Articles
Peramalan Beban Listrik Harian dengan Metode Adaptive Neuro Fuzzy Inferrence System
I G.M.W. Meindra Sidemen;
Rukmi Sari Hartati;
Linawati Linawati
Jurnal Teknologi Elektro Vol 13 No 1 (2014): (January - June) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Peningkatan kebutuhan tenaga listrik di Indonesia khususnya di Bali terjadi seiring dengan meningkatnya kegiatan ekonomi dan kesejahteraan masyarakat. Peningkatan kebutuhan tenaga listrik memerlukan suatu perencanaan sistem tenaga listrik yang tepat. Perencanaan yang baik bisa dilakukan melalui peramalan yang tepat untuk kebutuhan beban listrik. Peramalan berdasarkan rentang waktu dikategorikan menjadi peramalan jangka pendek, jangka menengah dan jangka panjang. Salah satu metode dalam sistem cerdas yang dapat digunakan untuk peramalan beban adalah metode adaptive neuro fuzzy inference sistem (ANFIS). Pada penelitian ini, metode tersebut digunakan untuk peramalan beban listrik jangka pendek atau harian. Data yang dipergunakan untuk pembelajaran pada peramalan ini adalah data sebenarnya (actual data) yang diambil dari PT.PLN (Persero) Area Pengatur Distribusi Bali, mulai dari bulan Juni 2011 sampai dengan Agustus 2011 dan Desember 2012 sampai dengan Februari 2013. Pembangunan Model ANFIS menggunakan program Matlab. Fuzzy Inference System yang digunakan adalah tipe Takagi-Sugeno orde satu, dengan membership function yaitu time (jam), data histori beban listrik sebelumnya, kondisi hari sebelumnya (hari kerja atau hari libur) dan kondisi hari ini (hari kerja atau hari libur) masing-masing berjumlah 7, 10, 2 dan 2 dengan tipe fungsi keanggotan gbellmf. Penelitian ini menghasilkan nilai MAPE terbesar yaitu 9,92% yang terjadi pada bulan Juni 2011 yang nilainya lebih baik dibandingkan dengan metode JST backpropagation. Jadi metode peramalan beban listrik harian menggunakan metode ANFIS cukup akurat.
Penentuan Harga Jual Biji Jarak Kering Dengan Metode Activity Based Cost System Untuk Mendukung Pertanian Berkelanjutan: Studi Kasus Nusa Penida, Bali
Radha Kurniawan;
Rukmi Sari Hartati;
I N Satya Kumara
Jurnal Teknologi Elektro Vol 18 No 2 (2019): (Mei-Agustus) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24843/MITE.2019.v18i02.P05
The aim of this research was to determine the price of jatropha seed at farmer level in Nusa Penida island, Bali, where the location was highly potential for Jatropha seed development as the main raw material of biodiesel production. The biodiesel price wass calculated using Activity-Based Cost System method. This method calculated biodiesel’s prices more accurate because it took a detailed activities costing in the biodiesel processing. The activities consisted of the production process activities at the unit and batch levels, where the costs were grouped by activity type according to Activity-Based Cost System, then grouped again into activity cost pool. It was found that the biodiesel price was Rp. 4,076.25 per liter which was higher than the current price of Rp. 3,893.9 per liter. The calculated biodiesel prices was then calibrated against with the price of industrial diesel oil in Nusa Penida island further calculation was carried out to determine the price of dried Jathropha seeds.
Penerapan Metode Prototype dalam Perancangan Sistem Informasi Penerimaan Karyawan Berbasis Website pada Berlian Agency
Ni Luh Ade Mita Rahayu Dewi;
Rukmi Sari Hartati;
Yoga Divayana
Jurnal Teknologi Elektro Vol 20 No 1 (2021): (Januari - Juni ) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24843/MITE.2021.v20i01.P17
This research discusses the application of the prototype method in designing a website-based employee recruitment information system. The case study used in this research is Berlian Agency, which is a business in the field of digital services that serves various needs, especially in the field of marketing such as website creation, social media management, and so on. The research carried out was adjusted to the business processes that were running at Berlian Agency, as well as by ongoing events and current conditions, therefore researchers used a descriptive method in this study. The descriptive research method is a research method that aims to describe and explain objects based on existing conditions. In this study, the development method chosen was the prototype method. The application of the prototype method makes it easier to design this system because there is a prototyping section that makes it easier for researchers to make the necessary designs. This study aims to produce an information system design for employee acceptance. Based on the research results, this system has been successfully designed as expected so that the prototype that has been made can be used as a basis for continuing to work on the system to be made. The results of this design can be used as a blueprint or documentation for the company with the hope that later it can be used as literature to develop a real employee recruitment system.
