Jurnal Nasional Teknik Elektro
Not a member yet
359 research outputs found
Sort by
Water Quality Control in Carp Fish Ponds Using Fuzzy Logic
No full textRegularly monitoring pond water quality in fish farming is a crucial practice often neglected, negatively impacting goldfish yields. Addressing this issue, a sophisticated device leveraging fuzzy logic has been engineered to accurately regulate acidity, temperature, and water levels, with real-time data accessible through the Blynk smartphone application. This innovative system employs a trio of sensors—namely an acidity sensor, a DS18B20 temperature sensor, and an HCSR04 ultrasonic sensor—coupled with three output mechanisms: an inlet pump, an outlet pump, and a heater, to ensure precise control. Rigorous testing under various conditions at different times of the day, lasting approximately one hour each, demonstrated the device's capability to adjust water's acidity by about 0.1 units per minute, reflecting the influences of fish activity and water temperature, with a deficient accuracy error of 0.19%. Additionally, the system's effectiveness in maintaining a consistent water level was confirmed, exhibiting a refill rate of 1.2 cm per minute as detected by the sensor. This integrated system is instrumental in safeguarding goldfish health and optimizing their productivity by ensuring water quality remains within the desired acidity, temperature, and volume parameters.Pembudidaya ikan mas koki tidak rutin memperhatikan kualitas air di kolam ikan sehingga mempengaruhi hasil panen ikan mas. Untuk itu dibuatlah suatu alat pengontrol pH, suhu dan ketinggian air dengan menggunakan metode fuzzy logic dan mampu melakukan pemantauan menggunakan aplikasi Blynk melalui smartphone. Pada penelitian ini digunakan 3 buah sensor yaitu sensor pH, sensor suhu DS18B20 & sensor ultrasonik HCSR04, dan 3 keluaran yaitu In pump, Out pump, & heater. Pengujian sistem kontrol menggunakan metode Fuzzy Logic pada pagi, siang, sore dan malam hari serta pada kondisi air asam & basa untuk setiap waktu pengujian ± 1 jam. Dari hasil penelitian diketahui bahwa waktu untuk menurunkan dan menaikkan nilai pH sekitar 0,1/menit. Hal ini antara lain disebabkan oleh pergerakan ikan yang aktif pada siang hari dan suhu air yang tinggi pada siang hari dengan error 0,19%. Untuk ketinggian air di kolam ikan mas di segala kondisi masih dikontrol sesuai ketinggian yang diinginkan dengan waktu pengisian air ke dalam kolam ikan mas yaitu 1,2 jarak terukur dari sensor/menit. Dengan demikian, air kolam ikan mas selalu terjaga dalam toleransi pH, suhu, dan tinggi (volume) air
Bahasa Inggris
No full textWounds result from physical violence that damages the continuity of body tissues and are frequently observed in forensic medicine and medicolegal science. In forensic medicine and medicolegal science, wounds play a significant role in creating a medicolegal examination and report (VeR) for deceased individuals and living victims. However, research findings indicate that the quality of clinical forensic descriptive results in VeR needs to improve in several hospitals in Indonesia. Meanwhile, high-quality VeR results are crucial in determining penalties for perpetrators in court, and poor VeR results can hinder the legal process. The application of information technology in medicine has yielded numerous tools that can assist experts in carrying out their duties. Likewise, clinical forensics, a generally conservative forensic pathology practice, can be enhanced through image-processing techniques and machine learning. Digital technology support for forensic cases has been available previously, such as in forensic photography; however, its application still needs improvement, and further development is required. This study applied a Yolo V4-based machine