1
Development of a Framework to Integrate Smart Home and Energy Operation Systems to Manage Energy Efficiency Through AI

Sasikala P.1*, S. Sivakumar2, Murali Kalipindi3 and Makhan Kumbhkar4

1Department of Computer Science, Government Science College (Nrupathunga University), Bangalore, Karnataka, India

2Department of Electrical and Electronics Engineering, VelTech Rangarajan Dr. Sagunthala R and D Institute of Science and Technology, Avadi, Chennai, Tamil Nadu, India

3Department of Artificial Intelligence and Machine Learning, Vijaya Institute of Technology for Women (Affiliated to JNTUK), Enikepadu, Andhra Pradesh, India

4Department of Computer Applications, ICAR-Indian Institute of Soybean Research, Indore, India

Abstract

Diverse smart home and Internet of things (IoT)-related products are fast evolving to provide customers with enhanced convenience and ease. However, the current crop of intelligent houses needs to be improved by a lack of operating systems that can adequately connect many components of the smart home ecosystem. This is true since these devices are manufactured using self-service modules and run on proprietary IoT platforms built by the manufacturer. Without a unified OS, managing many components of a smart home may quickly become a bureaucratic nightmare. Moreover, this gives rise to issues such as an overwhelming amount of traffic on the intelligent network for home and inefficiency in energy use. To address these challenges, it is essential to establish a comprehensive management system that facilitates seamless connectivity among IoT devices. To effectively manage the IoT, they suggest implementing three sophisticated models as application services inside an IoT platform designed for smart homes. The three methods under consideration are intelligence awareness targeting as a service (IAT), intelligence energy consumption as a service (IE2S), and intelligence service total access system (TAS) (IST). The IoT oversees the management of the “things” stage. Intelligent Adaptive Technology (IAT) employs advanced machine learning techniques to effectively develop a comprehensive understanding of the contextual information associated with the data values produced by various sensors. This enables the system to gather data on the prevailing environmental conditions efficiently. The IE2 system is a server designed for the IoT platform. Its primary responsibility is to handle and analyze the information gathered by the IAT. The server employs Mobius, an open-source platform by international standards, and a TensorFlow engine for information learning. The IE2S analyses customer use patterns to automate service delivery. Intelligence service TAS manages and provides care for the service period. Three clever models enable IoT devices in smart homes to collaborate actively. Innovative methods may reduce network congestion and energy waste by minimizing needless jobs and adjusting energy use to IoT usage trends in smart homes.

Keywords: Internet of things, smart home, IAT, IE2S, machine learning

1.1 Introduction

Internet of things has recently advanced to the point where it can autonomously recognize and assess its surroundings and apply this understanding to new settings. Because of advancements in the performance of computers, wireless network processing, efficient algorithms, and so on, the speed and precision with which IoT receives and processes instructions has increased dramatically. As more and more algorithms are developed, the IoT is expanding exponentially.

Internet of things applications in smart homes are many. In contrast to the traditional house, which consists mainly of a structure and specific furnishings for living, the smart home prioritizes the user’s happiness and comfort. Beyond its primary function as a place to sleep, eat, and shower, a “smart home” is an expansive collection of IoT apps that enable individuals to regulate their daily cycles, seek delight, and take advantage of novel services.

Numerous studies highlight issues with congestion in networks and power utilization despite the expansion of the IoT industry and related technologies. Currently, the more available technology to mitigate the effects of the widespread adoption of smart home services on energy and data networks is required. As the IoT continues to grow, so does the need for electricity, leading to a surge in the number of studies focusing on renewable energy sources. However, because of the time and resources needed for this new energy generation, it is a risky, long-term undertaking. To address these issues, need network processing that is both autonomous and efficient, as well as a means of controlling the IoT that minimizes wasted energy use by learning from users’ habits.

A smart manager is needed for controlling and managing the data from IoT users, which is crucial for finding solutions to the issues above. For this article, “intelligent manager” is not a reference to an IoT platform that only links IoT devices; instead, it refers to a manager who offers services that create a personalized environment for the customer in keeping with the smart home’s stated goal and utilizes network technology to reduce energy consumption.

The cognitive services provided by IoT platforms nowadays mainly consist of processing massive amounts of data and performing complex mathematical computations much more rapidly and correctly than a human can. Utilizing a set of rules that learn from and analyze user data, a smart home’s smart manager cannot only coordinate the operation of several strategies at once but also tailor its services to each resident. In a smart home, this allows for the anticipation and readiness for a wide range of scenarios and settings. Network and energy consumption may be reduced by appropriately operating and controlling IoT applications in advance, based on analysis of relevant data. Additionally, automated services are provided in a comfortable and pleasant setting by anticipating the user’s sentiments, ideas, and requirements.

This research takes into account prior work to learn about cutting-edge tech and industry trends across several IoT platforms and to identify the need for offering innovative home-specific IoT energy-saving services. This study presents a model for service with the necessary artificial intelligence to improve network and energy efficiency, to the research above.

1.2 Research Idea Definitions

1.2.1 A Service for Intelligence Awareness

A model that processes data from intelligent situational awareness systems at the item level is called intelligence awareness target as a service (IAT). Information assurance technology (IAT) sifts through data produced by objects, collecting just the information that is required in advance. This allows data to be processed in a manner suitable for the circumstance since needless processing is avoided and the essential actions and procedures are carried out.

To anticipate and respond to potential problems, IAT classifies users’ actions and routines into four broad categories. According to this, IAT is conscious of the circumstances, and the necessary data is processed.

Internet of things is broken down in this research into its parts: sensors, mobile devices, and home electronics.

These three gadgets (sensors, smartphones, and smart appliances) are among those that generate and process the most data in a smart home, and they are also among the easiest to access and gather data from about the environment and its inhabitants. The IAT, in broad terms, is split into stationary and non-stationary components. The sensor IAT is standard in stationary IAT, whereas the smartphone IAT is more common in mobile IAT.

Meanwhile, the concept of an IAT that controls a smart appliance may be articulated similarly. Being both an IAT and an IST (intelligence service target as a type of service) sets it apart from similar technologies. This study expands the scope of the IAT discussion to include smart appliances because of their unique characteristics.

1.2.2 IAT Sensor

The sensor IAT model is intended to be used by stationary items to gather primary sensor data. To assess the interior setting and provide concrete numerical values and ideas, a collection of sensors is necessary.

The fixed sensor boards may be seen in each room of the intelligent house, as shown in Figure 1.1. The smart boards have absolute sensors connected to them, and this data may be used to assess the present state of the intelligent house. Sound, motion and rotational sensors are employed. These sensors are used to learn about the state of each smart home’s local area network and to compile data for analysis.

Figure 1.1 Sensor for each network area.

Figure 1.2 Situation types according to detector.

Figure 1.2 illustrates how the ITA may be aware of the interior context in advance by classifying user actions and lifestyle habits into four broad categories. The first scenario involves an action being taken by a user while inside. In the second scenario, the user is at home, relaxing. The third scenario is a user who is sleeping in a home. In the fourth scenario, there is no one home. These grouped scenarios comprehend the user’s whereabouts and movement patterns.

The sensor’s checking shown in Figure 1.2 is straightforward and does not need a massive quantity of data...