Open Journal System Institut Teknologi PLN d/h. Sekolah Tinggi Teknik-PLN
Not a member yet
1182 research outputs found
Sort by
Evaluasi Kinerja Pembangkit Listrik tenaga uap ultra-Supercritical 1000 MW menggunakan GADS dan OEE untuk meningkatkan Efisiensi
No full textABSTRACT
With the advancement of time, the demand for electrical energy has become a primary necessity that is inseparable from Indonesian society. However, electricity cannot be produced instantly; rather, it requires a meticulous production process. The Cilacap Steam Power Plant (PLTU) implements the BERKAH guidelines (Happy, Efficient, Reliable, Competent, Adaptive, and Harmonious), with one of its important aspects being adaptability to changing times. Cilacap PLTU continues to upgrade its efficiency from the 2x300MW (Sub-Critical) power plant established in 2006, to the 1x660MW (Supercritical) in 2013, and the 1x1000MW (Ultra Supercritical) in 2019. Nevertheless, the newest unit, the 1x1000MW PLTU, still experiences disruptions such as Force Outage (FO) and Force Derating (FD), resulting in decreased efficiency. Over the past three years, there has been no trend indicating whether the plant's condition falls within the efficient category or not. It is hoped that the EFOR disruptions can remain below 7% and the EAF above 80%, in accordance with the contract with PLN.
This study aims to evaluate the plant’s performance using Generating Availability Data System (GADS)[2] data that complies with the standards of the North America Electric Reliability Council (NERC) and analyze disruptions using the Overall Equipment Effectiveness (OEE) method [3]. This research measures various parameters such as Equivalent Force Derating Hour (EFDH), Equivalent Plant Derating Hour (EPDH), Equivalent Availability Factor (EAF), and Force Outage Factor (EFOR). The results of this study are expected to provide a reference for improving plant performance and reducing the number of unit failures over the last three years. This research will also produce graphs of GADS and OEE values as a baseline for planning in the upcoming year, which can be valuable information and input for the company
Analisis Algoritma Prediksi Sistem Tenaga Listrik Menggunakan Machine Learning dan Implikasi Aksiologis
No full textDeteksi gangguan listrik merupakan aspek penting dalam memastikan stabilitas dan keamanan sistem tenaga listrik. Penelitian ini bertujuan untuk mengklasifikasikan jenis gangguan listrik menggunakan algoritma data mining dengan pendekatan machine learning. Data dikumpulkan melalui simulasi rangkaian listrik dalam kondisi normal dan gangguan, kemudian dianalisis menggunakan algoritma Decision Tree, Random Forest, Naive Bayes, dan K-Nearest Neighbors. Hasil pengujian menunjukkan bahwa algoritma Decision Tree dan Random Forest mencapai tingkat akurasi dan performa 100%, sedangkan Naive Bayes juga menunjukkan performa optimal dengan akurasi 100%. Sebaliknya, K-Nearest Neighbors menunjukkan performa yang lebih rendah dengan akurasi 82,20%. Temuan ini memperlihatkan bahwa algoritma Decision Tree dan Random Forest sangat efektif dalam mendeteksi dan mengklasifikasi tipe gangguan listrik secara akurat. Hasil penelitian ini memberikan kontribusi penting dalam pengembangan sistem proteksi otomatis untuk sistem tenaga listrik yang lebih andal dan adapti
Analisis Percepatan Waktu Pekerjaan Rigid Pavement Menggunakan Critical Path Method (CPM): (Studi Kasus: Proyek Pembangunan Jalan Kawasan Heartful Town Depok)
No full textRoad construction projects often encounter delays, particularly in rigid pavement development, where precise scheduling and execution are essential. The Heartful Town Road Construction Project in Depok experienced delays due to limited control over critical activities and the absence of scheduling analysis using the Critical Path Method (CPM). This study aims to evaluate project acceleration through CPM and crash duration analysis. Data from the project management team were used to build a work network and determine the critical path. Two acceleration strategies were examined: extending working hours and increasing workforce allocation. Results showed the initial project duration of 130 days was reduced to 127 days with one hour of daily overtime, and to 114 days with three hours of overtime. Alternatively, increasing the workforce by 50% on critical activities shortened the subbase layer from 33 to 18 days, compaction from 21 to 14 days, lean concrete casting from 28 to 15 days, and rigid pavement casting from 42 to 27 days. These reductions produced a total critical path duration of 80 days. The findings highlight that combining CPM with crash duration methods, either through overtime or additional workers, effectively accelerates project completion without changing the activity sequence
Penerapan Model Machine Learning Untuk Prediksi Kekuatan Beton
