Issuu

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 12 Issue: 05 | May 2025

p-ISSN: 2395-0072

www.irjet.net

AI-POWERED ENERGY MANAGEMENT SYSTEM FOR RESIDENTIAL AREA Prof. J. Kalidass 1, S. Karthika2, S. Kiruthika 3, L.S. Shanu4. 1 Assistant Professor, Department of the Computer Science and Engineering, Government College of Engineering

Srirangam, Tamilnadu, India

2,3,4 Final year UG Student, Department of the Computer Science and Engineering, Government College of

Engineering, Srirangam, Tamilnadu, India -----------------------------------------------------------------------***-----------------------------------------------------------------------

ABSTRACT

the real-time human presence or appliance status hence leading to poor energy management.

In the last years, there is a need for intelligent systems to be developed to make the electricity consumed in the homes optimized and rate wastage reduced. In this paper, we propose an AI-Powered Energy Management System (AIEMS), a novel system, which is computer vision and a deep learning-based system, designed for residential areas. The suggested system combines Convolutional Neural Networks (CNNs), real-time image processing, Internet of Things (IoT), and smart grid ideas to recognize a human presence and appliance status by classification from images. The system utilizes object detection frameworks such as YOLO and face recognition libraries to track the active appliances and the presence of occupants, hence ensuring the appliances are off when not in use. It also comes with fire and smoke detection mechanism that provides real time alerts using email, sms or mobile notifications. Saving Appliance Solution is an interface for visualizing appliance activities, alerts, and recommendations for energy saving. Based on the experimental results, the proposed solution has proven its effectiveness in minimizing the unnecessary energy consumption and the improved monitoring, safety, and decision process in the smart homes. [1]

Smart homes stand to benefit greatly from Artificial Intelligence (AI) and Computer Vision that have the potential to change energy monitoring and control forever. These technology allows systems to make informed decisions based on the analysis of real time visual and contextual data thus ensuring effective energy expenditure without compromising the comfort or security of the user. The role of AI into energy. management systems enables real-time detection of human activities and appliance usage, as well as environmental change/conditions based on real-time video feeds and trained models. This paper introduces an AI-Powered Energy Management System for residential areas that are built on the use of image classification, real-time object and face detection, and IoT-based alerts. Advanced deep learning models like YOLO for object detection and Media Pipe or Haar cascades for pose and face recognition are used for human presence and appliance status monitoring by the system . In case an appliance is determined ON, but there is no human presence, the system gives the alerts and has realtime notifications using various channels such as email, SMS and mobile notifications. Apart from optimizing the consumption of energy, the system includes fire and smoke detection modules for enhanced home safety. Monitoring data, alerts, and any system decision are output through a web dashboard, which provides transparency as well as actionable insights to users. In this paper, the architecture, the methodology, algorithms, techniques, implementation, and programming as well as the performances are explained, showing us the efficiency of AI to manage residential energy consumption efficiently and intelligently. [2][3]

KEYWORDS Artificial Intelligence (AI), Energy Management System (EMS), Smart Home, Computer Vision, YOLO, Deep Learning, Human Detection, Appliance Monitoring, IoT, Fire Detection, Face Recognition, Real-Time Alerting, Energy Efficiency, Smart Grid, Dashboard Visualization.

1. INTRODUCTION The pressure towards generation of more power together with the growth of smart devices as well as home automation has led to the need for formulation of intelligent power management systems. Residential energy consumption plays a significant role in the total energy demand consumption and inefficient consumption patterns end up wasting energy, increased carbon footprints and increased bills. Traditional forms of energy monitoring have always used manual inputs or sensors for detecting energy if this is not properly received, it might not be able to capture

Impact Factor value: 8.315

2. LITERATURE SURVEY Lots of research works have focused on the integration of Artificial Intelligence and IoT in the energy management systems in order to enhance efficiency and automation. The below studies provide a fundamental comprehension of the existing methodologies and their deficit, that our proposed system intends to overcome.

ISO 9001:2008 Certified Journal

Page 887