Articles
<![CDATA[Smart Aquaculture Vision: Deteksi dan Klasifikasi Ikan Otomatis Menggunakan YOLOv8 ]]>
<![CDATA[Riyo Irawan]]>;
<![CDATA[Tresna Dewi]]>;
<![CDATA[Pola Risma]]>;
<![CDATA[Yurni Oktarina]]>
<![CDATA[ Jurnal Rekayasa Elektro Sriwijaya Vol. 6 No. 2 (2025): Jurnal Rekayasa Elektro Sriwijaya ]]>
Publisher : <![CDATA[Jurusan Teknik Elektro Fakultas Teknik Universitas Sriwijaya ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.36706/jres.v6i2.157
<![CDATA[Akuakultur modern menuntut sistem pemantauan ikan yang efisien dan akurat guna meningkatkan produktivitas dan keberlanjutan. Penelitian ini mengusulkan pemanfaatan algoritma deteksi objek berbasis deep learning, yaitu YOLOv8, untuk mendeteksi dan mengklasifikasikan tiga jenis ikan secara otomatis: Black Spotted Barb, Gourami, dan Mosquito Fish. Dataset yang digunakan berasal dari Kaggle dan terdiri atas 730 gambar yang telah dilabeli ulang menggunakan Roboflow. Proses pelatihan dilakukan di Google Colab dengan konfigurasi GPU, batch size 32, selama 100 epoch. Model dievaluasi menggunakan metrik presisi, recall, dan mAP. Hasil evaluasi menunjukkan performa yang sangat baik dengan nilai precision sebesar 0.978, recall sebesar 0.928, mAP50 sebesar 0.973, dan mAP50-90 sebesar 0.616. Temuan ini membuktikan bahwa YOLOv8 mampu memberikan deteksi objek yang akurat dan efisien, serta berpotensi untuk diterapkan dalam sistem pemantauan akuakultur berbasis visi komputer secara real-time.]]>
<![CDATA[Model Prediksi Deep Learning dengan Pendekatan Feedforward Neural Network ]]>
<![CDATA[Lukman Nul Hakim]]>;
<![CDATA[Tresna Dewi]]>;
<![CDATA[Pola Risma]]>;
<![CDATA[Yurni Oktarina]]>
<![CDATA[ Jurnal Rekayasa Elektro Sriwijaya Vol. 6 No. 2 (2025): Jurnal Rekayasa Elektro Sriwijaya ]]>
Publisher : <![CDATA[Jurusan Teknik Elektro Fakultas Teknik Universitas Sriwijaya ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.36706/jres.v6i2.158
<![CDATA[Penelitian ini bertujuan untuk menerapkan model deep learning, khususnya Feedforward Neural Network (FNN), dalam meramalkan nilai irradiance berdasarkan data waktu. Solar irradiance sangat penting dalam pengembangan sistem energi terbarukan seperti panel surya untuk meningkatkan efisiensi sistem tenaga surya dan mengoptimalkan perencanaan sumber daya energi. Penggunaan model ini diharapkan dapat memberikan prediksi yang lebih akurat dan handal, sehingga mendukung pengambilan keputusan dalam pengelolaan energi terbarukan secara berkelanjutan. Untuk meningkatkan akurasi prediksi, penelitian ini menerapkan teknik preprocessing data yang mencakup penghapusan nilai hilang dan normalisasi menggunakan MinMaxScaler guna meningkatkan stabilitas pelatihan model. Model FNN yang diusulkan terdiri dari beberapa lapisan tersembunyi dengan aktivasi non-linear untuk menangkap pola kompleks dalam data, serta lapisan output untuk menghasilkan prediksi akhir. Pelatihan model dilakukan menggunakan algoritma optimasi seperti Adam, dengan fungsi aktivasi ReLU untuk meningkatkan konvergensi. Evaluasi model dilakukan menggunakan metrik RMSE, MSE, MAE, dan R-squared (R²) sebagai indikator utama keakuratan model dalam peramalan irradiance. Hasil evaluasi menunjukkan bahwa model ini mampu memberikan prediksi yang akurat terhadap pola irradiance, dengan nilai RMSE dan MAE yang rendah serta R² mendekati satu, menandakan kinerja yang sangat baik dalam menangkap dinamika data.]]>
<![CDATA[Simulation Design of Artificial Intelligence Controlled Goods Transport Robot ]]>
<![CDATA[Oktarina, Yurni]]>;
<![CDATA[Sastiani, Destri Zumar]]>;
<![CDATA[Dewi, Tresna]]>;
<![CDATA[Kusumanto, RD]]>
<![CDATA[ Computer Engineering and Applications Journal (ComEngApp) Vol. 11 No. 2 (2022) ]]>
Publisher : <![CDATA[Universitas Sriwijaya ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
