Claim Missing Document
Check
Articles

Found 2 Documents
Search

A Concept of Household Gas Leak Safety System Using MQ-2 Sensor and Embedded-Based Alarm Simanjuntak, Christopel; Kumaat, Arief Perdana; Joroh, Adelaida; Mpila, Deby A.
Jurnal Elektrik Vol. 5 No. 1 (2026): Vol.5 No.1 1 Juni 2026
Publisher : Jurusan Teknik Elektro - Politeknik Negeri Manado

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65485/elektrik.v5i1.1284

Abstract

Liquefied Petroleum Gas (LPG) leakage in domestic and small-industrial sectors constitutes a major threat that frequently triggers fatal accidents, including severe fires and structural explosions. The primary factor behind these high accident rates is the limitation of human senses in detecting gas accumulation during the early stages of micro-leaks, compounded by the lack of continuous environmental monitoring systems. This study aims to formulate a comprehensive concept for an early detection system design for LPG leakage, integrating the MQ-2 semiconductor gas sensor with an Arduino microcontroller as the primary controller. The core focus of this paper is on modeling the hardware architecture, designing the system programming logic, and conducting a theoretical analysis of the integrated warning alarm scheme.The approach method applied in this conceptual research includes extensive literature reviews, logical flowchart design, electronic circuit schematic modeling, and mathematical simulation of component response times. The proposed system concept is engineered to monitor gas particle concentration fluctuations in the air in real-time. Based on the established theoretical parameters, when the MQ-2 sensor detects LPG particles exceeding the predefined safe threshold (set above 400 ppm), the Arduino microcontroller will immediately process the analog signal into a digital command to execute emergency protocols. This action triggers a high-frequency audible alarm via a buzzer and a visual indicator through a red LED light to provide instant warnings to building occupants.The conceptual design results demonstrate that the integration between Arduino and the MQ-2 sensor yields a security system scheme that is not only highly responsive and rapid in calculation but also economical and easy to mass-produce. This paper contributes significant scientific value by delivering a solid, valid, and applicable conceptual blueprint, which can serve as a primary reference for the physical development and manufacturing of smart gas leak detection devices in the future to enhance residential safety standards.
a Development of a Self-Sufficient AI Ecosystem Concept for Msty-based Multi-Models Simanjuntak, Christopel; Kumaat, Arief Perdana; Joroh, Adelaida; Rumokoy, Stieven N; Mpila, Deby A; Warokka, Adriyan
Prosiding Seminar Nasional Produk Terapan Unggulan Vokasi Vol 5 No 1 (2026): Prosiding Seminar Nasional Produk Terapan Unggulan Vokasi Ke-6 Politeknik Negeri
Publisher : Pusat Penelitian dan Pengabdian Kepada Masyarakat Politeknik Negeri Manado

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Infrastruktur cloud pihak ketiga menuntut pengembangan Artificial Intelligence (AI), tetapi sering kali hal ini disertai kekhawatiran privasi, pelanggaran kedaulatan data, dan biaya operasional yang mahal. Namun demikian, belakangan ini, paradigma pengembangan AI lokal (on-premise) telah mendapatkan daya tarik sebagai alternatif yang lebih baik. Makalah ini membahas gagasan dan fondasi pembangunan ekosistem AI otonom untuk lingkungan multi-model di Msty. Metode penelitian adalah studi konseptual dan pendekatan eksperimental untuk menguji ruang kerja desktop terintegrasi seperti itu dari model bahasa besar untuk pembuatan teks dan model lokal untuk menghasilkan gambar. Hasil penelitian menunjukkan bahwa Msty dapat mengoordinasikan interaksi multi-model secara bersamaan melalui metode manajemen memori internal yang efisien tanpa ketergantungan komputasi eksternal. Menerapkan ekosistem ini pada akhirnya mencapai kedaulatan data penuh dan fleksibilitas dalam alur kerja bagi pengembang. Penelitian ini menyimpulkan bahwa Msty efektif mempercepat demokratisasi AI mandiri, sekaligus menjadi aplikasi tepat untuk pengembangan kecerdasan buatan yang aman, privat, dan hemat biaya pada perangkat keras tingkat konsumen.