Implemented IoT based Self-learning Home Management System (SHMS) for INDIA

ABSTRACT

Internet of things (IoT) makes deployment of smart home concept easy and real. Smart home concept ensures residents to control , monitor and manage their energy consumption without any wastage. This paper presents a self-learning Home Management System (SHMS). In the proposed system, a Home Energy Management System (HEMS), Demand Side Management (DSM) system, and Supply Side Management (SSM) system were developed and integrated for real time operation of a smart home. This integrated system has some capabilities such as Price Forecasting (PF), Price Clustering (PC) and Power Alert System (PAS) which to enhance its functions. These enhancing capabilities were developed and implemented using computational and machine learning technologies. In order to vali date the proposed system, real-time power consumption data was collected from a Singapore smart home and a realistic experimental case study was carried out. The case study has shown that the developed system has performed well and created energy awareness to the residents. This proposed system also displays its ability to customize the model for different types of environments compared to traditional smart home models.

EXISTING SYSTEM

In recent years, Internet of Things (IoT) had been widely recognized as an important factor to smart homes. It contributes to allow smart homes to monitor, control and manage the house environment according to homeowner’s lifestyles. Thus, Home Energy Management System (HEMS) in smart home has become the emerging trend research topic in most countries. Most research in smart home focuses on the establishment of communication infrastructure, speed of communication, data transmission reliability, cyber security and enhanced hardware development. It also mentioned the future research on IoT systems for next generation smart homes. However, the data usage methods were not mentioned for smart homes concept. Machine learning is known to be a fascinating technique for smart homes. Its functionality for self-decision making can be incorporated to smart homes energy management systems. Machine learning uses data to learn behaviours patterns to offer optimized solutions. This customized solution accommodating different environment and situations offers HEMS to be a self-learning system. Several researchers had shown that most industrial companies are interested in agents for various applications such as scheduling, resource, strategic planning, control and real-time planning. One of the solutions provided by S.D.J.Mc Arthur,et al.,  was the implementation of Multi-Agent System(MAS). Designing MAS into smart homes systems allows flexible modifications to the agent behaviours when there are changes in environment. MAS network management, control, parallel program design and computer communication were well known to be strong in un uniformed environment such as smart homes.

PROPOSED SYSTEM:

This paper proposed a novel smart home system called Self-learning Home Management System (SHMS). This model proposed using multi-agent system communication network. It includes rule based classifiers technique in the supply and demand side management system. With machine learning  functions in home energy management system. In addition, the functionalities uses computational and machine learning techniques in Price Clustering (PC), Price Forecasting (PF) and Power Alert System (PAS) enhance the capability.

 

CONCLUSIONS :

Self-learning home management system (SHMS) was developed on a Multi-Agent System (MAS) platform, the communications and interactions a agents were implemented on Internet of Things (IoT) principles. Intelligent agents and their interaction play an important part in efficient operation of the smart home. Real simulations based on india smart home shows the SHMS concept capability in using a Demand Side Management (DSM) system, Supply Side Management (SSM) system and Home Energy Management System (HEMS) to make adaptive and intelligent decision based. In addition, the functionalities uses computational and machine learning techniques in Price Clustering (PC), Price Forecasting (PF)

and Power Alert System (PAS) enhance the capability of them proposed system.