling are possible through short-term load forecasting. Energy supply stability and efficiency depend on accurate forecasting, which balances demand and supply. In this paper, a novel day-ahead residential load forecasting method is introduced. A new parallel deep learning network is presented that is based on CNN and GRU networks. Some features are extracted from the dataset during pre-processing. The CNN models extract more information from these features in two parallel paths. By observing the input both forward and in reverse directions, bi-GRU networks are used to learn long dependency patterns. The proposed method is evaluated using real-world data collected by the Mashhad energy distribution company. In comparison with state-of-the-art methods, the proposed method has the lowest RMSE, MAE, and MAPE values of 49.04, 34.37, and 3.81, respectively.