Deep learning for load forecasting with smart meter data: Online Adaptive Recurrent Neural Network

• Published in: Applied energy Vol. 282; p. 116177
• Main Authors: Fekri, Mohammad Navid, Patel, Harsh, Grolinger, Katarina, Sharma, Vinay
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.01.2021
• Subjects: Adaptive learning; Bayesian optimization; Concept drift; Online Adaptive RNN; Online learning; Recurrent Neural Network; Adaptive learning; Bayesian optimization; Recurrent Neural Network; Concept drift; Online Adaptive RNN; Online learning
• ISSN: 0306-2619; 1872-9118
• DOI: 10.1016/j.apenergy.2020.116177

Electricity load forecasting has been attracting research and industry attention because of its importance for energy management, infrastructure planning, and budgeting. In recent years, the proliferation of smart meters and other sensors has created new opportunities for sensor-based load forecasting on the building and even individual household level. Machine learning approaches such as Recurrent Neural Networks (RNNs) have shown great successes in load forecasting, but these approaches employ offline learning: they are trained once and miss on the opportunity to learn from newly arriving data. Moreover, they are not well suited for handling the concept drift; for example, their predictive performance will degrade if the load changes due to the installation of new equipment. Consequently, this paper proposes Online Adaptive RNN, an approach for load forecasting capable of continuously learning from newly arriving data and adapting to new patterns. RNN is employed to capture time dependencies while the online aspect is achieved by updating the RNN weights according to new data. The performance is monitored; if it degrades, online tuning is activated to adapt the RNN hyperparameters to changes in data. The proposed approach was evaluated with data from five individual homes: the results show that the proposed approach achieves higher accuracy than the standalone offline long short term memory network and five other online algorithms. Moreover, the time to learn from new samples is only a fraction of the time needed to re-train the offline model. •Energy consumption patterns change over time. Energy data contain concept drift.•Traditional machine learning models require re-training to benefit from new patterns.•Proposed Online Adaptive recurrent neural network learns from data as they arrive.•The proposed approach is developed to handle concept drift in data.•Our model outperforms offline Long Short Term Memory and five other online algorithms.