<![CDATA[Technological advances enable scientists and researchers to develop more automated systems for life's convenience. Transportation is among those conveniences needed in daily activities, including warehouses. The easy-to-build and straightforward transport robot are desired to ease human workers' working conditions. The application of artificial intelligence (AI), Fuzzy Logic Controller, and Neural Network ensures the robot is able to finish assigned tasks better and faster. This paper shows the concept design of an AI-controlled good transport robot applied in the warehouse. The design is made as fast and straightforward forward possible, and the feasibility of the proposed method is proven by simulation in Scilab FLT and Neuroph.]]>
<![CDATA[SYNERGY OF GREEN ENERGY AND SMART TECHNOLOGY: APPLICATION OF RECURRENT NEURAL NETWORKS IN SOLAR-POWERED AGRICULTURE ]]>
<![CDATA[Maulidina, Elfira]]>;
<![CDATA[Dewi, Tresna]]>;
<![CDATA[Kusumanto, Raden]]>
<![CDATA[ International Journal of Mechanics, Energy Engineering and Applied Science (IJMEAS) Vol. 3 No. 2 (2025): IJMEAS - May ]]>
Publisher : <![CDATA[Yayasan Ghalih Pelopor Pendidikan (Ghalih Foundation) ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.53893/ijmeas.v3i2.406
<![CDATA[In an effort to improve energy efficiency and sustainability in the agricultural sector, smart technology has been integrated into the greenhouse system. The research utilizes the Recurrent Neural Network (RNN) algorithm to forecast values of irradiance on a time principal. The RNN algorithm is chosen for its ability to handle time-series data and predict patterns based on historical data. By using the RNN algorithm, the system can predict real-time needs and then use this information to optimally distribute power from solar power plants. Additionally, this system is equipped with Internet of Things (IoT)-based monitoring capabilities, allowing remote monitoring and control of the research object. Connected IoT sensors collect real-time environmental data and send it to the data server for analysis. The data is also used to update the model of RNN, making supply prediction more accurate over time. The implementation results show increased energy efficiency and reduced operational costs in Green House management. By leveraging AI and IoT technology, model evaluation is conducted using RMSE, MSE, MAE, and R-squared (R²) metrics as important indicators of model accuracy. The evaluation results indicate that this model can provide accurate predictions of irradiance patterns, with low RMSE and MAE values and R² approaching one, signifying excellent implementation in capturing data dynamics.]]>
<![CDATA[Internet of Things Based Temperature and pH Stabilization Control System in The Pome Biodigester Fermentation Process at PLTBg ]]>
<![CDATA[Mustofa]]>;
<![CDATA[Dewi, Tresna]]>;
<![CDATA[Bow, Yohandri]]>
<![CDATA[ International Journal of Mechanics, Energy Engineering and Applied Science (IJMEAS) Vol. 3 No. 3 (2025): IJMEAS - September ]]>
Publisher : <![CDATA[Yayasan Ghalih Pelopor Pendidikan (Ghalih Foundation) ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.53893/ijmeas.v3i3.433
<![CDATA[The Government of Indonesia is striving to reduce dependence on fossil fuels by increasing the use of renewable energy (RE), with a target of achieving 23% by 2025. Biogas Power Plants (PLTBg) are one of the solutions that utilize palm oil mill effluent (POME) to produce biogas through anaerobic fermentation. The currently operating systems face problems and challenges in monitoring and operating mesophilic digesters, particularly covered lagoon types, which are managed conventionally, resulting in frequent drops in temperature and pH levels. To address these issues, automation was implemented in the temperature and pH control systems, with the expectation of increasing biogas production. This study adopted an approach that involved recording temperature and pH data, analyzing their upper and lower thresholds, and developing a