Please use this identifier to cite or link to this item: https://repositori.uma.ac.id/handle/123456789/29662
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorIdris, Muhammad-
dc.contributor.authorKesuma, Gian Raja-
dc.date.accessioned2026-04-02T03:57:56Z-
dc.date.available2026-04-02T03:57:56Z-
dc.date.issued2025-08-
dc.identifier.urihttps://repositori.uma.ac.id/handle/123456789/29662-
dc.description104 Halamanen_US
dc.description.abstractPenelitian ini bertujuan mengoptimasi efisiensi boiler tipe SFW produksi PT Atmindo yang digunakan di PT. Pabrik pengolahan kelapa sawit dengan menggunakan metode Taguchi dan analisis regresi linear berganda. Variabel yang diteliti meliputi jumlah cangkang (kg), fiber (kg), dan temperatur steam superheater (°C), dengan kinerja turbin uap (MW) sebagai variabel respon. Eksperimen dirancang menggunakan orthogonal array L9 untuk tiga faktor dan tiga level. Data historis enam bulan telah dibersihkan dari outlier dan diuji normalitas. Hasil regresi menunjukkan nilai koefisien determinasi (R²) sebesar 93,12%, menandakan model yang baik. Variabel cangkang dan fiber berpengaruh signifikan terhadap kinerja turbin uap, sedangkan temperatur steam superheater tidak signifikan. Model regresi bebas multikolinearitas dan valid untuk prediksi serta optimasi. Evaluasi model dengan MAE, MSE, RMSE, dan MAPE menunjukkan tingkat kesalahan rendah. Penelitian menyimpulkan bahwa optimalisasi penggunaan cangkang dan fiber dapat meningkatkan efisiensi boiler, sementara pengaturan temperatur steam superheater memberikan dampak kecil. Metode Taguchi dan regresi berganda efektif dalam menentukan parameter operasi optimal untuk meningkatkan efisiensi boiler. This study aims to optimize the efficiency of the SFW type boiler produced by PT Atmindo, used at PT. Palm oil processing plant, by applying the Taguchi method and multiple linear regression analysis. The independent variables examined are the amount of shells (kg), fiber (kg), and steam superheater temperature (°C), with steam turbine performance (MW) as the dependent variable. The experiment was designed using the L9 orthogonal array for three factors at three levels. Six months of historical data were cleaned from outliers and tested for normality. Regression results showed a coefficient of determination (R²) of 93.12%, indicating a good model fit. Shell and fiber variables significantly affect steam turbine performance, whereas steam superheater temperature is not significant. The regression model is free from multicollinearity and valid for prediction and optimization. Model evaluations using MAE, MSE, RMSE, and MAPE indicate low prediction error. The study concludes that optimizing the use of shells and fiber can improve boiler efficiency, while steam superheater temperature control has minimal impact. Taguchi and multiple regression methods are effective in determining optimal operating parameters to improve boiler efficiency.en_US
dc.language.isoiden_US
dc.publisherUniversitas Medan Areaen_US
dc.relation.ispartofseriesNPM;218130017-
dc.subjectBoiler SFWen_US
dc.subjectmetode Taguchien_US
dc.subjectregresi linear bergandaen_US
dc.subjectefisiensi boileren_US
dc.subjectkinerja turbin uapen_US
dc.subjectoptimalisasien_US
dc.subjectmultiple linear regressionen_US
dc.subjectsteam turbine performanceen_US
dc.titleOptimasi Efisiensi Boiler Berbasis Metode Taguchi terhadap Pengaruh Cangkang, Fiber dan Temperatur Steam Superheater untuk Meningkatan Kinerja Turbin Uapen_US
dc.title.alternativeBoiler Efficiency Optimization Based on Taguchi Method on the Effect of Shell, Fiber and Steam Superheater Temperature to Improve Steam Turbine Performanceen_US
dc.typeThesisen_US
Appears in Collections:SP - Mechanical Engineering

Files in This Item:
File Description SizeFormat 
218130017 - Gian Raja Kesuma - Fulltext.pdfCover, Abstract, Chapter I, II, III, V, Bibliography1.84 MBAdobe PDFView/Open
218130017 - Gian Raja Kesuma - Chapter IV.pdf
  Restricted Access
Chapter IV941.48 kBAdobe PDFView/Open Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.