Dimas Irsyad Maulana, Muhammad Zakky Ihsan Udin , Maulana Albaihaqi Artanto , Muhammad Ilman Salamun, Mirza Syahir Nur Ramadhan

Implementasi Sistem Smarthome Berbasis Ai dan IoT untuk Deteksi Keborosan Listrik Menggunakan K-Means Clustering dan Kontrol Real-Time melalui Aplikasi Android

Introduction

Implementasi sistem smarthome berbasis ai dan iot untuk deteksi keborosan listrik menggunakan k-means clustering dan kontrol real-time melalui aplikasi android. Sistem smarthome AI & IoT deteksi keborosan listrik real-time. K-Means clustering & kontrol Android tingkatkan efisiensi energi rumah. Pantau & hemat listrik secara cerdas.

Abstract

In the modern era, energy management in households has become a significant challenge, especially in reducing electricity wastage. This study proposes a smart home system based on AI and IoT capable of detecting electricity wastage in real-time using the KMeans clustering algorithm. The system utilizes IoT sensors such as DHT, PIR, and ACS712 to measure environmental data, human presence, and current consumption. The received data is analyzed on the server using a machine learning model trained with historical datasets. The electricity wastage predictions are sent to an Android application, enabling users to monitor and control household devices such as lights and fans in real-time. This system aims to provide an effective solution to enhance household energy efficiency while delivering a smarter and more integrated control experience.

Review

This study presents a timely and highly relevant solution for a significant contemporary challenge: household electricity wastage. The authors propose an integrated smart home system leveraging both Artificial Intelligence (AI) and the Internet of Things (IoT) to achieve real-time detection and control. The core strength lies in its practical approach, combining sensor-based data acquisition with machine learning-driven analysis to empower users with immediate insights and capabilities to manage their energy consumption via a dedicated Android application. This foundational concept promises a tangible impact on enhancing household energy efficiency and fostering a smarter, more integrated living experience. Methodologically, the system is well-structured, employing a combination of crucial IoT sensors including DHT for environmental parameters, PIR for human presence, and ACS712 for current consumption, ensuring comprehensive data capture. The data collected is then processed on a server where a machine learning model, specifically utilizing the K-Means clustering algorithm, is trained on historical datasets to identify patterns indicative of electricity wastage. The predictions derived from this analysis are subsequently relayed to an Android application, providing users with a real-time interface to monitor their usage and directly control household appliances such as lights and fans. This end-to-end architecture, from data acquisition to intelligent analysis and user-friendly control, highlights a practical and implementable design. While the abstract outlines a promising system, a more detailed elaboration on certain aspects would strengthen the overall contribution. For instance, clarifying how the K-Means clustering algorithm specifically defines and categorizes "electricity wastage" within its clusters, along with the characteristics and size of the historical datasets used for training, would provide deeper insight into the model's efficacy. Future work could also benefit from discussing the system's performance metrics, such as the accuracy of wastage detection and the latency of real-time control, to fully validate its claims. Additionally, an exploration into the security and privacy implications inherent in smart home systems would be a valuable addition to ensure user confidence and adoption.

Full Text

Comments

You need to be logged in to post a comment.