Systematic Indoor Room Temperature Forecasting for Smart Buildings Using Machine Learning
Proceedings of 3rd International Conference on Internet of Things and Connected Technologies (ICIoTCT), 2018 held at Malaviya National Institute of Technology, Jaipur (India) on March 26-27, 2018
8 Pages Posted: 8 May 2018
Date Written: April 29, 2018
The growing trend within corporate and residential building structures in India is moving towards self-sustainable techniques like intelligent temperature control, usage of renewable sources of energy and efficient load forecasting. Out of the multiple factors governing the smart buildings, controlled indoor room temperature is considered important, as it drives the need of cooling or heating through air conditioners or heat pumps. Increase in need of self sustainability of building structures and considering Indian climatic and geographical conditions, existing literature suggests that there is a demand to contribute further in this area. This paper explores the application of efficient computational data analysis techniques like Machine Learning to linear time series forecasting methods. It presents the results of a pilot study conducted to perform systematic indoor room temperature forecasting of a corporate building structure, considering Indian climatic and geographical conditions in mind. Out of three models selected for the study, the results show that Multiple Exogenous Inputs fed Auto-Regressive Integrated Moving Average model is the best fit model.
Suggested Citation: Suggested Citation