E-Journal Politeknik Negeri Cilacap
Not a member yet
919 research outputs found
Sort by
Comparison Method of Convolutional Neural Network and Support Vector Machine for Facial Expression Recognition
Get PDFFacial expressions are an important component of nonverbal communication that enable humans to understand each other\u27s emotional states intuitively. Although humans can easily recognize expressions such as smiles or frowns, replicating this ability in computational systems remains a complex challenge. Therefore, an automated system capable of accurately and efficiently identifying facial expressions is needed. This research aims to compare the accuracy of CNN and SVM methods in facial expression recognition using the JAFFE dataset, which is limited to one demographic (Japanese women) with 284 images (80% for training, 10% for validation, and 10% for testing). CNN extracts features through convolution and pooling processes, while SVM is used as a classification algorithm based on statistical learning. The recognition process is divided into three main stages: data preprocessing, feature extraction, and facial expression classification. The system recognizes seven emotional categories: anger, disgust, fear, happiness, neutral, sadness, and surprise. Results show that CNN outperforms SVM with an accuracy of 86%, while SVM achieves 81%. The limitations of the dataset may affect generalizability, and further research can use larger, more diverse dataset
A Comparative Analysis of KIP-K Acceptance Prediction Based on School Type Using XGBoost, Random Forest, and SVM-RBF: Evaluation Through Accuracy and Data Visualization
Get PDFThe Indonesia Smart College Card (Kartu Indonesia Pintar-Kuliah / KIP-K) is a national initiative aimed at expanding access to higher education for students from socioeconomically disadvantaged backgrounds. This study, conducted at Politeknik Negeri Cilacap, investigates the prediction of KIP-K acceptance based on the type of high school attended by applicants. A comparative analysis was carried out using three supervised machine learning algorithms: Extreme Gradient Boosting (XGBoost), Random Forest, and Support Vector Machine with Radial Basis Function (SVM-RBF). The dataset, sourced from institutional admission records between 2022 and 2024, comprises information on school types (public, private, vocational, madrasah, and others), demographic attributes, and the KIP-K acceptance status. The data were split into training and testing sets using a 50:50 stratified sampling technique to preserve class distribution. Model performance was evaluated using standard classification metrics, including accuracy, precision, recall, and F1-score. Additionally, confusion matrices, ROC curves, and feature importance visualizations were used to enhance model interpretability. The experimental results demonstrate that the XGBoost algorithm consistently outperformed the other models across all performance metrics. Specifically, XGBoost exhibited the highest discriminatory power with an AUC of 0.93, followed by Random Forest (0.90) and SVM-RBF (0.85). These findings affirm the suitability of tree-based ensemble methods for classification tasks in educational domains and emphasize the predictive relevance of school type in determining KIP-K eligibility. The study presents a data-driven decision support framework that can contribute to more objective, transparent, and equitable scholarship allocation practices, particularly within the context of vocational higher education institutions in Indonesi
Development of a Hybrid CNN–SVM-Based Acute Lymphoblastic Leukemia Detection System on Hematology Image Data
Get PDFAcute Lymphoblastic Leukemia (ALL) is among the most common pediatric blood cancers and progresses rapidly, necessitating early and accurate detection. Manual diagnosis via microscopic analysis of blood samples is time-consuming and highly dependent on specialist expertise. This study proposes a hybrid model that combines a Convolutional Neural Network (CNN) with a Support Vector Machine (SVM) to automatically detect ALL from blood-cell images. The CNN performs deep feature extraction from images, while the SVM serves as the classifier to determine ALL status. The dataset comprises microscopic images labeled as ALL or normal and is processed through preprocessing steps such as augmentation and normalization. The adopted CNN produces optimized feature representations. Experimental results show that the hybrid CNN–SVM model with an RBF kernel achieves the best performance, with an accuracy of 96.4%, precision of 95.8%, recall of 96.1%, and an F1-score of 96.0%, surpassing pure CNN-based baselines. Training converged at the 41st epoch, with a training accuracy of 97.2%, validation accuracy of 95.9%, training loss of 0.09, and validation loss of 0.11, indicating stable learning without overfitting. The model’s ROC curve lies well above the chance diagonal, with an Area Under the Curve (AUC) of 0.914, means there is a 91.4% chance the model assigns a higher score to a truly positive (leukemia) image than to a negative (normal) image.These findings suggest that the CNN–SVM hybrid approach enhances leukemia detection performance compared with conventional CNN-only methods and holds promise as a fast, accurate, and efficient image-based decision-support tool for early leukemia diagnosis in digital hematology
