Husnain, Husnan Ali (2025) Pengendalian Kualitas Sistem Heat Recovery Steam Generator di PLTGU PT PLN Nusantara Power Unit Pembangkit Gresik Menggunakan Diagram Kontrol Max-MEWMA Berbasis Model Random Forest Regressor. Other thesis, Institut Teknologi Sepuluh Nopember.
![]() |
Text
5003211133-Undergraduate_Thesis.pdf Download (8MB) |
Abstract
Sistem Heat Recovery Steam Generator (HRSG) pada PLTGU PT. PLN Nusantara Power UP Gresik berperan penting dalam meningkatkan efisiensi energi dan menurunkan emisi karbon. Namun, pengendalian kualitas pada sistem ini menghadapi tantangan signifikan, seperti autokorelasi tinggi dan deteksi pergeseran kecil pada parameter kritis, yaitu Fuel Gas Flow (FGF), Combustion Pressure (CP), dan Gas Temperature (GT). Penelitian ini menggunakan penerapan diagram kontrol Max-MEWMA berbasis residual dari model Random Forest Regressor untuk mengatasi permasalahan tersebut. Model dibangun berdasarkan lag-lag signifikan dari analisis PACF, dan residual yang dihasilkan diuji untuk memenuhi asumsi independensi. Hasil pengujian menunjukkan bahwa residual telah bebas dari autokorelasi, meskipun tidak sepenuhnya memenuhi asumsi normal multivariat. Diagram kontrol Max-MEWMA kemudian diterapkan pada data fase I dan II untuk mendeteksi sinyal out-of-control secara simultan. Hasil analisis menunjukkan bahwa proses belum sepenuhnya terkendali secara statistik, dengan karakteristik Fuel Gas Flow (FGF) dan Combustion Pressure (CP) sebagai penyumbang sinyal out-of-control terbanyak. Temuan ini divalidasi melalui diagram Ishikawa, yang mengidentifikasi faktor manusia, mesin, metode, dan pengukuran sebagai penyebab utama. Meskipun indeks kapabilitas multivariat (MCp dan MCpk) bernilai > 1, namun proses belum dapat dinyatakan kapabel karena belum memenuhi kestabilan statistik. Penelitian ini menunjukkan bahwa penggabungan metode Random Forest Regressor dan Max-MEWMA efektif dalam meningkatkan sensitivitas deteksi dan mendukung pengendalian kualitas sistem HRSG secara real-time.
=================================================================================================================================
The Heat Recovery Steam Generator (HRSG) system at PLTGU PT. PLN Nusantara Power UP Gresik plays a critical role in improving energy efficiency and reducing carbon emissions. However, quality control in this system faces significant challenges, such as high autocorrelation and difficulty detecting small shifts in critical parameters, namely Fuel Gas Flow (FGF), Combustion Pressure (CP), and Gas Temperature (GT). This study uses the application of a Max-MEWMA control chart based on residuals from a Random Forest Regressor model to address these issues. The model was constructed using significant lags identified through PACF analysis, and the resulting residuals were tested to satisfy the independence assumption. The tests confirmed that residuals were free from autocorrelation, although not fully normally distributed. The Max-MEWMA control chart was applied to both Phase I and Phase II data to simultaneously detect out-of-control signals. The analysis revealed that the process was not fully in statistical control, with Fuel Gas Flow (FGF) and Combustion Pressure (CP) being the main contributors to the out-of-control signals. These findings were validated using an Ishikawa diagram, which identified human, machine, method, and measurement factors as the primary causes. Although the multivariate process capability indices ( ! and !") were greater than one, the process could not be deemed capable due to a lack of statistical stability. This research demonstrates that combining the Random Forest Regressor with the Max-MEWMA control chart enhances detection sensitivity and supports real-time quality control of HRSG systems.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Heat Recovery Steam Generator (HRSG), Random Forest Regressor, Max-MEWMA, Pengendalian Kualitas, Residual, Kapabilitas Proses, Out-of-Control.Heat Recovery Steam Generator (HRSG), Random Forest Regressor, Max-MEWMA, Quality Control, Residual, Process Capability, Out-of-Control. |
Subjects: | Q Science > Q Science (General) |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
Depositing User: | Husnan Ali Husnain |
Date Deposited: | 01 Aug 2025 09:25 |
Last Modified: | 01 Aug 2025 09:25 |
URI: | http://repository.its.ac.id/id/eprint/125798 |
Actions (login required)
![]() |
View Item |