Federated deep contrastive learning for mid-term natural gas demand forecasting

Abstract

<p>Accurate mid-term gas demand forecasting plays a crucial role for gas companies and policymakers to achieve reliable gas supply plans, supply contracts management, and efficient operation to meet the increasing gas demand. However, mid-term gas demand forecasting faces the problems of data paucity caused by the low frequency of collecting monthly data and heterogeneous consumption patterns of various usage categories. This paper proposes a novel <strong>Fed</strong>erated <strong>Con</strong>trastive pretraining - <strong>L</strong>ocal <strong>C</strong>lustered <strong>F</strong>inetuning paradigm (FedCon-LCF) by integrating federated learning, deep contrastive learning, and clustering approaches. The proposed method can utilize data from multiple gas companies to overcome data paucity issues in a privacy-preserving way, and high-performance forecasting can be achieved by local clustered regression considering the heterogeneous patterns. An improved hierarchical contrastive loss and multi-scale regression loss are integrated to develop the <strong>F</strong>orecasting-<strong>O</strong>riented <strong>C</strong>ontrastive <strong>L</strong>earning model (FOCL), which can effectively extract information and generate fine-grained representations of time series for accurate forecasting. The proposed method is evaluated on a dataset collected from 11 gas companies in 11 different Chinese cities with a total of 17648 clients over 10 usage categories. The proposed method outperforms the benchmark <a href="https://www.sciencedirect.com/topics/engineering/long-short-term-memory" title="Learn more about LSTM from ScienceDirect's AI-generated Topic Pages">LSTM</a> model with an average improvement of 25.30% in <a href="https://www.sciencedirect.com/topics/engineering/mean-squared-error" title="Learn more about MSE from ScienceDirect's AI-generated Topic Pages">MSE</a> and 16.52% in <a href="https://www.sciencedirect.com/topics/engineering/mean-absolute-error" title="Learn more about MAE from ScienceDirect's AI-generated Topic Pages">MAE</a> for 3-month-ahead, 6-month-ahead, 9-month-ahead, and 12-month-ahead gas demand forecasting.</p>