No full textDeveloping a concrete strength prediction model using three machine learning approaches, namely Linear Regression, Neural Network, and Multi-Layer Perceptron (MLP). The three models were tested using a dataset containing information about the composition of concrete materials and concrete strength test results. Evaluation is carried out using Precision, Recall, F1-Score, Mean Squared Error (MSE), and R-squared metrics to measure the accuracy of model predictions. The research results show that the MLP model provides the best performance, with very high Precision, Recall and F1-Score, as well as low MSE and R-squared reaching 0.97. Compared with Neural Network and Linear Regression, the MLP model shows better generalization ability on test data. Although other models also gave good results, MLP proved to be more effective in predicting concrete strength with higher accuracy. This research indicates that the use of MLP in concrete strength prediction can increase accuracy and efficiency in construction applications
Pengembangan Aplikasi Manajemen Konten Digital Signage Berbasis CMS untuk Optimalisasi Komunikasi Internal Perusahaan
No full textEffective internal communication is a crucial factor in improving a company's operational efficiency. However, limited access to communication devices for some employees, particularly those working in the field, poses a challenge in ensuring fast and equitable information dissemination. This study aims to develop a Content Management System (CMS)-based digital signage that enables real-time, automated, and centralized information delivery within a corporate environment. The system is designed using the CodeIgniter framework and PostgreSQL as the database, integrating various information sources such as photos, videos, and employee attendance data. Key features developed include the Admin Dashboard for monitoring company activities, the Attendance Dashboard displaying employee attendance, visitors, and multimedia information, and communication-related features such as Communication, YouTube, Television, Photo, Guest, and Business Trip (Dinas) to support more effective information distribution. Functional testing and User Acceptance Testing (UAT) were conducted to ensure the system meets its specifications. The test results indicate that all features function properly without significant errors and that the system is user-friendly for HR staff. Compared to conventional methods that rely on external storage media such as USB drives or memory cards, this system provides a more flexible and efficient solution for managing and distributing digital content. With centralized content management, companies can update information quickly without requiring reconfiguration on each digital signage screen. The successful implementation of this system demonstrates that it can be an effective solution for improving internal communication in various corporate environments
Power Generation system pada Gas Turbine Generator SGT- 800 Pada PT Kilang Pertamina Balikpapan
No full textThis research aims to evaluate the performance and efficiency of the power generation system in meeting the refinery's energy needs. The research method used involves in-depth analysis of operational and technical data from gas turbine generators, as well as factors that influence performance and efficiency. The research results show that this power generation system is able to provide stable and efficient power output in accordance with refinery demand. Operational parameters such as temperature, pressure and gas flow stability have an important role in maintaining optimal performance. Apart from that, regular maintenance and routine monitoring of critical components are also needed to ensure the smooth operation of the power generation system. These findings provide valuable insights for the energy industry regarding the operation and maintenance of gas turbine generator-based power generation systems. Optimizing operational processes and proper maintenance can increase the efficiency and productivity of power generation systems, thereby supporting smooth refinery operations and minimizing potential disruptions to energy supply. In conclusion, this research highlights the importance of effective management of power generation systems to achieve the desired sustainability and energy efficiency goals
A Literature Review : Komparasi Algoritma K-Means dan DBSCAN dalam Klustering untuk Strategi Promosi
