IoT and Machine Learning on Smart Home-Based Data and a Perspective on Fog Computing Implementation

IoT and Machine Learning on Smart Home-Based Data and a Perspective on Fog Computing Implementation

Asha Rajiv, Abhilash Kumar Saxena, Digvijay Singh, Aishwary Awasthi, Dharmesh Dhabliya, R. K. Yadav, Ankur Gupta
Copyright: © 2023 |Pages: 14
DOI: 10.4018/978-1-6684-8785-3.ch017
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This study emphasises the need for energy efficiency in buildings, focusing primarily on the heating, ventilation, and air conditioning (HVAC) systems, which consume 50% of building energy. A predictive system based on artificial neural networks (ANNs) was created to generate short-term forecasts of indoor temperature using data from a monitoring system in order to reduce this energy use. The technology seeks to estimate inside temperature in order to determine when to start the heating, ventilation, and air conditioning system, potentially reducing energy use dramatically. The chapter describes the system's code implementation, which includes data pre-processing, model training and evaluation, and result visualisation. In terms of evaluation metrics, the model performed well and revealed the potential for large energy savings in buildings.
Chapter Preview
Top

Introduction

The article addresses the use of an Artificial Neural Network (ANN) system to estimate the indoor temperature of a room, using data from a complex monitoring system serving as input. The purpose is to save energy by anticipating the indoor temperature and deciding whether or not to turn on the Heating, Ventilation, and Air Conditioning (HVAC) system. Because building energy consumption accounts for a large fraction of overall global energy consumption, the article emphasises the necessity of energy efficiency-related methods and optimisation approaches in buildings to minimise global energy consumption. The installation of such devices could serve as a wake-up call for building owners, urging them to reassess their operations and begin investigating new technologies. In addition to the ANN system, the paper discusses fog computing, a decentralised computing infrastructure that allows data processing and storage to be closer to the data source. Fog computing has the ability to increase the ANN system's efficiency and accuracy, as well as other building automation systems.

Fog Computing to Analyse Smart Home Based Data

Fog computing is a paradigm that has evolved as a solution to the centralised cloud computing model's difficulties. It entails deploying computing resources near the network's edge, closer to the data source, in order to give a more distributed and decentralised approach to data processing and analysis (Lamiae et al., 2023). Fog computing can provide an efficient approach to process data received from numerous sensors and devices deployed within a building in the context of building energy management. Data may be processed and analysed in real-time by putting fog nodes closer to these devices, lowering latency and enhancing the overall efficiency of the building energy management system (Aqib et al., 2023). Fog computing can also offer a more scalable and adaptable approach to building energy management, allowing building owners to swiftly deploy new sensors and devices as needed and react to changing energy use trends. Fog computing is a type of distributed computing infrastructure that extends cloud computing to the network's edge (Katal et al., 2023). It allows for data processing and analysis to take place closer to the source of the data (Padhy et al., 2023). Fog computing can be used to process and analyse data from sensors and devices located in buildings in the context of building energy usage (Anoushee et al., 2023). Fog computing, for example, may analyse data from temperature sensors, occupancy sensors, lighting systems, and HVAC systems to find patterns and anomalies in energy consumption (Anoushee et al., 2023).

IoT in Smart Homes Data and Fog Computing

The application of machine learning and IoT on smart home data to reduce energy usage is demonstrated in this code implementation. The goal is to estimate indoor temperature using previous data and then utilise the projections to make more energy-efficient HVAC decisions. To provide a short-term forecast of indoor temperature, the system employs an Artificial Neural Network (ANN) model. This forecast enables the smart home to adjust to future temperature circumstances and operate the HVAC system in the most energy-efficient way possible. Furthermore, this application emphasises the significance of energy efficiency tactics and optimisation approaches in structures. HVAC's high power use, accounting for a large percentage of total consumption, mandates the employment of these strategies to reduce building energy consumption. The implementation analyses the data and generates helpful insights by utilising several numerical and data visualisation libraries. Data pretreatment techniques such as feature engineering, date encoding, and ordinal encoding are also used to prepare the data for the ANN model. Finally, the implementation assesses the ANN model's performance using several measures where the model's evaluation measures demonstrate its ability to predict indoor temperature and reduce energy consumption.

Complete Chapter List

Search this Book:
Reset