A review of the machine learning approach for predicting residential natural gas demand and consumption
The only byproducts of burning natural gas are carbon dioxide, water vapor, and very little amounts of nitrogen oxide, making it the cleanest fossil fuel in the world. Natural gas is also used to power a large variety of consumer goods, including furnaces, dryers, fireplaces, and stoves. You definitely use natural gas to power at least one of your appliances. This study provides a complete evaluation of the machine learning forecasting approaches currently in use for estimating residential natural gas demand and consumption. The paper discusses the main difficulties in estimating natural gas demand, including the complicated and dynamic interaction between weather patterns and energy consumption as well as the diversity of homes and their energy consumption habits. The paper also provides an overview of the various machine learning methods and algorithms, such as artificial neural networks, decision tree models, and regression-based approaches that are utilized for extrapolating natural gas demand. The research examines these methods' precision, memory, and accuracy in addition to their benefits and drawbacks. In its last section, the report makes several recommendations for more research, including the use of data from smart meters and the implementation of advanced machine learning techniques, such as deep learning, to improve forecast accuracy. The state-of-the-art in machine learning for natural gas prediction and consumption is generally covered in this review, which is an essential resource for academics and business experts.
Keywords: Demand, Consumption, Energy, Machine learning and Natural Gas
*Correspondence: aminuagabus@gmail.com, 07032132336
Download Paper