Claim Missing Document
Check
Articles

Found 34 Documents
Search

IMPLEMENTASI IOT DAN GAUSSIAN NAÏVE BAYES UNTUK SISTEM MONITORING KANDANG AYAM BROILER Randily, Dhea Ramadhanti Putri; Rismawan, Tedy; Sari, kartika
Jurnal Informatika dan Teknik Elektro Terapan Vol 13, No 2 (2025)
Publisher : Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/jitet.v13i2.6243

Abstract

Abstrak. Ayam Broiler merupakan unggas penghasil daging yang sensitif terhadap suhu, kelembapan, dan amonia. Apabila suhu dan kelembapan kandang tidak berada dalam kondisi ideal, maka ayam akan melakukan panting dan memperlambat metabolisme dengan cara menjauhi tempat pakan. Panting adalah kondisi dimana mulut ayam selalu terbuka untuk menstabilkan suhu tubuh. Selain itu, gas amonia yang tinggi dapat menyebabkan gangguan pernapasan pada ayam. Berdasarkan masalah tersebut, dibangun sebuah sistem monitoring kandang ayam broiler berbasis Internet of Things (IoT)dengan metode Gaussian Naive Bayes untuk membantu peternak dalam memonitoring kondisi kandang agar selalu berada dalam kondisi ideal dengan mengklasifikasikan menjadi dua kategori yaitu ideal dan buruk berdasarkan tiga parameter yaitu suhu, kelembapan, dan amonia. Sistem ini menggunakan NodeMCU ESP32 dengan output sistem yaitu lampu pijar, exhaust fan, dan mist maker. Lampu pijar digunakan sebagai pemanas kandang. Exhaust fan digunakan untuk menurunkan suhu kandang saat berada di atas kondisi ideal, menjaga sirkulasi udara dan mengontrol amonia kandang. Mist maker digunakan untuk meningkatkan kelembapan udara di dalam kandang. Pengujian dilakukan dengan data pelatihan sebanyak 573 data dan data pengujian sebanyak 246 data, diperoleh akurasi sebesar 91,87%. Hasil pengujian menunjukkan bahwa sistem mampu dalam mengatur suhu, kelembapan, dan amonia pada kandang ayam broiler.
Utilization of Artificial Intelligence as a Tool to Assist in Preparation of School Learning Materials on the West Kalimantan Border Ristian, Uray; Ruslianto, Ikhwan; Bahri, Syamsul; Hidayati, Rahmi; Rismawan, Tedy; Suhery, Cucu; Marisa Midyanti, Dwi; Nirmala, Irma; Suhardi; Hasfani, Hirzen; Sari, Kartika; Kasliono; Muhardi, Hafiz
MEKONGGA: Jurnal Pengabdian Masyarakat Vol. 2 No. 1 (2025): April 2025
Publisher : Digital Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69616/mekongga.v2i1.217

Abstract

The use of Artificial Intelligence (AI) as a tool in compiling learning materials in schools has great potential to improve the quality of education, especially in border areas that are often faced with challenges of accessibility and availability of resources. This community service activity aims to fulfill the Tri Dharma of Higher Education by providing contributions to the community while assisting teachers in utilizing AI tools. Through a preliminary study, a needs mapping was carried out together with teachers of SDN 03 Balai Karangan to ensure that the development of AI tools is relevant to learning needs on the West Kalimantan border. The development of this AI tools integrates natural language processing technology to produce learning materials that are in accordance with the curriculum and local characteristics. After development, the AI ??tools is disseminated and practiced in the school environment. It is hoped that through this activity, teachers at SDN 03 Balai Karangan on the West Kalimantan border will be better able to compile quality and relevant learning materials with the help of AI tools, can improve the quality of education and provide real benefits to local students.
LITERATURE REVIEW: IOT-BASED GEOGRAPHIC INFORMATION SYSTEM FOR MONITORING SOIL CHEMICAL PROPERTIES IN OIL PALM PLANTATIONS Banyuriatiga, Banyuriatiga; Sari, Kartika; Syaddam, Syaddam; Rahman, Arief; Adlian, Adlian
Jurnal Dialektika Informatika (Detika) Vol 5, No 2 (2025): Jurnal Dialektika Informatika(Detika) Vol.5 No.2 Mei 2025
Publisher : Universitas Muria Kudus