laboratory-scale model that simulates industrial conditions. This model was equipped with a temperature and pH control system, along with monitoring and control coding. In the laboratory-scale POME digester prototype using actual POME liquid as the test medium, it was demonstrated that temperature and pH could be effectively monitored and controlled by automatically regulating the POME pump motors. The study results show that temperature control was maintained within the range of 38–41 °C, and pH control within the range of 6.5–8. The temperature drop from 40 °C to 35 °C occurred over 274 minutes. The temperature control response time was 5.6 seconds. The pH decrease rate due to the addition of fresh POME was 2.04, with a pH control response time of 8.5 seconds.]]>
<![CDATA[Implementasi Deep Learning Dalam Prediksi Real-Time Iradian Surya ]]>
<![CDATA[Liwijaya, Angga]]>;
<![CDATA[Risma, Pola]]>;
<![CDATA[Oktarina, Yurni]]>;
<![CDATA[Dewi, Tresna]]>
<![CDATA[ Journal of Applied Smart Electrical Network and Systems Vol. 6 No. 2 (2025): Vol. 6 No. 02 (2025): Vol 06, No. 02 Desember 2025 ]]>
Publisher : <![CDATA[Indonesian Society of Applied Science (ISAS) ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52158/pvdpsr36
<![CDATA[Accurate prediction of solar irradiance plays a critical role in the planning and operation of renewable energy systems, particularly for photovoltaic integration and energy management. This study investigates the use of a deep learning approach based solely on Convolutional Neural Networks (CNN) to forecast short-term solar irradiance values. The model is trained using normalized multivariate time series data, which include several meteorological parameters as input features. The CNN architecture is designed to extract temporal patterns from the input sequences and predict radiation intensity at the next time step. Experimental results show that the proposed model achieves strong predictive performance, with a Mean Squared Error (MSE) of 0.0006, Root Mean Squared Error (RMSE) of 0.0242, Mean Absolute Error (MAE) of 0.0184, and a coefficient of determination (R²) of 0.9607. These findings demonstrate that CNN, despite its simplicity, is capable of effectively learning complex temporal relationships in solar irradiance data. Furthermore, the loss curves for both training and validation sets indicate stable convergence without signs of overfitting. The results suggest that CNN-based forecasting models can offer a lightweight and accurate solution for real-time solar prediction applications, especially when computational resources are limited.]]>
<![CDATA[SIMULASI KEMACETAN LALU LINTAS PADA LOKASI BUNDARAN BALTOS BANDUNG ]]>
<![CDATA[Dewi, Tresna]]>;
<![CDATA[Badruzzaman, Farid]]>;
<![CDATA[Fajar, Yusuf]]>;
<![CDATA[Suhaedi, Didi]]>;
<![CDATA[Harahap, Erwin]]>
<![CDATA[ Smart Comp :Jurnalnya Orang Pintar Komputer Vol 9, No 2 (2020): Smart Comp :Jurnalnya Orang Pintar Komputer ]]>
Publisher : <![CDATA[Politeknik Harapan Bersama ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30591/smartcomp.v9i2.1768
<![CDATA[Pada masa kini perkembangan teknologi telah memacu meningkatnya kepemilikan kendaraan, terutama di beberapa kota seperti Jakarta, Bandung, Yogyakarta, dan kota besar lainnya. Situasi ini menimbulkan masalah yaitu kemacetan lalu lintas dimana salah satu penyebabnya adalah kepemilikan kendaraan yang semakin bertambah setiap tahun. Kondisi ini tidak didukung dengan keadaan infrastruktur dan sumber daya yang terbatas, disamping fasilitas lalu lintas yang pengoperasiannya masih belum optimal. Penyebab lain dari kemacetan lalu lintas ini adalah banyaknya angkutan umum yang sering berhenti sembarangan, keluar masuk area parkir, dan persilangan kendaraan di persimpangan. Oleh karena itu pada artikel ini dilakukan penelitian untuk mengetahui penyebab dalam masalah kemacetan dan diharapkan ditemukan solusi pemecahan masalah. Dalam penelitian ini, dirancang sistem simulasi arus lalu lintas dengan menggunakan aplikasi SimEvents yang dijalankan pada software MATLAB Simulink. Berdasarkan simulasi, dapat diprediksi penyebab kemacetan lalu lintas khususnya di lokasi Bundaran Balubur Town Square (BALTOS) Kota Bandung Jawa Barat.Kata kunci :Â kemacetan, lalu lintas, simevents, matlab]]>