No full textDalam dunia pendidikan tinggi, strategi promosi yang efektif menjadi kunci dalam menarik calon mahasiswa. Salah satu cara yang dapat dilakukan adalah dengan memanfaatkan teknik analisis data, seperti klustering, untuk memahami pola dan segmentasi calon mahasiswa. Penelitian ini melakukan kajian literatur terhadap algoritma K-Means dan DBSCAN guna mengevaluasi kelebihan dan kekurangan masing-masing dalam proses segmentasi. Hasil studi menunjukkan bahwa K-Means lebih cepat dan efisien dalam mengelompokkan data yang memiliki distribusi rapi, sedangkan DBSCAN lebih unggul dalam menangani data dengan kepadatan yang tidak merata serta mampu mengidentifikasi data outlier. Beberapa penelitian yang ditinjau juga menunjukkan bahwa kombinasi kedua algoritma ini dapat meningkatkan akurasi segmentasi, di mana K-Means digunakan sebagai langkah awal untuk membentuk klaster awal, sementara DBSCAN digunakan untuk validasi dan penyempurnaan hasil klasterisasi. Kesimpulannya, pemilihan algoritma harus disesuaikan dengan karakteristik data yang dianalisis. Perguruan tinggi dapat memanfaatkan kombinasi kedua metode ini untuk mengembangkan strategi promosi yang lebih efektif dan terarah. Penelitian selanjutnya dapat mengeksplorasi metode lain, seperti machine learning, untuk lebih meningkatkan akurasi segmentasi dan efektivitas strategi pemasaran pendidikan tinggi
Analisis Tingkat Kemacaetan Pada Ruas Jalan Margonda Raya Kota Depok
No full textMargonda Raya Street is one of the traffic infrastructures that affect road users in Depok City and neighboring towns, especially road users working in Jakarta. The occurrence of congestion is due to the imbalance in the traffic network, namely the accumulation of vehicles leading to high traffic density on a certain road network, causing traffic flow to be blocked or even stopped. At the time of congestion, the saturation value on the road is greater than 0.75. This method is used to evaluate the Indonesian Road Management Manual (MKJI 1997) by observing the geometric conditions of the 6/2D main access road classification in the field, the traffic volume is monitored for 2 weeks on Monday, Wednesday, Friday and Sunday during peak hours. The capacity of Margonda Raya Street is 8351.64 pcu/h. EnglishThe traffic volume on Margonda Raya Road in the Jakarta-Depok direction was 2736.20 pcu/h and in the Depok-Jakarta direction was 3049.30 pcu/h. The best service level on Margonda Raya Road was on Sunday, May 26 (morning, Jakarta-Diepok) at 0.29 service level B, with stable traffic conditions. However, the trial operation was limited by traffic conditions. Meanwhile, the worst service level was on Friday, May 17 (afternoon, Jakarta-Diepok) at 1.12 service level F, with forced traffic conditions, low speed, excess traffic capacity, and long queues (traffic jams)
Uji Mutasi pada Penerapan Token Mitigasi Kerentanan Cross Site Request Forgery
No full textKeamanan aplikasi web merupakan perhatian kritis karena banyaknya kerentanan seperti SQL Injection (SQLi), Cross-Site Scripting (XSS), dan Cross-Site Request Forgery (CSRF). Kerentanan-kerentanan ini dieksploitasi oleh penyerang untuk mendapatkan akses yang tidak sah dan merusak aplikasi web. Penelitian kami berfokus pada analisis kerentanan Cross-Site Request Forgery (CSRF) dengan menggunakan pendekatan pengujian mutasi, yang menerapkan 5 operator mutasi yang memutasi token input pada form input. Kami memperkenalkan alat otomatis untuk mengidentifikasi dan mengatasi kerentanan CSRF menggunakan pola token rahasia. Alat ini meningkatkan keamanan aplikasi web berbasis PHP tanpa mengorbankan fungsionalitasnya. Ketika kerentanan berhasil dideteksi, aplikasi akan memberi tahu pengguna agar segera dapat diperbaiki, indikator skor mutasi kami gunakan sebagai alat pengukuran sejauh mana pengujian mutasi berhasil dilakukan, hasilnya dari 1022 mutasi yang dihasilkan seluruhnya dapat dihentikan dengan presentasi 100% menunjukan bahwa operator mutasi yang digunakan dapat bekerja dengan baik untuk mendeteksi mutan yang dihasilkan
Penerapan YOLOv8 Dalam Deteksi Penyakit Tanaman Daun Jambu Air Secara Real-time
No full textPendeteksian penyakit pada tanaman adalah tantangan utama dalam pertanian untuk menjaga kesehatan dan produktivitas tanaman. Penelitian ini mengimplementasikan metode YOLOv8 untuk mendeteksi penyakit pada daun jambu air dan mengevaluasi akurasi menggunakan mean Average Precision (mAP). Dataset yang terdiri dari 754 gambar daun jambu air diperluas menjadi 1229 gambar melalui proses augmentasi. Preprocessing dataset dilakukan di Roboflow, sementara pelatihan model dilakukan di google collab. Model YOLOv8 dilatih untuk mendeteksi enam jenis penyakit—embun jelaga, berlubang, antraknosa, mosaic virus, layu fusarium, dan gall—serta satu kategori daun sehat. Pengujian dilakukan terhadap 75 daun untuk memperoleh hasil mAP, yang menunjukkan nilai sebesar 82%. Sistem deteksi dibangun menggunakan framework Flask untuk integrasi dengan antarmuka pengguna berbasis web. Hasil penelitian menunjukkan bahwa YOLOv8 efektif dalam mendeteksi penyakit pada daun jambu air dengan akurasi tinggi, memberikan kontribusi signifikan dalam pengelolaan kesehatan tanaman.
Kata kunci: deteksi penyakit tanaman, YOLOv8, mean average precision, daun jambu air, framework flas