A Hybrid LSTM Neural Network for Energy Consumption Forecasting of Individual Households

Ke Yan,Meng Qi,Wei Li,Yang Du,Zhiwei Ji

doi:10.1109/access.2019.2949065

Abstract

Irregular human behaviors and univariate datasets remain as two main obstacles of data-driven energy consumption predictions for individual households. In this study, a hybrid deep learning model is proposed combining an ensemble long short term memory (LSTM) neural network with the stationary wavelet transform (SWT) technique. The SWT alleviates the volatility and increases the data dimensions, which potentially help improve the LSTM forecasting accuracy. Moreover, the ensemble LSTM neural network further enhances the forecasting performance of the proposed method. Verification experiments were performed based on a real-world household energy consumption dataset collected by the ‘UK-DALE project. The results show that, with a competitive training efficiency, the proposed method outperforms all compared state-of-art methods, including the persistent method, support vector regression (SVR), long short term memory (LSTM) neural network and convolutional neural network combining long short term memory (CNN-LSTM), with different step sizes at 5, 10, 20 and 30 minutes, using three error metrics.

Highlights

Data-driven energy consumption forecasting methods, which predicts the short-term future energy consumption values based on historical data, are important elements in the process of building information modeling (BIM) [1], [21], [36]
We propose a hybrid deep learning forecasting framework that combines multiple long short term memory (LSTM) neural networks with stationary wavelet transforms (SWT) to solve the irregular and univariate individual household energy consumption forecasting problem
A realworld dataset containing energy consumption data of five households located in London, United Kingdom (UK) is utilized for verification purposes

Summary

Introduction

Data-driven energy consumption forecasting methods, which predicts the short-term future energy consumption values based on historical data, are important elements in the process of building information modeling (BIM) [1], [21], [36]. The short-term and very short-term energy consumption forecasting techniques are useful for residential energy demand-side management, electricity price market design, energy efficiency and maintenance scheduling of large-scale complex smart power grids [12], [28], [37]. It provides extra reliability, security and protection for the smart grid to handle the increasing energy demand from residential households. Along with the fast development of artificial intelligence (AI) technology, the extended deep learning methods nowadays are capable of performing very short-term energy consumption forecasting results with spectacularly high prediction accuracy [7], [22], [29], [34]. Univariate time series data, e.g., the energy consumption data, forecasting is a challenging problem even for deep learning technologies

Methods

Results

Conclusion