Pemanfaatan Energi Panas Pada Tungku Biomassa Guna Pengisian Aki 12V Pada Generator Termoelektrik
Get PDFTermoelektrik merupakan suatu alat yang berbentuk modul atau peltier, yang dapat secara langsung mengubah energi panas menjadi energi listrik. Termoelektrik dapat dibangkitkan dengan memanfaatkan proses pembakaran biomassa pada tungku. Energi panas dari hasil pembakaran biomassa digunakan sebagai energi primer untuk membangkitkan generator termoelektrik. Pada penelitian ini, dilakukan pembuatan biomassa dari limbah kulit pisang dan daun kering dengan enam jenis briket. Komp osisi jenis briket terbagi menjadi, jenis A (dengan perbandingan 20% perekat: 60% kulit pisang: 40% daun kering), jenis B (dengan perbandingan 20% perekat: 50% kulit pisang: 50% daun kering), jenis C (dengan perbandingan 30% perekat: 30% kulit pisang: 70% daun kering), jenis D (briket arang batok kelapa), jenis E (briket arang bambu), dan jenis F (briket arang kayu). Pembakaran briket maksimal adalah briket jenis D (briket batok kelapa) dengan tegangan 19,88 V dan arus 0,19 A pada perbendaan suhu (ΔT) 80oC. Pembakaran briket terendah adalah briket jenis A dengan tegangan 12,55V dan arus 0,12A pada perbedaan suhu (ΔT) 30oC. Sedangkan daya terbesar untuk modul termoelektrik adalah sebesar 3,76 Watt/Jam dengan spesifikasi aki 12V, 7.2Ah. Sehingga untuk melakukan pengisian aki dibutuhkan waktu 22,9 jam
Design and Implementation of a Web-Based Relay Maintenance Information System at the Substation
Get PDFProtection relays detect disturbances or abnormal conditions in the electric power system, isolate disturbances, eliminate abnormal conditions, and produce signals or indications. One of the protection relays used in substations is the Over-Current Relay (OCR) and Ground Fault Relay (GFR). Routine maintenance of OCR and GFR is usually done by resetting or replacing the relay. However, the maintenance process was carried out manually using special test equipment, with the results recorded manually in a form, which was then digitized. This procedure is considered less effective. This study develops a web-based relay maintenance application using the waterfall method. This method was chosen because of its structured approach, complete documentation, and cost and processing time stability. The test results show that the web-based system functions according to the expected functionality. This application has been proven to facilitate and increase officers\u27 effectiveness in relay maintenance based on a user survey. This application\u27s OCR and GFR setting results are simulated using the ETAP 19.01 application for a short circuit fault scenario. The simulation results show that the relay works according to the inverse standard characteristics. The relay trip time for a fault on the 150kV side of the transformer is 2.2 seconds, while for a fault on the 20kV side, it is 7 seconds on Circuit Breaker 2 and 26 seconds on Circuit Breaker 1.Relai proteksi berfungsi untuk mendeteksi gangguan atau kondisi ketidaknormalan pada sistem tenaga listrik, dengan tujuan mengisolasi gangguan, menghilangkan kondisi tidak normal, serta menghasilkan sinyal atau indikasi. Salah satu relai proteksi yang digunakan di gardu induk adalah Over-Current Relay (OCR) dan Ground Fault Relay (GFR). Pemeliharaan rutin OCR dan GFR biasanya dilakukan melalui pengaturan ulang (setting) atau penggantian relai. Namun, proses pemeliharaan selama ini dilakukan secara manual menggunakan alat uji khusus, dengan hasil yang dicatat secara manual ke dalam formulir, yang kemudian didigitalisasi. Prosedur ini dinilai kurang efektif. Penelitian ini mengembangkan aplikasi pemeliharaan relai berbasis web menggunakan metode waterfall. Metode ini dipilih karena pendekatannya yang terstruktur, dokumentasi yang lengkap, serta kestabilan dalam biaya dan waktu pengerjaan. Hasil pengujian menunjukkan bahwa sistem berbasis web yang dikembangkan berfungsi sesuai dengan fungsionalitas yang diharapkan. Berdasarkan survei pengguna, aplikasi ini terbukti memudahkan dan meningkatkan efektivitas petugas dalam melakukan pemeliharaan relai. Nilai hasil setting OCR dan GFR dari aplikasi ini disimulasikan menggunakan aplikasi ETAP 19.01 untuk skenario gangguan hubung singkat. Hasil simulasi menunjukkan bahwa relai bekerja sesuai dengan karakteristik standar invers. Waktu trip relai untuk gangguan pada sisi 150kV transformator adalah 2,2 detik, sedangkan untuk gangguan pada sisi 20kV transformator adalah 7 detik pada Circuit Breaker 2 dan 26 detik pada Circuit Breaker
Korelasi Nilai Oktan dan Kandungan Oksigen Biofuel Mangrove dengan Brake Torque dan Emisi NOx Mesin Bensin 150cc
Get PDFAlternative fuels are a priority to reduce the problem of fuel stock shortages. Bioethanol has potential because of its octane number specifications, oxygen content, and simple production techniques. Mangrove bioethanol is produced from mangrove fruit, which is non-edible and abundant. This study aims to observe the correlation of Mangrove Biofuel specifications (mixture of gasoline and 5% Mangrove Bioethanol) to Brake Torque and NOx Emissions produced by gasoline engines. Experimental methods are applied, from Mangrove Bioethanol production, octane value, oxygen content testing, and Mangrove Biofuel production to performance and emission testing. The results of the study showed that the correlation of the 101 octane number specifications and 30.31% oxygen content of Mangrove Biofuel increased the highest Brake Torque, reaching 5.45%, and increased the highest NOx emissions up to 6.54% at engine speeds of 5000 rpm