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24176/detika.v5i2.15401

Abstract

Oil palm has a crucial role in the National economy, where its productivity depends on the soil chemical properties of the soil. In practice, monitoring methods carried out using conventional methods can be time-consuming and expensive. Therefore, a more efficient approach is needed. By reviewing previous research, this study highlights the latest developments, challenges, opportunities, and future directions in utilising the Internet of Things and Geographic Information Systems for monitoring soil chemical properties in oil palm plantations, thereby supporting more productive and sustainable management. The descriptive method was applied in this literature review research to 26 relevant articles selected from various publication databases, covering the time frame from 2020 to 2024. This study examines the development of IoT sensors for monitoring soil parameters, including humidity, pH, and temperature, in real-time, as well as the application of GIS for spatial analysis and data visualization. The results highlight the significant potential of integrating IoT and GIS to provide efficient real-time data and spatial analysis, thereby supporting more precise land management and informed decision-making, particularly concerning soil fertility and fertilizer use.
Sistem Klasifikasi Kualitas Udara dengan Integrasi Sensor menggunakan Metode K-Nearest Neighbor Achyar, Athif Tafrihan; Hidayati, Rahmi; Sari, Kartika
JEPIN (Jurnal Edukasi dan Penelitian Informatika) Vol 11, No 2 (2025): Volume 11 No 2
Publisher : Program Studi Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26418/jp.v11i2.90728

Abstract

Kualitas udara yang buruk dapat berdampak negatif pada sistem respirasi manusia dan dampak kesehatan lainnya, sehingga pemantauan kualitas udara berperan penting terhadap permasalahan ini. Penelitian ini mengembangkan sistem klasifikasi kualitas udara secara real-time menggunakan metode K-Nearest Neighbor (KNN) dan data dari 4 sensor: GP2Y1010AU0F (PM₂.₅), MQ-135 (CO₂), MQ-131 (O₃), dan MQ-7 (CO). Sistem terdiri dari sensor yang terintegrasi dengan ESP32, penyimpanan cloud menggunakan Firebase, dan antarmuka web untuk pemrosesan serta visualisasi data. ESP32 berfungsi mengumpulkan data dari sensor dan mengirimkannya ke Firebase, yang kemudian diakses oleh aplikasi web untuk klasifikasi menggunakan algoritma KNN. Klasifikasi dilakukan dalam 4 kategori: baik, sedang, buruk, dan sangat buruk, dengan hasil ditampilkan di antarmuka web untuk pemantauan selama 24 jam terakhir. Pengujian menggunakan confusion matrix dengan 900 data latih dan 600 data uji menunjukkan tingkat akurasi sebesar 98,67%. Selain itu, pengujian crossvalidation dengan k=3 menghasilkan akurasi sebesar 99,39%.
Combining IoT and Time Series Model for Minute-Level Outlier Detection in Wind Speed Forecasting Huda, Nur'ainul Miftahul; Imro'ah, Nurfitri; Hidayati, Rahmi; Sari, Kartika
CAUCHY: Jurnal Matematika Murni dan Aplikasi Vol 10, No 2 (2025): CAUCHY: JURNAL MATEMATIKA MURNI DAN APLIKASI
Publisher : Mathematics Department, Universitas Islam Negeri Maulana Malik Ibrahim Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.18860/cauchy.v10i2.35768