Studi Pengelolaan Energi Listrik di Perusahaan Pengolahan Daging PT. Soejasch Bali
Hendri Noviyanto Pratama;
Rukmi Sari Hartati;
I Nyoman Satya Kumara
Jurnal Teknologi Elektro Vol 16 No 2 (2017): (May - Agustus) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24843/MITE.2017.v16i02p05
PT. Soejasch Bali adalah perusahaan pengolahan daging yang berdiri sejak 1983. Dalam paper ini dilakukan audit energi listrik dan deph interview berdasarkan kuesioner USAID – ICED untuk mengetahui Intensitas Konsumsi Energi di bangunan non produksi sebesar IKE 87,412 kWh/m2 dan mengetahui Konsumsi Energi Spesifik untuk perhitungan secara keseluruhan karena berupa perusahaan produksi sebesar 887,62 kWh/Ton. Peluang Hemat Energi yang bisa diterapkan di PT. Soejasch Bali yaitu dengan pengaturan jam operasional yang menghasilkan 5% penghematan energi yang dikonsumsi (turnitin chek 14%, 11okt2016)
Pendekatan Diagnostik Berbasis Extreme Learning Machine dengan Kernel Linear untuk Mengklasifikasi Kelainan Paru-Paru
Putu Prima Winangun;
I Made Oka Widyantara;
Rukmi Sari Hartati
Jurnal Teknologi Elektro Vol 19 No 1 (2020): (Januari - Juni ) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24843/MITE.2020.v19i01.P12
Abstract— an expert system can be used as a second opinion for comparison or supporting diagnosis from experts. Data mining is used to obtain information applied to this system. Whereas in conducting learning using Artificial Neural Networks which apply the Extreme Learning Machine method so that it can accelerate learning up to thousands of times. In this paper, software development is carried out to test the activation functions used in conducting learning and the variables used as input during learning.
Diagnosa Tumor Otak Berdasarkan Citra MRI (Magnetic Resonance Imaging)
Ida Bagus Leo Mahadya Suta;
Rukmi Sari Hartati;
Yoga Divayana
Jurnal Teknologi Elektro Vol 18 No 2 (2019): (Mei-Agustus) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24843/MITE.2019.v18i02.P01
Brain tumors are one of the most deadly diseases, one of the most common types is glioma, about 6 out of 100,000 patients are glioma sufferers. Digital imagery through Magnetic Resonance Imaging (MRI) is one method to help doctors analyze and classify brain tumor types. However, manual classification requires a long time and has a high risk of errors, so an automatic and accurate method is needed to classify MRI images. Convolutional Neural Network (CNN) is one of the solutions for automatic classification in MRI images. CNN is a deep learning algorithm that has the ability to learn on its own from the previous case. And from the research that has been done, the results obtained that CNN is able to complete the classification of brain tumors with high accuracy. Accuracy enhancements are obtained by developing the CNN algorithm either by determining the kernel value and / or activation function.
PERBANDINGAN PENGGUNAAN METODE OPTIMASI EXTENDED QUADRATIC INTERIOR POINT DENGAN NEURAL NETWORK UNTUK ECONOMIC DISPATCH PEMBANGKIT THERMAL DI BALI
Rukmi Sari Hartati;
I Wayan Sukerayasa
Jurnal Teknologi Elektro Vol 3 No 2 (2004): (July - December) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Makalah ini membahas alokasi pembebanan perbandingan penggunaan metode optimasi extended quadratic interior point dengan neural network untuk economic dispatch pembangkit termal di bali. Metode exstended quadratic interior point (EQIP) adalah metode deterministik yang merupakan pengembangan metodeKarmakar oleh James A. Momoh dkk. dengan berdasarkan pada perbaikan kondisi awal sehingga bisa digunakan untuk menyelesaikan permasalahan pemrograman kuadratik (non linier). Metode jaringan Syaraf Tiruan model Hofield (HNN) adalah suatu metode kecerdasan buatan yang berdasarkan pada prinsip kerja jaringan saraf manusia, yang telah banyak digunakan sebagai metode alternatif untuk menyelesaikan permasalahan-permasalahan optimasi yang sulit dimana cara konvensional gagal untuk mencapai kecepatan, keakuratan atau efficiency. Dari hasil ujicoba pada sistem pembangkit termal dibali diperoleh hasil yang mendekati sama.