<![CDATA[Model Deep Learning Hybrid CNN-AE untuk Klasifikasi Presisi Warna Buah Melon ]]>
<![CDATA[Oktarina, Yurni]]>;
<![CDATA[Dewi, Tresna]]>;
<![CDATA[Septiyani AR, Dini]]>
<![CDATA[ Journal of Applied Smart Electrical Network and Systems Vol. 6 No. 2 (2025): Vol. 6 No. 02 (2025): Vol 06, No. 02 Desember 2025 ]]>
Publisher : <![CDATA[Indonesian Society of Applied Science (ISAS) ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.52158/d3hydf28
<![CDATA[Melon fruit color classification is a critical step in assessing fruit ripeness and quality. This study proposes a hybrid deep learning model that integrates Convolutional Neural Network (CNN) and Attention Enhancement (AE) for accurate classification of melon fruit color. The model leverages CNN’s strength in visual feature extraction while enhancing focus on crucial image regions through the attention mechanism. A diverse image dataset of melon fruits was collected under various lighting conditions and angles. Pre-processing steps, including data augmentation, normalization, and image scaling, were applied to improve model generalization. The CNN-Attention hybrid architecture incorporates an attention module into the CNN layers to emphasize significant features. Comparative experiments between the standard CNN and the hybrid model demonstrate that the latter achieves superior classification accuracy, with an average improvement of 5%. Moreover, the hybrid model exhibits better robustness against image noise and lighting variations. These results indicate that incorporating Attention Enhancement can yield a more adaptive and reliable model for melon fruit color classification. The proposed approach is expected to support the development of automated systems for fruit sorting in agriculture and distribution, enhancing speed, accuracy, and efficiency for farmers, traders, and consumers.]]>
<![CDATA[Eligibility Study on Floating Solar Panel Installation over Brackish Water in Sungsang, South Sumatra ]]>
<![CDATA[Sasmanto, Andri Agus]]>;
<![CDATA[Dewi, Tresna]]>;
<![CDATA[Rusdianasari]]>
<![CDATA[ EMITTER International Journal of Engineering Technology Vol 8 No 1 (2020) ]]>
Publisher : <![CDATA[Politeknik Elektronika Negeri Surabaya (PENS) ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24003/emitter.v8i1.514
<![CDATA[Electric generation using the photovoltaic (PV) effect is considered ideal in South Sumatra as a response to the government policy to increase the utilization of renewable energy to support the depletion of conventional energy. PV panels can be installed in a fishing village in the Sungsang Estuary. This paper examined the eligibility analysis for the installation of PV panels on brackish water. In this research, two Panels are installed, the first one is floating over a water body, and the second is ground mounted as a comparison of electricity produced and efficiency. The Jsc floating and ground mounting differ in 0.4435 A. The measured Jload in floating PV panels is 0.3900 A higher than the ground mounting. The measured Voc at the floating PV panels is 0.2935 V higher, and the Vload of the floating PV panel is 3.0742 V higher than the ground mount. The differences are due to the floating PV panel surface temperature being lower than ground mounting. Electricity generated by floating PV panels is averagely 11.89 Watt higher, and the efficiency is 4% higher than that of ground installation. This experiment also shows that PV panels can be installed over brackish water in the fishing village of Sungsang Estuary.]]>