Abstract

Renewable energy optimisation and early warning systems require accurate short-term wind speed forecast. Anomalies in environmental data impair forecasting model reliability. This paper presents an integrated approach using IoT-based remote sensing and time series modelling to address the issue. IoT-based anemometer sensors collected wind speed data at one-minute intervals from December 24, 2024, to January 10, 2025. Aggregating the raw data into 5-minute intervals prepared it for the ARIMA model. This model determined temporal patterns and predicted short-term wind speeds. Analyzing residuals between observed and predicted results helped identify wind outliers. This approach is novel because it uses IoT-based continuous sensing and time series modeling for real-time environmental monitoring. Studies showed that a 65-minute frame with 5-minute intervals was best for replicating wind speed dynamics. Six cycles of outlier detection found 87 outliers. The ARIMA model improved predictions by include these outliers as exogenous variables. This emphasizes the importance of fixing time series model anomalies to improve prediction. The augmented ARIMA model with outlier corrections provides minute-level forecasts and reliable anomaly identification for renewable energy optimization and early warning systems. This study shows that new statistical methods and the Internet of Things (IoT) can improve real-time environmental and energy decisions.
Implementasi mekanisme sleep-wake pada sistem pendeteksi kebocoran gas LPG Mahardika, Nanda Rizky; Nirmala, Irma; Sari, Kartika
JITEL (Jurnal Ilmiah Telekomunikasi, Elektronika, dan Listrik Tenaga) Vol. 5 No. 3: September 2025
Publisher : Jurusan Teknik Elektro, Politeknik Negeri Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35313/jitel.v5.i3.2025.221-232

Abstract

Liquefied Petroleum Gas (LPG) merupakan salah satu sumber energi yang paling banyak digunakan di seluruh dunia, baik di sektor rumah tangga maupun industri. Pemanfaatannya memberikan manfaat besar dalam pemenuhan kebutuhan energi, peningkatan kualitas hidup, dan mendukung perkembangan industri. Namun demikian, kebocoran pada tabung, regulator, maupun selang penghubung LPG dapat menimbulkan risiko kebakaran. Penelitian ini mengimplementasikan mekanisme sleep-wake pada sistem pendeteksi kebocoran gas berbasis mikrokontroler ESP32 untuk mengoptimalkan konsumsi daya pada perangkat elektronik, khususnya mikrokontroler dan sensor. Sensor MQ-6 digunakan untuk membaca kadar LPG di udara sekitar dalam satuan ppm, sementara sensor suara FC-04 berfungsi mendeteksi bunyi desis kebocoran dan mengirimkan sinyal external interrupt untuk membangunkan ESP32 dari kondisi deep sleep . Data hasil pengukuran dikirim ke server ThingSpeak dan peringatan dikirimkan melalui Telegram sesuai dengan kondisi kadar gas yang terdeteksi. Hasil penelitian menunjukkan bahwa sistem mampu mendeteksi kadar LPG dengan akurasi sebesar 97%. Sistem juga berhasil mengirimkan data kadar gas ke ThingSpeak dan notifikasi Telegram secara bersamaan. Berdasarkan pengujian, sistem dengan mode deep sleep  dapat beroperasi selama 5 jam 47 menit (347 menit), sedangkan sistem tanpa deep sleep  hanya bertahan 4 jam 11 menit (251 menit), dengan tingkat optimisasi daya sebesar 27,7%.
Perbandingan Kalman Filter dan Exponential Moving Average pada Sensor Ultrasonik dalam Sistem Smart Waste ATM Wijaya, Andy; Suhardi; Sari, Kartika
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 2 (2025): July
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jst-undiksha.v14i2.101049

Abstract

Waste level monitoring is still often done manually, making it inefficient in preventing accumulation. Ultrasonic sensors are widely used because they are practical and affordable, but their accuracy is often affected by environmental and hardware conditions. This study aims to compare the Kalman Filter and Exponential Moving Average methods to improve the accuracy of ultrasonic sensor readings in an automated waste monitoring system. The type of research used is an experiment with a microcontroller-based system that is tested on various waste height variations. The Kalman Filter combines previous estimates with new data, while the Exponential Moving Average gives more weight to the most recent value. The performance of both methods is assessed based on measurement consistency and error rate.The data was then analyzed quantitatively using Root Mean Square Error (RMSE).The results show that the Kalman Filter produces lower errors and more stable data compared to the Exponential Moving Average or raw data. In conclusion, the Kalman Filter is more effective in improving the reliability and accuracy of the automated waste monitoring system. The implications of this research suggest that selecting the right sensor type can significantly improve system performance in detecting waste capacity in real time.
Sistem Deteksi Kebocoran Gas LPG Berbasis IOT Menggunakan Flame Sensor, MQ-6, dan LM35 Baiti, Nurul; Suhery, Cucu; Sari, Kartika
Jurnal Rekayasa Teknologi Informasi (JURTI) Vol 9, No 3 (2025): Jurnal Rekayasa Teknologi Informasi (JURTI)
Publisher : Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30872/jurti.v9i3.17767

