Perancangan Sistem Prediksi Keandalan Generator Turbin Uap Di PT. PJB UP Gresik Berbasis Jaringan Syaraf Tiruan

Al Hamda, Moch Hamdani (2023) Perancangan Sistem Prediksi Keandalan Generator Turbin Uap Di PT. PJB UP Gresik Berbasis Jaringan Syaraf Tiruan. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02311940000077-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy

Abstract

PT. PJB Unit Pembangkitan Gresik merupakan salah satu perusahaan yang beroperasi pada bidang pembangkitan listrik yang menyumbang 31.45% pembangkitan dari semua unit pembangkitan yang dimilki oleh PT. PJB. PT. PJB Unit Pembangkitan Gresik memilki beberapa pembangkit, salah satunya pembangkit listrik tenaga gas dan uap (PLTGU) dimana
memiliki generator turbin uap sebagai salah satu komponen penting.Pentingnya evaluasi keandalan dari generator turbin uap minimal selama satu masa turn around (5000 jam) dapat mempengaruhi kinerja generator turbin uap. Sistem prediksi keandalan dibuat berbasis jaringan syaraf tiruan, dimana jaringan syaraf tiruan digunakan untuk memprediksi daya keluaran dari turbin dengan 9 masukkan data proses. Variasi node yang digunakan dalah 5,10,dan 15. Node berjumlah 15 dipilih karena semakin banyak node semakin kecil nilai mean square error (MSE). Deteksi anomali dilakukan pada data asliterhadap hasil jaringan syaraf tiruan dan setiap anomali dianggap kegagalan. Nilai i merupakan kegagalan ke-i pada satu periode turn around dan nilai n merupakan total kegagalan terbanyak yang terjadi dari 9 periode turn around dari 2018-2022. Keandalan dihitung dan didapat nilai terbaik pada periode Januari 2018 – Juli 2018 dengan nilai 0.874
=====================================================================================================================================
PT. PJB Gresik Generation Unit is one of the companies operating in the electricity generation sector which accounts for 31.45% of generation from all generation units owned by PT. PJB. PT. PJB Gresik Generation Unit has several generators, one of which is a gas and steam power plant which has a steam turbine generator as an important component. The importance of evaluating the reliability of a steam turbine generator for at least one turn around period (5000 hours) can affect performance steam turbine generator. The reliability prediction system is made based on artificial neural networks, where artificial neural networks
are used to predict the output power of the turbine with 9 inputs of process data. Variations of the nodes used are 5, 10, and 15. The number of nodes 15 was chosen because the more nodes the smaller the mean square error (MSE). Anomaly detection is performed on the original data on the results of the artificial neural network and any anomaly is considered a failure. The value of i is the i-th failure in one turn around period and the value of n is the most total failures that have occurred out of the 9 turn around periods from 2018-2022. Reliability is calculated and the best value is obtained in the period January 2018 - July 2018 with a value of 0.874.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Deteksi Anomali, Generator Turbin Uap, Jaringan Syaraf Tiruan, Keandalan, Anomaly Detection, Artificial Neural Network, Reliability, Steam Turbine Generator
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7 Turbogenerators. Steam-turbines T Technology > TS Manufactures > TS173 Reliability of industrial products
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Moch Hamdani Al Hamda
Date Deposited:	20 Jul 2023 07:42
Last Modified:	20 Jul 2023 07:42
URI:	http://repository.its.ac.id/id/eprint/98731

Actions (login required)

View Item