learning and image processing algorithm to classify and detect types of wounds. This algorithm was chosen for its high speed and accuracy in classification and detection tasks. The research results showed that the learning model's performance, measured in accuracy, precision, recall, and average F1 score, reached 92%. Usability testing showed that the system performed well and could be helpful with minor improvements.Luka merupakan hasil kekerasan fisik yang merusak kontinuitas jaringan tubuh. Luka menjadi kasus yang sering terjadi dalam Ilmu Kedokteran Forensik dan Medikolegal. Dalam Ilmu Kedokteran Forensik dan Medikolegal, luka memiliki peranan penting dalam pembuatan Visum et Repertum (VeR) baik bagi jenazah maupun korban hidup. Sementara itu, dari hasil penelitian memperlihatkan kualitas hasil deskripsi forensik klinis VeR dibeberapa rumah sakit di Indonesia berada pada kategori sedang dan buruk, sedangkan kualitas hasil VeR yang baik akan sangat dibutuhkan untuk menentukan hukuman bagi pelaku di pengadilan, hasil VeR yang buruk akan menyebabkan sulitnya proses hukum. Pengaplikasian teknologi informasi dalam bidang kedokteran telah menghasilkan banyak tools yang dapat membantu ahli dalam menjalankan tugasnya. Begitu juga dalam bidang forensik klinis, kegiatan forensik patologis yang umumnya bersifat konservatif, dapat ditingkatkan kualitasnya dengan adanya teknik pengolahan citra dan machine learning. Bantuan teknologi digital processing untuk kasus forensik sudah ada sebelumnya seperti pada fotografi forensik, namun pengaplikasiannya masih minim dan perlu dikembangkan lebih lanjut. Pada penelitian ini menggunakan algoritma pembeljaran berbasis Yolo V4, algoritma ini dipilih karena memiliki kecepatan dan keakuratan yang tinggi dalam melakukan klasifikasi dan deteksi. Hasil penelitian menujukkan peforma model pembelajaran yang diukur berdasarkan akurasi, presisi, recall dan F1 Score rata-rata mencapai 92%. Kemudian hasil Usability testing menunjukkan sistem dinilai memiliki kinerja yang baik dan dianggap memiliki kegunaan yang cukup baik dengan ruang untuk perbaikan minor
Korelasi Jumlah Kendaraan Terhadap Kualitas Udara, Suhu, Dan Kebisingan Di Kota Malang Dengan Pendekatan Berbasis Internet Of Things
No full textIn response to the issues of air pollution, temperature, and noise, this project attempts to create an air quality, temperature, and noise monitoring system using Internet of Things (IoT) technology. This system will comprise physical components and a web platform that delivers real-time environmental reports. Users can readily obtain information regarding air quality, temperature, and noise levels via this platform, which takes advantage of the internet's accessibility. This Internet of Things-based device monitors environmental quality in six high-traffic areas in Malang, Indonesia. The system uses various sensors to monitor air quality, temperature, humidity, dust levels, carbon monoxide (CO), carbon dioxide (CO2), and noise pollution in real-time. Data was collected during peak traffic hours, demonstrating the direct influence of car emissions on air quality. The findings show that some regions' CO and particulate matter levels surpass safe criteria, notably during peak traffic periods, but CO2, humidity, and noise levels are below acceptable norms. These findings highlight the necessity for urban air pollution reduction initiatives. Additional sensor calibration and communication modifications are recommended to increase system accuracy and dependability. This study gives significant insights for local authorities to manage urban environmental quality and safeguard human health.Menanggapi masalah polusi udara, suhu, dan kebisingan, proyek ini berupaya menciptakan sistem pemantauan kualitas udara, suhu, dan kebisingan dengan