Abstract

Liquefied Petroleum Gas (LPG) merupakan sumber energi yang banyak digunakan di rumah tangga dan industri, namun berpotensi menimbulkan bahaya kebocoran yang dapat menyebabkan kebakaran dan ledakan. Penelitian ini bertujuan mengembangkan sistem deteksi kebocoran gas berbasis Internet of Things (IoT) dengan tiga jenis sensor, yaitu MQ-6 untuk mendeteksi kadar gas LPG, flame sensor untuk mendeteksi keberadaan api, dan sensor LM35 untuk mengukur suhu. Data dari sensor dikirim ke NodeMCU ESP32 melalui koneksi WiFi dan diproses untuk menghasilkan notifikasi secara real time melalui Bot Telegram. Sistem juga dilengkapi aktuator otomatis berupa kipas, buzzer, dan servo-regulator sebagai respons awal terhadap potensi bahaya. Hasil pengujian menunjukkan tingkat akurasi keseluruhan sebesar 93,33%, dengan sensor MQ-6 mendeteksi gas di atas 200 ppm, flame sensor berfungsi akurat, dan sensor LM35 memiliki akurasi lebih dari 99,5% dibandingkan termometer konvensional. Sistem mampu mendeteksi gas pada rentang 23-403 ppm dan suhu 26,05°C-38,98°C. Temuan ini menunjukkan bahwa integrasi multi-sensor, aktuator otomatis, dan notifikasi real time dapat meningkatkan efektivitas deteksi dini serta memperkuat keamanan penggunaan LPG di rumah tangga maupun industri kecil.
Control and Monitoring System for Lighting on Dragon Fruit Plants Kafriandeni, Okhari; Suhery, Cucu; Sari, Kartika; Malik, Abdul
Jurnal Media Informasi Teknologi Vol. 2 No. 2 (2025): Oktober 2025
Publisher : Digital Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69616/mit.v2i2.238

Abstract

Dragon fruit plants require prolonged sunlight exposure to grow and develop optimally. One common practice among farmers is to provide artificial lighting at night. However, this lighting is still operated manually, which is considered less effective. To address this issue, this study developed a lighting control and monitoring system for dragon fruit plants. The system uses a NodeMCU ESP32 as the main controller, an LDR sensor for light detection, a relay module for lamp control, and Firebase with Flutter as the storage platform and real-time monitoring interface. The system is designed to automatically turn on the lamp when the light intensity detected by the LDR sensor falls below a predetermined threshold and to turn it off when the light is sufficient. Implementation of the system over two weeks produced a positive response to dragon fruit flowering, with blossoms appearing at several points on the plants. Accuracy testing of both the sensor readings and the automatic control system was conducted by comparing the number of correct data points with the total collected data. From a total of 5,131 data points recorded during testing, 4,745 matched the system’s logic, resulting in an accuracy of 92.47%. This result indicates that the system performs reliably in providing artificial lighting to extend the photosynthesis process in dragon fruit plants.
Implementation of Door Lock System Using Keyword and RFID Suwanty, Erina; Hidayati, Rahmi; Sari, Kartika
Jurnal Media Informasi Teknologi Vol. 2 No. 2 (2025): Oktober 2025
Publisher : Digital Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69616/mit.v2i2.240

Abstract

Doors are an important element in a building that serve as a connection between rooms. Generally, door locking systems are still conventional, which have several shortcomings, such as being easy to duplicate. Therefore, this study developed a door locking system using a keyword and RFID (Radio Frequency Identification), as combining two security systems enhances security quality. Testing was conducted by collecting data on the performance of the Voice Recognition module at randomly selected distances ranging from 20 to 100 cm, with intervals of 20 cm between the user's voice and the microphone. Then, the RFID capability was tested under two different conditions: the first condition where the RFID card was not covered by a cover, and the second condition using a cover to simulate the condition of the card being inside a wallet or case, and finally, the overall system testing. Based on the test results conducted in this study, using both sensors with a total of 40 data points, the overall system achieved 100% accuracy.