Analisa Pengukuran Ketebalan Steel Block, Aluminium Block dan Steel Plate Menggunakan Ultrasonic Thickness Gauge
Get PDFThickness measurement using an ultrasonic sensor is a type of non-destructive test (NDT) that is commonly used in various industries. Measurement of thickness (thickness) on steel blocks, aluminum blocks and steel plates using an ultrasonic thickness gauge. The method used is the contact testing method which is then compared with manual measurements. Based on data analysis from the measurement results, it was found that the thickness with the highest percent error was in the Steel S3 sample, namely 93.87%. Meanwhile, the thickness measurement with the lowest percent error in the Steel plate sample was 0%. Meanwhile, for measuring the dimensions of artificial corrosion defects, the smallest error percentage obtained was 0% for several dimensions, with the average measurement error obtained being 22.82%. The profile shape of the detected artificial corrosion defects is exactly the same as the reference profile. So, measuring the thickness of steel blocks, aluminum blocks and steel plates as well as measuring artificial corrosion defects located in steel plates can be detected well and the measuring results can be trusted
Analisis Hasil Pengujian Tahanan Isolasi pada Pemutus Tenaga (PMT) Kubikel 20 kV Setia di Gardu Induk Budi Kemuliaan
Get PDFPermintaan listrik di Indonesia terus meningkat, terlihat dari penambahan daya pelanggan rumah tangga serta meningkatnya kebutuhan pasokan tegangan menengah bagi sektor industri. Oleh karena itu, pemeliharaan peralatan di Gardu Induk, seperti pengujian tahanan isolasi pada Pemutus Tenaga (PMT), sangat diperlukan untuk menjaga keandalan distribusi listrik dan mencegah terjadinya kebocoran arus, korsleting, serta risiko kejutan listrik. Penelitian ini menganalisis hasil pengujian tahanan isolasi PMT Kubikel 20 kV Setia di Gardu Induk Budi Kemuliaan pada tahun 2024 dan tahun 2025 guna mengevaluasi kinerja PMT tersebut. Hasil penelitian menunjukkan adanya penurunan nilai pada terminal atas-bawah pada tahun 2025, yang mengindikasikan degradasi isolasi pada bagian atas akibat faktor usia dari PMT tersebut. Meskipun seluruh nilai pengujian masih berada jauh di atas batas minimum yang diperbolehkan, disarankan untuk melakukan pergantian unit kubikel serta uji ulang dan pembersihan isolator PMT guna mempertahankan keandalan sistem isolasi. Hasil penelitian ini memberikan dasar praktis bagi unit pemeliharaan untuk menentukan kebutuhan penggantian PMT dan penjadwalan ulang inspeksi berkala guna menjaga keandalan sistem distribusi listrik
Investigation on Exhaust Emission and Performance of SI-Matic Engine Applied Acetone-Butanol-Ethanol (ABE) Fuel Mixtures
Get PDFAcetone-butanol-ethanol (ABE) is a preferred alternative energy for SI engines. This research uses an experimental method with an automatic SI engine, RON-90 fuel mixed with ABE1 (12:8:1)v/v, ABE2 (15:10:1)v/v and ABE3 (1i8:12:1) v/v. Engine speed is 4000-10000rpm, and compression ratio is 11.6:1. Emission and engine performance testing used EPSG4-Gas analyzer and Dynotest-chassis type 50L-BRT. This research aims to explore the mixture of RON-90 and ABE to optimize performance and exhaust emissions. This research shows that torque increases by an average of 14.7% at an engine speed of 6000rpm. Power increased significantly with an average value of 9.5% at an engine speed of 8000rpm, MEP increased by 0.5%, and thermal efficiency increased by 7%. SFC experienced a fairly optimal decrease of 15.6% on average. The exhaust gas emissions produced are CO and HC. The reduction in CO and HC occurred in the ABE3 variant with values of 8.2% and 1.6%, respectively
Perbandingan Kinerja Antara Gatling dan Apache JMeter pada Uji Beban RESTful API
Get PDFThis research explores and compares the performance of two popular load testing tools, namely Gatling and Apache JMeter, with a focus on API performance testing. The rapid growth in web and mobile application development highlights the urgent need to ensure optimal API performance. This research was conducted to provide in-depth insight into the advantages and disadvantages of these two testing tools through the use of similar testing scenarios. The experimental method involves implementing test scenarios that include load variations and high demands on both devices. The main parameters observed include API response time, throughput, and latency. In-depth analysis was carried out on the data obtained to evaluate the reliability and efficiency of each tool. The results of this research provide a comprehensive understanding of the performance of Gatling and Apache JMeter in the context of API performance testing. These findings can provide practical guidance for software developers and testing practitioners in selecting load testing tools that suit their project needs. Recommendations for future research include expanding exploration of other load testing tools, comparison with more complex test scenarios, and integration with performance monitoring tools for more holistic analysis. Thus, this research is expected to make a significant contribution to the understanding and selection of effective load testing tools in web and mobile application development.