menggunakan teknologi Internet of Things (IoT). Sistem ini akan terdiri dari komponen fisik dan platform web yang memberikan laporan lingkungan secara real-time. Pengguna akan dapat dengan mudah mendapatkan informasi mengenai kualitas udara, suhu, dan tingkat kebisingan melalui platform ini, yang memanfaatkan akses internet. Perangkat berbasis Internet of Things ini memantau kualitas lingkungan di enam area dengan lalu lintas tinggi di Malang, Indonesia. Sistem ini menggunakan berbagai sensor untuk memantau kualitas udara, suhu, kelembapan, tingkat debu, karbon monoksida (CO), karbon dioksida (CO2), dan polusi suara secara real time. Data dikumpulkan selama jam-jam sibuk lalu lintas, yang menunjukkan pengaruh langsung dari emisi mobil terhadap kualitas udara. Temuan menunjukkan bahwa tingkat CO dan materi partikulat di beberapa wilayah melampaui kriteria aman, terutama selama periode lalu lintas puncak, tetapi tingkat CO2, kelembaban, dan kebisingan berada di bawah norma yang dapat diterima. Temuan ini menyoroti perlunya inisiatif pengurangan polusi udara perkotaan. Kalibrasi sensor tambahan dan modifikasi komunikasi direkomendasikan untuk meningkatkan akurasi dan ketergantungan sistem. Studi ini memberikan wawasan yang signifikan bagi pemerintah daerah untuk mengelola kualitas lingkungan perkotaan dan menjaga kesehatan manusi
APD-BayTM: Prediksi Indeks Kualitas Udara Jakarta Menggunakan Bayesian Optimized LSTM
No full textThe Air Quality Index (AQI) is a metric for evaluating air quality in a region. Jakarta holds the fifth position globally in terms of air pollution. Several studies have been performed to forecast pollution levels in Jakarta. However, existing studies exhibit limitations such as outdated datasets, lack of data normalization, absence of machine learning parameter setting, neglect of k-fold cross-validation, and a failure to incorporate deep learning algorithms for pollution detection. This study introduces an air quality detection system called APD-BayTM to address these issues. This proposed system leverages Long Short-Term Memory (LSTM) and uses Bayesian Optimization (BO) to enhance the performance of air pollution detection. The methodology used in this research involves four key steps: data preprocessing, LSTM model development, hyperparameter tuning through BO, and performance assessment using 5-fold cross-validation. APD-BayTM exhibits robust performance that is comparable to previous research outcomes. The LSTM model in APD-BayTM on the training dataset achieved average precision, recall, F1 score, and accuracy values of 93.29%, 91.41%, 91.89%, and 95.90%, respectively. These metrics improved on the test dataset, reaching 97.44%, 99.71%, 98.52%, and 99.34%, respectively. These findings show the robustness of APD-BayTM across datasets of varying sizes, encompassing both large and small datasets.Indeks Kualitas Udara (IKU) berfungsi sebagai metrik untuk mengevaluasi kualitas udara, mengukur konsentrasi kumulatif berbagai polutan udara di suatu wilayah tertentu. Jakarta, ibu kota Indonesia, menempati peringkat kelima secara global dalam hal polusi udara. Beberapa penelitian telah dilakukan untuk memprediksi tingkat polusi di Jakarta menggunakan teknik pembelajaran mesin. Meskipun penelitian yang ada telah menunjukkan akurasi yang cukup dalam memprediksi IKU, mereka mengalami berbagai keterbatasan. Ini termasuk penggunaan dataset yang sudah ketinggalan zaman, ketidakhadiran normalisasi data, tidak ada penyesuaian parameter pembelajaran mesin, kurangnya implementasi validasi silang k-fold, dan kegagalan dalam menggunakan algoritma pembelajaran mendalam untuk deteksi polusi. Oleh karena itu, sistem deteksi polusi udara yang dikembangkan dalam penelitian sebelumnya cenderung suboptimal, menghasilkan akurasi yang lebih rendah.
