Sistem Keamanan dan Monitoring Kualitas Air Kolam Ikan Lele Berbasis IoT

Anwar, Rahmat Fairun and , Ir. Pratomo Budi Santosa, M.T (2026) Sistem Keamanan dan Monitoring Kualitas Air Kolam Ikan Lele Berbasis IoT. Skripsi thesis, Universitas Muhammadiyah Surakarta.

	PDF (Naskah Publikasi) Artikel Publikasi Ilmiah.pdf Download (749kB)
	PDF (Surat Pernyataan Publikasi) surat pernyataan publikasi perpus.pdf Restricted to Repository staff only Download (664kB)

Abstract

This research discusses the design and implementation of an Internet of Things (IoT)-based fish pond security and monitoring system using an ESP32 microcontroller integrated with a pH sensor, a DS18B20 temperature sensor, a DS3231 Real-Time Clock (RTC) module, an I2C LCD, relays, a servo motor, a Telegram Bot, and Computer Vision technology based on the YOLOv8 algorithm. The system is designed to monitor fish pond conditions in real time by measuring water pH and temperature, displaying the data on an LCD, and sending information to users via Telegram, enabling remote monitoring and device control. In addition, the system provides both online and offline operating modes, which can be selected using a push button. For the security aspect, the YOLOv8 algorithm is employed to detect the presence of humans around the fish pond in real time. When a human is detected, the system automatically records a video, sends a notification via Telegram, and triggers the ESP32 to actuate a servo motor automatically. The experimental results show that the DS18B20 temperature sensor achieved an average error of 0.5%, the pH sensor produced an average error of 2.25%, and the DS3231 RTC module exhibited an average time deviation of 1–2 seconds compared with the reference time. Furthermore, testing of the mode selection button and relay control demonstrated that both functions operated normally. Communication testing through Telegram indicated that the system successfully transmitted monitoring data and detection notifications, while also receiving commands for relay and servo motor control with a 100% success rate and a response time ranging from 0 to 1 second. Moreover, human detection testing using the YOLOv8 algorithm achieved a precision of 100% and a recall (sensitivity) of 92–99%, with no False Positive detections but 1–16 False Negative frames across three testing scenarios. These results demonstrate that the proposed system is capable of real-time water quality monitoring, human detection, automatic video recording, notification delivery, and reliable remote device control, thereby improving the efficiency of monitoring and security for IoT-based fish ponds.

Item Type:	Thesis (Skripsi)
Uncontrolled Keywords:	ESP32, Internet of Things, Fish Pond Monitoring, Security System, YOLOv8, Telegram.
Subjects:	S Agriculture > SH Aquaculture. Fisheries. Angling T Technology > TK Electrical engineering. Electronics Nuclear engineering T Technology > Information Technology
Divisions:	Fakultas Teknik > S1 Teknik Elektro
Depositing User:	RAHMAT FAIRUN ANWAR
Date Deposited:	04 Aug 2026 03:12
Last Modified:	04 Aug 2026 03:12
URI:	http://eprints.ums.ac.id/id/eprint/146715

Actions (login required)

View Item