Jurnal Infotel (Sekolah Tinggi Teknologi Telematika Telkom Purwokerto)
Not a member yet
392 research outputs found
Sort by
Prototype of cascade level and flow control system on steam drum based on IoT
No full textIn the industrial field, boiler functions to heat a fluid in the form of water, the boiler has a part which is a steam drum which has a function to produce steam for use for utility needs, and a steam turbine, in practice, the state of the water level must be maintained at the desired value or set. point so that carryover does not occur, and in overcoming these problems a control system is needed. This control works by comparing the value of the sensor and the set point, then gives an output signal to correct that to speed up the response, so it is necessary to use a cascade control configuration that adds an input flow control as a slave control. In this prototype, the cascade level control serves to control the level process. In addition, the human-machine interface has been designed to monitor processes in real-time. In addition, this prototype is equipped with an Internet of Things system that functions for the monitoring process as long as it is always connected to the internet. To run the control system, parameter control is needed, in this project the PID parameter setting uses the Ziegler-Nichols method with the parameter Kp level=20.25; Ki level = 1.51; Kp Flow = 5.14; Ki flow = 2.2
Classification of diabetic foot ulcer using convolutional neural network (CNN) in diabetic patients
Full text linkThe image of chronic wounds on human skin tissue has the similar look in shape, color and size to each other even though they are caused by different diseases. Diabetic ulcer is a condition where peripheral arterial blood vessels are disrupted due to hyperglycemia in people with diabetes mellitus. This research was aimed to analyze the accuracy of the Convolutional Neural Network algorithm in classifying diabetic ulcer disease with a transfer learning approach based on the appearance of the image of the wound on the sole in people with diabetes mellitus. By applying the transfer learning approach, the results showed that the Resnet152V2 model achieved the best accuracy value of 0.993 (99%), precision of 1.00, recall of 0.986, F1-Score of 0.993 and Support of 72. Therefore, the ResNet152V2 model was highly considered for classifying diabetic ulcer in patients with diabetes melitus.Citra luka kronis pada jaringan kulit manusia memiliki bentuk, warna dan besar luka yang terlihat menyerupai satu sama lain walau ditimbulkan oleh penyakit berbeda. Ulkus diabetik adalah kondisi dimana pembuluh darah arteri perifer terganggu disebabkan oleh hiperglikemia pada penderita Diabetes Melitus. Penelitian ini bertujuan untuk menganalisa akurasi yang dihasilkan algoritma Convolutional Neural Network dalam mengklasifikasi penyakit ulkus diabetik dengan pendekatan transfer learning berdasarkan penampakan citra luka pada telapak kaki pada penderita diabetes melitus. Dengan menerapkan pendekatan transfer learning, diperoleh hasil bahwa model Resnet152V2 meraih nilai akurasi terbaik yakni sebesar 0.993 (99%), precision 1.00, recall 0.986, F1-Score 0.993 dan Support sebesar 72. Oleh sebab itu, model ResNet152V2 sangat dipertimbangkan untuk mengklasifikasikan penyakit ulkus diabetik pada penderita diabetes melitus
Analysis and small-signal modeling of simplified cascade multiphase DC-DC buck converter
Full text linkOne of the power converters that are often implemented in renewable energy applications is a DC-DC power converter. One of such converters is a step-down converter or buck converter whose output voltage is lower than its input. A novel DC-DC buck converter for low output-voltage and high output current applications is presented in this paper. When compared to the conventional buck converter, the voltage ratio of the proposed topology is higher. The output of this converter also has lower ripple. Thus, the proposed topology is appropriate for renewable applications. The operating principle and small-signal model analysis are discussed in detailed. Finally, a simulation studies is carried out by PSIM to verify performances of the offered topolog
Performance of the K-Nearest Neighbors method on identification of maize plant nutrients
Full text linkMaize is one kind of commodity consumption in domestic as well as export that has high economic value. However, the low productivity is caused by the main factor, namely the decreased level of soil fertility, so that it has the same effect on crop yields. These problems require the application of technology with the K-Nearest Neighbor (KNN) method. The method of study is based on 17 signs of nutrient deficiencies with Minkowski distance calculation process, calculation of deficiency of soil nutrients based on the value of K determined. The test results of the research use K = 75 to get an accuracy of 92.40. Comparative analysis of the K-nearest neighbor (K-NN) and NB methods by looking for the closeness between the criteria for new cases and old case criteria based on the criteria for the closest cases. The results showed that the K-Nearest Neighbor (K-NN) Algorithm had a better accuracy value than NB.Jagung merupakan salah satu jenis komoditas konsumsi dalam negeri maupun ekspor yang memiliki nilai ekonomi tinggi. Namun rendahnya produktivitas tersebut disebabkan oleh factor utama yaitu tingkat kesuburan tanah yang menurun, sehingga berpengaruh sama hasil panen. Permasalahan tersebut memerlukan penerapan teknologi dengan metode K-Nearest Neighbor (KNN). Metode penelitian didasarkan pada 17 tanda kekurangan unsur hara dengan proses perhitungan jarak minkowski, perhitungan kekurangan unsur hara tanah berdasarkan nilai K yang ditentukan. Hasil pengujian penelitian menggunakan K = 75 untuk mendapatkan akurasi sebesar 92,40. Analisis komparatif metode K-nearest neighbor (K-NN) dan NB dengan mencari kedekatan antara kriteria kasus baru dan kriteria kasus lama berdasarkan kriteria kasus terdekat. Hasil penelitian menunjukkan bahwa Algoritma K-Nearest Neighbor (K-NN) memiliki nilai akurasi yang lebih baik daripada NB