Untuk mengatasi masalah ini, penelitian ini memperkenalkan sistem deteksi kualitas udara, disebut sebagai APD-BayTM. Sistem yang diusulkan ini memanfaatkan arsitektur Long Short-Term Memory (LSTM) dan menggabungkan Bayesian Optimization (BO) untuk meningkatkan kinerja deteksi polusi udara. Penelitian ini juga melakukan analisis perbandingan APD-BayTM terhadap sistem deteksi polusi udara sebelumnya. Metodologi yang digunakan dalam penelitian ini melibatkan empat langkah kunci: pra-pemrosesan data, pengembangan model LSTM, penyetelan hiperparameter melalui BO, dan penilaian kinerja menggunakan validasi silang 5-fold. Melalui eksperimen yang ketat, APD-BayTM menunjukkan kinerja yang kokoh, sebanding atau bahkan lebih unggul dibandingkan dengan hasil penelitian sebelumnya. Pada dataset pelatihan, model LSTM dalam APD-BayTM mencapai nilai rata-rata presisi, recall, skor F1, dan akurasi masing-masing sebesar 93,29%, 91,41%, 91,89%, dan 95,90%. Terutama, metrik-metrik ini lebih meningkat pada dataset uji, mencapai 97,44%, 99,71%, 98,52%, dan 99,34%. Temuan ini menunjukkan ketangguhan APD-BayTM di berbagai ukuran dataset, mencakup dataset besar maupun kecil
MQTT Broker Optimization: Comparative Analysis of Round Robin and Least Response Time
Full text linkOptimizing MQTT broker performance is crucial for maintaining efficient message routing in IoT systems, especially under varying workloads and QoS levels. This study compares the Round Robin (RR) and Least Response Time (LRT) algorithms to evaluate their performance across QoS levels 0, 1, and 2 and client loads ranging from 500 to 2,500 clients. Using Apache JMeter, key metrics such as CPU usage, throughput, delay, jitter, and response time were assessed. LRT was found to excel in enhancing response time and reducing delay, particularly under high client loads and in applications requiring minimal latency. However, this comes at the cost of higher CPU usage under heavy loads. In contrast, RR demonstrated optimal performance in maintaining balanced CPU utilization and predictable performance, though with slightly higher response times. Both algorithms demonstrated linear scalability in throughput, confirming their ability to handle increasing workloads without bottlenecks. These findings offer practical guidance for IoT developers: in latency-sensitive environments such as industrial automation, LRT is preferable due to its low-latency benefits, while RR is better suited for resource-constrained IoT systems like environmental monitoring, where stability and even load distribution are prioritized. The trade-offs identified provide valuable insights for selecting appropriate algorithms based on specific application requirements
Strategi pengoptimalan QCI untuk meminimalkan latensi
No full textLimited QCIs (QoS Class Identifiers) restrict the handling different service types with varying quality requirements. This necessitates research on QoS management to minimize latency and improve user experience, particularly for real-time applications like video conferencing and online gaming. This paper proposes a combined optimization scheme targeting QCI 3 to reduce latency. The approach involves disabling DRX, optimizing pre-allocation, and reducing the PDCP discard timer. The optimization performance is studied by taking the case of an e-sport game that demands low network latency, affecting the quality of the players' experience. The optimization scheme was validated through functionality, resource allocation, and air interface latency tests conducted under actual e-sport gaming conditions. Network latency was measured every minute to evaluate the impact of optimization on esports games running under QCI 7, QCI 3, and optimized QCI 3. In addition, air interface latency for optimized QCI 3 under networks with poor coverage and very high-capacity networks was compared to latency under QCI 8 (basic), QCI 7, and regular QCI 3. The optimization strategy demonstrated a significant reduction in air interface latency, up to 19% improvement compared to non-optimized QCI 3. It has reduced air interface latency's maximum, minimum, and standard deviation values during gameplay. The strategy also ensured concurrent operation with multiple QCI values without compromising other application’s throughput. The proposed optimization strategy effectively enhances the user experience by significantly reducing average latency and jitter.Jumlah QCI yang terbatas menyebabkan keterbatasan dalam menangani berbagai jenis