Rancang Bangun Sistem Monitoring Deteksi Dini Gangguan Overload dan Power Fluctuation pada Sistem Listrik 1 Fasa Berbasis Threshold, Simple Moving Average, dan Fuzzy Sugeno

Peningkatan penggunaan energi listrik pada sektor rumah tangga maupun industri menyebabkan risiko terjadinya gangguan seperti overload dan power fluctuation yang dapat menurunkan keandalan sistem serta berpotensi merusak peralatan listrik. Sistem proteksi konvensional umumnya hanya bekerja ketika gangguan telah terjadi secara langsung, sehingga diperlukan suatu sistem early warning yang mampu mendeteksi perubahan kondisi beban secara lebih dini. Penelitian ini bertujuan merancang dan mengimplementasikan sistem monitoring serta deteksi dini gangguan overload dan power fluctuation pada instalasi listrik satu fasa berbasis Internet of Things (IoT) menggunakan metode hybrid Threshold, Simple Moving Average (SMA), dan Fuzzy Sugeno. Sistem dikembangkan menggunakan mikrokontroler ESP32, sensor PZEM-004T untuk akuisisi data kelistrikan, OLED sebagai media tampilan lokal, serta platform Blynk untuk monitoring jarak jauh. Metode Threshold digunakan sebagai klasifikasi awal kondisi sistem, SMA untuk menghitung nilai acuan dan deviasi daya sebagai indikator fluktuasi, serta Fuzzy Sugeno untuk menghasilkan keputusan status akhir sistem. Hasil penelitian menunjukkan bahwa sistem berhasil melakukan monitoring parameter kelistrikan secara real-time dengan rata-rata error pengukuran tegangan yang sangat kecil, yaitu sebesar 0,55% pada OLED dan 0,56% pada dashboard Blynk. Integrasi metode hybrid terbukti andal dalam mendeteksi dan mengklasifikasikan kondisi sistem menjadi Normal, Waspada, dan Kritis. Sistem berhasil memberikan peringatan dini (early warning) dengan mendeteksi kondisi Waspada Fluktuasi Daya pada rentang deviasi daya 9,5% hingga 19%, dan Waspada Overload pada rentang arus 2,96 A hingga 3,79 A, sebelum akhirnya berubah menjadi status Overload (Kritis) pada arus 4,05 A. Output kondisi sistem tersebut berhasil divisualisasikan dengan akurat melalui LED indikator, buzzer, layar OLED, serta dashboard Blynk.

=====================================================================================================

Rising electricity consumption in both the household and industrial sectors increases the risk of faults such as overloads and power fluctuations, which can compromise system reliability and potentially damage electrical equipment. Conventional protection systems typically operate only after a fault has already occurred; therefore, an early warning system capable of detecting changes in load conditions at an earlier stage is required. This research aims to design and implement an IoT-based monitoring and early warning system for overload and power fluctuation faults in single-phase electrical installations, utilizing a hybrid method combining Threshold, Simple Moving Average (SMA), and Fuzzy Sugeno logic. The system was developed using an ESP32 microcontroller, a PZEM-004T sensor for electrical data acquisition, an OLED display for local monitoring, and the Blynk platform for remote monitoring. The Threshold method serves as the initial system condition classifier; SMA calculates reference values and power deviations to indicate fluctuations; and Fuzzy Sugeno determines the final system status. Research results demonstrate that the system successfully monitors electrical parameters in real-time with minimal measurement error averaging 0.55% on the OLED display and 0.56% on the Blynk dashboard. The integration of these hybrid methods proved reliable in detecting and classifying system conditions into Normal, Alert, and Critical states. The system successfully provided early warnings by detecting Power Fluctuation Alert conditions within a power deviation range of 9.5% to 19% and Overload Alert conditions within a current range of 2.96 A to 3.79 A, prior to transitioning to an Overload (Critical) status at a current of 4.05 A. These system status outputs were accurately visualized via indicator LEDs, a buzzer, the OLED screen, and the Blynk dashboard.