Secure protection for covid-19 infographic using blockchain and discrete cosine transform-singular value decomposition (DCT-SVD) watermarking
Full text linkCovid-19 infographics have a crucial role in mitigating the covid-19 pandemic by conveying the complex Covid19 information in a form of a simple yet understandable image. However, keenly to contribute to mitigating Covid-19, numerous parties and agencies had released Covid-19 infographics that might contain incorrect or inaccurate information. To prevent such recurrent, this paper proposed an authentication system by using a blockchain-based authorization service that lets the authority guarantee the correctness and validity of the infographics in a transparent manner. We proposed smart contract-based watermarking requests and approval management that let anyone track the watermarking process. To prevent unauthorized infographic fabrications, we use the DCT-SVD method considering its robustness against various attacks. We deployed and evaluated the smart contract on Ethereum test networks (Ropsten, Rinkeby, Goerli, and Kovan) to compare the efficiency and the ease of use. The result showed that the test networks have similar efficiency while the Ropsten and Goerli have better ease of use. The watermark validation service is accessible via a web-based interface for anyone to check the validity of the infographic’s watermark
Peer to peer (P2P) and cloud computing on infrastructure as a service (IaaS) performance analysis
Full text linkThe resources of information technology and the availability of services on non-cloud network systems are limited. This constitutes problems for companies, especially in the efficient management of information technology. The high investment in infrastructure procurement is an obstacle in building centralized systems, including the adoption of cloud computing through Infrastructure as a Service (IaaS), as an elective solution. This research aims to analyze the performance of cloud servers on IaaS services using the parameters of cloud service availability, resource utilization, and throughput transfer which were implemented in companies engaged in the toll road concession sector. Furthermore, the results are expected to be a reference in supporting company decisions/policies related to cloud system adoption. The methodology involved the Network Development Life Cycle (NDLC), a system constituted by 6 (six) stages of management, namely user, proxy server, database, web service, monitoring service, and Remote Desktop Protocol (RDP). The results of cloud service availability indicate that the cloud system provides service availability (system interface, broad network access, and resource pooling). Furthermore, cloud systems have a significant performance on resource utilization (CPU) and throughput transfer parameters, while non-cloud systems only excel in response time and resource utilization (Memory) parameters. The overall result analysis based on this research scenario showed that the cloud system provides services according to user needs and has a better speed in data transmission, but has shortcomings in response time
Detection of learning styles with prior knowledge data using the SVM, K-NN and Naïve Bayes algorithms
Full text linkThe two types of automatic learning style detection approaches are data driven (DD) and literature based (LB). Both methods of automatic learning style detection have advantages over traditional learning style detection methods because they use external data sources, such as forums, quizzes and views of teaching materials, that are more accurate than the questionnaires used in traditional styles of detection. The results of automatic detection, on the other hand, do not always reflect learning styles. This paper presents a learning style recognition method that uses data from the learner’s internal source, namely prior knowledge, to overcome these challenges. Prior knowledge is proposed because it is based on the learner’s knowledge or skills, which better reflect the learner’s characteristics, rather than on the learner’s behaviour, which tends to be dynamic. By using past knowledge, this paper presents a method for detecting automatic learning patterns. The learning style detection framework is unique in that it consists of three stages: prior knowledge question development, prior knowledge measurement and learning style detection using the Support Vector Machine (SVM), Naïve Bayes and K-Nearest Neighbour (K-NN) classification methods. The accuracy of learning style detection using prior knowledge data was higher than detection results using behavioural data or hybrid data (prior knowledge + behaviour) in this studyMetode deteksi gaya belajar otomatis dapat dibagi menjadi dua yakni Metode Data-Driven (DD) dan Literature Based (LB). Kedua metode deteksi gaya belajar otomatis tersebut memiliki kelebihan dibandingkan metode de- teksi gaya belajar konvensional, karena metode deteksi gaya belajar otomatis menggunakan sumber data eksternal seperti forum, kuis dan kunjungan bahan ajar yang lebih akurat dibandingkan dengan kuesioner yang digunakan dalam metode deteksi gaya