layanan dengan syarat kualitas yang berbeda. Studi terkait skema, algoritma, dan optimasi telah diusulkan untuk memenuhi persyaratan QoS pada nilai QCI yang berbeda. Dalam makalah ini, skema pengoptimalan gabungan ditujukan untuk meningkatkan pengalaman pengguna. Kinerja pengoptimalan dikaji dengan menggunakan kasus permainan e-sport yang menuntut latensi jaringan yang rendah, sebab hal ini mempengaruhi kualitas pengalaman pemain. Strategi yang diusulkan divalidasi melalui uji fungsi dan kinerja pada QCI 3 yang dilakukan pada kondisi permainan e-sport yang sebenarnya. Hasil pengujian menunjukkan bahwa pengoptimalan pada QCI yang dipilih dapat beroperasi secara bersamaan dengan beberapa nilai QCI tanpa memengaruhi throughput aplikasi lain. Hal ini telah mengurangi nilai maksimum, minimum, dan deviasi standar dari latensi antarmuka udara selama bermain game. Optimasi QCI 3 pada cakupan yang buruk dan kapasitas sangat tinggi dapat mengurangi latensi rata-rata hingga 19% dibandingkan dengan QCI 3 tanpa pengoptimalan. Strategi pengoptimalan secara signifikan meningkatkan pengalaman pengguna dengan mengurangi rata-rata latensi dan jitter
Portable Stress Detection System for Autistic Children Using Fuzzy Logic
No full textStress is prone to occur in children with autism. According to the study, around 85% of children who have autism suffer from anxiety disorders that can exacerbate their condition, leading to self-harm and harm to those in their vicinity. Heart rate, skin conductance, and finger temperature changes occur during stress. In this paper, we design a system to monitor heart rate, body temperature, and skin conductance to detect signs of stress. Subsequently, the measurement data is processed using the fuzzy logic (FL) method as a decision-maker algorithm. In particular, we use 64 fuzzy rules with membership functions for each parameter. Parameter measurement results will be displayed using a widget called Gauge, while stress conditions will be displayed using a label widget. The results will be displayed on the Blynk application with an IoT system and viewed remotely via Android devices. The test was conducted on five children aged 5-9 years with varying body conditions. From the test results, the mean accuracy of the heart rate sensor was 95.01%, the mean temperature sensor accuracy was 97.7%, and the mean conductance sensor accuracy was 93.75%. The stress levels range from a minimum of 25% to a maximum of 75%. These findings indicate that the developed tool has performed effectively, and it is feasible to monitor its operation remotely
Object Segmentation in Stunted Face Images using Deeplabv3+ with Resnet-50
Full text linkStunting is the impaired growth and development that children experience from poor nutrition, repeated infection, and inadequate psychosocial stimulation. This study explores the impact of data preprocessing, specifically using DeepLabV3+ segmentation, on the performance of ResNet-50 in classifying stunting and non-stunting facial images. Initially, ResNet-50 achieved 99% accuracy and a 3.22% loss with the unsegmented dataset. By applying DeepLabV3+ to remove irrelevant features and backgrounds, the model's performance improved to a perfect 100% accuracy and a reduced loss of 0.45%. These results underscore the importance of high-quality data preprocessing in enhancing model precision and reliability. The findings have significant implications for practical applications, particularly in medical imaging, where improved diagnostic accuracy can benefit patient outcomes. Further research is recommended to explore additional preprocessing methods and their effects on model performance across diverse domains. This study highlights the transformative potential of effective data preprocessing in optimizing deep learning models for more accurate and reliable machine learning solutions
Sebuah Identifikasi yang Ditingkatkan dari Penyakit Katup Jantung Dengan Selective Phonocardiogram Features Driven by Convolutional Neural Networks (SFD-CNN)
