Syahrozhadl, Nur Laili (2023) Pengendalian Kualitas Sistem Heat Recovery Steam Generator (HRSG) di PLTGU PT PLN Nusantara Power Unit Pembangkitan Gresik Menggunakan Diagram Kontrol MEWMV dan MEWMA Berbasis Residual Model Multioutput Least Square Support Vector Regression (MLS-SVR). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 06211940000070-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
06211940000070-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (4MB) | Request a copy |
Abstract
Pembangkit Listrik Tenaga Gas Uap (PLTGU) memiliki tingkat produksi listrik yang lebih tinggi dibandingkan Pembangkit Listrik Tenaga Uap dan Pembangkit Listrik Tenaga Gas. Pengendalian kualitas dalam bidang industri sangat diperlukan agar proses yang dilakukan oleh perusahaan telah sesuai dengan yang telah ditetapkan. Studi kasus penelitian ini adalah pada sistem Heat Recovery Steam Generator (HRSG) di PLTGU PT PLN Nusantara Power UP Gresik. Pada penelitian ini karakteristik kualitas yang digunakan yaitu laju aliran bahan bakar (Fuel Gas Flow, FGF), tekanan pembakaran (Combustion Pressure, CP), temperatur gas buang (Gas Temperature, GT). Hubungan pada ketiga variabel menunjukkan adanya kasus autokorelasi. Adanya autokorelasi menyebabkan false alarm dan mendapatkan pengambilan keputusan yang tidak tepat. Analisis yang digunakan yaitu model Multi-output Least Square Support Vector Regression (MLS-SVR) untuk mengatasi data berautokorelasi dan dilakukan pengendalian menggunakan diagram kontrol MEWMV dan MEWMA. Input model MLS-SVR ditentukan berdasarkan lag yang signifikan dari plot Partial Auto Correlation Function (PACF), dimana pada penelitian ini. PACF pada FGF signifikan pada lag-1, 5, 16, 21, 23, 46, 47, dan 51. PACF pada CP signifikan pada lag-1, 16, 21, 23, 27, dan 46. Sedangkan PACF pada GT signifikan pada lag-1, 13, 16, 21, 23, 25, dan 36. Hasil input model MLS-SVR dan kombinasi hyper-parameter optimal menghasilkan nilai residual yang dapat mengurangi autokorelasi. Nilai residual digunakan pada pembuatan diagram kontrol MEWMV dan MEWMA. Pada fase I diperoleh diagram kontrol MEWMV dengan pembobot ω = 0,2 dan λ = 0,1 yang terkendali secara statistik setelah dilakukan 7 kali proses iterasi. Pada fase II dengan menggunakan ω dan λ yang sama, dapat diperoleh beberapa pengamatan yang keluar batas kontrol. Selanjutnya residual MLS-SVR pada fase I dimonitor menggunakan diagram kontrol MEWMA dengan λ = 0,1 didapatkan diagram yang telah terkendali secara statistik setelah 27 kali proses iterasi. Diagram MEWMA berbasis residual MLS-SVR pada fase II menunjukkan pengamatan yang out of control. Hasil identifikasi variabel penyebab out of control menunjukkan bahwa variabel laju aliran bahan bakar dan tekanan pembakaran menyebabkan sistem HRSG belum terkendali secara statistik
=====================================================================================================================================
Steam Gas Power Plants (PLTGU) have a higher electricity production rate than Steam Power Plants and Gas Power Plants. Quality control in the industrial field is very necessary so that the processes carried out by the company are in accordance with the predefined specification. The case study of this research is on the Heat Recovery Steam Generator system at PLTGU PT PLN Nusantara Power UP Gresik. In this study, the quality characteristics used were fuel flow rate (FGF), combustion pressure (CP), exhaust gas temperature (GT) The relationship on all those variables indicates the presence of autocorrelation. The existence of autocorrelation causes false alarms and gets incorrect decision making. The analysis used was the Multi-output Least Square Support Vector Regression (MLS-SVR) model to overcome the autocorrelated data and the residual were monitored using MEWMV and MEWMA control chart. The inputs of the MLS-SVR model are determined based on the significant lag from the Partial Auto Correlation Function (PACF) plot. The PACF plots of FGF are significant at lag-1, 5, 16, 21, 23, 46, 47, and 51, The PACF plots of CP are significant at lag-1, 16, 21, 23, 27, and 46, while PACF of GT are significant lag-1, 13, 16, 21, 23, 25, and 36. The inputs of the MLS-SVR model and the optimal hyper-parameter combination produce residual values that can reduce autocorrelation. The residuals of MLS-SVR are monitored using MEWMV and MEWMA control charts. In phase I, a MEWMV control charts with weighting ω = 0.2 and λ = 0.1 was obtained statistically after 7 iteration processes were carried out. In phase II using the same ω and λ, several observations that are out of control can be obtained. Furthermore, the MLS-SVR residual in phase I was monitored using the MEWMA control charts with λ = 0.1 obtained a statistically controlled charts after27 iteration processes. The MLS-SVR residual-based MEWMA charts in phase II shows out of control observations. The results of the identification of variables that cause out of control show that the variables of fuel gas flow and combustion pressure cause the HRSG system to have not been statistically controlled
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Heat Recovery Steam Generator (HRSG), MEWMA, MEWMV, MLS-SVR, PLTGU |
| Subjects: | Q Science > QA Mathematics > QA9.58 Algorithms T Technology > TJ Mechanical engineering and machinery > TJ164 Power plants--Design and construction |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Nur Laili Syahrozhadl |
| Date Deposited: | 20 Feb 2023 02:25 |
| Last Modified: | 20 Feb 2023 02:25 |
| URI: | http://repository.its.ac.id/id/eprint/97617 |
Actions (login required)
 |
View Item |