Pengaruh Injeksi Daya Aktif Terhadap Keandalan Penyulang
Gesta Winantara;
Rukmi Sari Hartati;
I Wayan Sukerayasa
Jurnal Teknologi Elektro Vol 12 No 1 (2013): (January - June) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Peningkatan kesejahteraan masyarakat berimplikasi terhadap peningkatan kebutuhan tenaga listrik. Peningkatan pasokan tenaga listrik harus dibarengi dengan tingkat keandalan yang sesuai standar, sehingga masyarakat pengguna dapat menikmati energi listrik dengan kualitas yang memuaskan. Kualitas layanan dapat dilihat dari lamanya pemadaman dan seberapa sering pemadaman terjadi dalam satu satuan waktu. Indeks keandalan merupakan ukuran keandalan yang dinyatakan dalam besaran probabilitas. Keandalan suatu penyulang dipengaruhi oleh banyak faktor, baik faktor internal maupun faktor eksternal. Trend sistem distribusi menunjukkan semakin banyak penyulang yang mendapat sumber tambahan dari pembangkit-pembangkit kecil, dengan sumber energi primernya energi terbarukan yang disebut pembangkit tersebar (DG). Dalam penelitian ini dilakukan analisis bagaimana pengaruh DG terhadap keandalan penyulang. Penyulang Serangan merupakan salah satu penyulang yang mendapat pasokan sumber dari pusat pembangkit tenaga sampah (DG). Dari hasil analisis pada penyulang ini didapatkan nilai indeks titik beban (load point), SAIFI dan SAIDI yang lebih baik dari sebelum masuknya pembangkit tenaga sampah.
Penajaman Citra Satelit Landsat 8 Menggunakan Transformasi Brovey
I Made Dhanan Pradipta;
I Made Oka Widyantara;
Rukmi Sari Hartati
Jurnal Teknologi Elektro Vol 18 No 3 (2019): (September - Desember) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24843/MITE.2019.v18i03.P08
Abstract—Image sharpening is an image optimization technique by combining two satellite images by taking advantage of each images, so that the output image has good spectral and spatial quality. Image sharpening is an alternative to facilitate the recognition of an object, moreover for Landsat satellites has many bands that can be used as samples for the image sharpening process. One method used in image sharpening is Brovey Transformation. This method is widely used because it is very simple and easy in making the algorithm. The Bands which is used in the image sharpening process are composite bands (red, green, and blue) with panchromatic bands from the Landsat 8 satellite. The results of the image sharpening process are high spectral images obtained from composite images and spatial values obtained from panchromatic images. Analysis of the results of image sharpening using the Brovey Transform method is done by comparing the quality of the truecolor image before sharpened and the image that has been sharpened by calculating the maximum and minimum values of the red, green and blue bands, analyzing the mean values and also the standard deviation values of each image. Key Words— Brovey Transformation, Composite, Image Sharpening, Landsat 8
Rancang Bangun Aplikasi File Transfer Menggunakan Library ZeroMQ
I Gusti Ngurah Yoga Pawitra;
Rukmi Sari Hartati;
Yoga Divayana
Jurnal Teknologi Elektro Vol 19 No 1 (2020): (Januari - Juni ) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.24843/MITE.2020.v19i01.P08
Sending data with large data sizes between two or more computers in a company is affected by various factors, both the network connection of the two computers to the data exchange method used, one method that can be used is the ZeroMQ library. The method of sending files using the ZeroMQ method is relatively faster because the architecture is brokerless or without middleware as an intermediary so that the process of sending files can be done quickly and can blow up the resources used. The ZeroMQ method was originally created to send messages, but we can add file conversion algorithms to binary numbers, so the messages sent are binary numbers from encoded files. This application is built on the Python programming language so that it is easier to develop again. The results of the design of file transfer applications using the ZeroMQ library performed on two server computers and client computers explain that this method can be used to send files and from the results of the test the time required and the latency value is relatively faster and more stable.