No full textValvular Heart Disease (VHD) is a significant cause of mortality worldwide. Although extensive research has been conducted to address this issue, practical implementation of existing VHD detection results in medicine still falls short of optimal performance. Recent investigations into machine learning for VHD detection have achieved commendable accuracy, sensitivity, and robustness. To address this limitation, our research proposes utilizing Selective Phonocardiogram Features Driven by Convolutional Neural Networks (SFD-CNN) to enhance VHD detection. Notably, SFD-CNN operates on phonocardiogram (PCG) signals, distinguishing itself from existing methods based on electrocardiogram (ECG) signals. We present two experimental scenarios to assess the performance of SFD-CNN: one under default parameter conditions and another with hyperparameter tuning. The experimental results demonstrate that SFD-CNN surpasses other existing models, achieving outstanding accuracy (96.80%), precision (93.25%), sensitivity (91.99%), specificity (98.00%), and F1-score (92.09%). The outstanding performance of SFD-CNN in VHD detection suggests that it holds great promise for practical use in various medical applications. Its potential lies in its ability to accurately identify and classify VHD, enabling early detection and timely intervention. SFD-CNN could significantly improve patient outcomes and reduce the burden on healthcare systems. With further development and refinement, SFD-CNN has the potential to revolutionize the field of VHD detection and become an indispensable tool for healthcare professionals.Penyakit Jantung Katup (Valvular Heart Disease/VHD) adalah penyebab kematian global yang signifikan. Penelitian ekstensif telah dilakukan untuk mengatasi masalah ini, termasuk investigasi terbaru dalam menggunakan pembelajaran mesin untuk deteksi VHD. Meskipun penelitian-penelitian ini telah mencapai tingkat akurasi, sensitivitas, dan ketahanan yang patut dipuji, implementasi praktis dari hasil deteksi VHD yang ada di dunia kedokteran masih belum mencapai kinerja yang optimal. Untuk mengatasi keterbatasan ini, penelitian kami mengusulkan pemanfaatan Fitur Fonokardiogram Selektif yang Didorong oleh Jaringan Syaraf Tiruan Konvolusional (SFD-CNN) untuk meningkatkan deteksi VHD. Khususnya, SFD-CNN beroperasi pada sinyal fonokardiogram (PCG), yang membedakannya dengan metode yang sudah ada berdasarkan sinyal elektrokardiogram (EKG). Dalam penelitian ini, kami menyajikan dua skenario eksperimental untuk menilai kinerja SFD-CNN: satu di bawah kondisi parameter default dan satu lagi dengan penyetelan hiperparameter. Hasil eksperimen menunjukkan bahwa SFD-CNN melampaui model lain yang ada, mencapai akurasi yang luar biasa (96,60%), presisi (93,25%), sensitivitas (91,99%), spesifisitas (98,00%), dan F1-skor (92,09%). Hal ini menunjukkan bahwa SFD-CNN menjanjikan sebagai pendekatan yang unggul untuk deteksi VHD, menyoroti potensinya untuk implementasi praktis dalam aplikasi medis
Robot Tanggap Bencana Berbasis IoT untuk Identifikasi Korban pada Runtuhnya Bangunan
No full textNatural disasters like earthquakes frequently cause building collapses, trapping many victims under dense rubble. The first 72 hours are crucial for locating survivors, but the dangers of secondary collapse hinder direct access. Teleoperated robots can provide vital visual data to aid rescue efforts, though many prototypes remain constrained by high complexity, cost, and minimal customizability. This work investigates developing an Internet of Things (IoT) integrated disaster response robot that delivers accessible and remotely controllable capabilities for victim identification in hazardous collapse sites. Requirements analysis was conducted through a literature review and first responder interviews to determine the critical capabilities needed. The robot was designed using 3D modeling software and assembled using 3D printed and off-the-shelf components. It features remote-controllable movement, real-time video feed, geopositioning, and remote lighting toggling. Rigorous lab tests validated core functionalities, including camera image acquisition, Bluetooth communication ranges up to 10 meters, and comparable GPS coordinate accuracy to a smartphone. Further field experiments showcased the robot's ability to transmit smooth video signals over distances up to 12 meters and its adeptness at navigating complex terrains, evidenced by its proficient left/right panning and ability to surmount obstacles. An affordable Internet-of-Things integrated disaster robot tailored to victim identification was successfully designed, prototyped, and tested. This robot aids search and rescue operations by delivering visual and spatial data about hard-to-reach victims during the critical hours after disaster strikes. This confirms strong potential, accessibility, and customizability for professional and volunteer urban search and rescue teams across environments and economic constraints