belajar konvensional. Meskipun demikian, hasil deteksi otomatis tidak selalu mencerminkan gaya belajar.. Untuk mengatasi kendala tersebut, penelitian ini mengusulkan metode deteksi gaya belajar yang mengambil data dari sumber internal pembelajar yakni prior knowledge. Prior knowledge diusulkan karena tidak berdasarkan pada perilaku pembelajar yang cenderung bersifat dinamis, namun lebih kepada pengetahuan atau keterampilan yang dimiliki, yang lebih mencerminkan karakteristik pembelajar. Penelitian ini mengusulkan sebuah metode untuk mendeteksi gaya belajar otomatis dengan memanfaatkan prior knowledge. Kebaruannya terletak pada kerangka kerja (framework) deteksi gaya belajar yang terdiri dari tiga tahapan yaitu: tahapan pembangunan pertanyaan prior knowledge, tahapan pengukuran prior knowledge, dan tahapan deteksi gaya belajar dengan menggunakan metode klasifikasi svm, naïve bayes dan k-nn, Penelitian ini menghasilkan akurasi hasil deteksi gaya belajar dengan data prior knowledge yang lebih tinggi dibandingkan dengan hasil deteksi yang menggunakan data perilaku dan data hybrid (prior knowledge + perilaku
Performance improvement of the shredder machines using IoT-based overheating controller feature
Full text linkPlastic shredding plays an essential role in the plastic waste recycling process. Plastic waste can be enumerated manually using a knife and scissors or a crushing machine. The use of a shredder machine to chop plastic waste, especially those whose primary drive is an electric motor, often experience problems. The main obstacle is the need for high power consumption (more than 1 HP) and the reliability of the drive elements against overheating. Overheating can damage the electrical circuit components that connect the power supply to electric motors, especially AC electric motors, causing a lot of loss in terms of performance and user safety. Internet of Things (IoT) technology is widely used to minimize energy resources by automating various systems. This study proposed the design of a shredder machine with a control system using IoT technology integrated with a shredder and conveyor machine designed using the Quantity Functional Diagram (QFD) method. The advantages of the shredder machine presented in this study are that it can operate at home using electric power, is more flexible, and minimizes overheating with an IoT-based overheating controller. This research succeeded in keeping the temperature of the electric motor of the shredder machine stable at a temperature of 40℃-55℃. The average delay of the IoT module to control on and off the shredder machine design system in this study is 219 ms and 200 ms, which are in the good category according to the Telecommunications and Internet Protocol Harmonization Over Network (TIPHON) standard
Imputasi KNN terhadap Nilai yang Hilang dari Prediksi Durasi Hujan Berbasis Regresi pada Data BMKG
No full textThe prediction of rain duration based on data from the Meteorology, Climatology, and Geophysics Agency (BMKG) is an important issue but remains an open problem. At the same time, several studies have shown that missing values can cause a decrease in the performance of the model in making predictions. This study proposes k-nearest neighbors (KNN) imputation to overcome the problem of missing values in predicting rain duration. The source of the rain duration prediction dataset is the BMKG data. We compared gradient boosting regression (GBR), adaptive boosting regression (ABR), and linear regression (LR) for the regression model for predicting rain duration. We compared the KNN imputation method with several benchmark methods, including zero imputation, mean imputation, and iterative imputation. Parameters r2, mean squared error (MSE) and mean bias error (MBE) measure the performance of these imputation methods. The test results show that for rain duration prediction using the regression method, GBR shows the best performance, both for train data and test data with r2 = 0.915 and 0.776, respectively. Then our proposed KNN imputation has the best performance for missing value imputation compared to the benchmark imputation method. The prediction values of r2 and MSE when using KNN imputation at Missing Percentage = 90% are 0.71 and 0.36, respectively.Abstrak — Prediksi durasi hujan berdasarkan data Badan Meteorologi, Klimatologi, dan Geofisika (BMKG) merupakan isu penting namun tetap menjadi permasalahan terbuka. Pada saat yang sama, beberapa penelitian menunjukkan bahwa nilai yang hilang dapat menyebabkan penurunan kinerja model dalam membuat prediksi. Penelitian ini mengusulkan imputasi k-nearest neighbor (KNN) untuk mengatasi masalah missing value dalam memprediksi durasi hujan. Sumber dataset prediksi durasi hujan adalah dari data BMKG. Kami membandingkan gradient boosting regression (GBR), adaptive boosting regression (ABR), dan linear regression (LR) untuk model regresi untuk memprediksi durasi hujan. Kami membandingkan metode imputasi KNN dengan beberapa metode benchmark, termasuk imputasi nol, imputasi rata-rata, dan imputasi berulang. Parameter r2, mean squared error (MSE) dan mean bias error (MBE) mengukur kinerja metode imputasi ini. Hasil pengujian menunjukkan bahwa untuk prediksi durasi hujan dengan menggunakan metode regresi, GBR menunjukkan performa terbaik, baik untuk data latih maupun data uji dengan masing-masing r2 = 0,915 dan 0,776. Kemudian imputasi KNN yang kami usulkan memiliki kinerja terbaik untuk imputasi nilai yang hilang dibandingkan dengan metode imputasi benchmark. Nilai prediksi r2 dan MSE saat menggunakan imputasi KNN pada Missing Percentage = 90% masing-masing adalah 0,71 dan 0,36