KURNIAWAN, ASEP (2023) ANALISA KONDISI OPERASI CIRCULATED FLUIDIZED BED (CFB) BOILER MENGGUNAKAN MACHINE LEARNING. Master thesis, UNIVERSITAS SULTAN AGENG TIRTAYASA.

	Text ASEP KURNIAWAN_7780210001_fulltext.pdf Restricted to Registered users only Download (22MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_CP.pdf Restricted to Registered users only Download (17MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_01.pdf Restricted to Registered users only Download (55MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_02.pdf Restricted to Registered users only Download (55MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_03.pdf Restricted to Registered users only Download (55MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_04.pdf Restricted to Registered users only Download (55MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_05.pdf Restricted to Registered users only Download (55MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_Ref.pdf Restricted to Registered users only Download (55MB) | Request a copy
	Text ASEP KURNIAWAN_7780210001_Lamp.pdf Restricted to Registered users only Download (55MB) | Request a copy

Abstract

Circulated Fluidized Bed (CFB) boilers are a type of boiler used in several low-capacity steam power plants (PLTU) in the Indonesian archipelago. CFB boiler is a type of boiler that has a complex operating pattern because it involves two phenomena, namely hydrodynamics and combustion simultaneously in the furnace. Reliability problems arise due to the complexity of the CFB boiler operation pattern. Temperature and pressure throughout the furnace are important parameters to maintain because related to the coal combustion process. An accurate analysis of the operating conditions of the CFB boiler is needed to be able to find out the parameters that most influence the condition of the CFB boiler. The purpose of this study was to determine the operating pattern of CFB boiler pressure and temperature so that it can be used to predict optimal conditions in increasing the reliability and efficiency of the boiler. The method used in this study is an Artificial Neural Network (ANN) which is a method in machine learning that can be used to make predictions based on data. The operating conditions of pressure and temperature at several points and the rate of steam production at the CFB Boiler can be predicted by going through the dataset processing stages, namely training, validating, and testing of the specified operating data parameters. The results of this study indicate that machine learning models can accurately predict CFB Boiler temperature conditions with an error R of 0.9993. Parameters of coal feed flow rate, primary air flow rate (PA flow), and secondary air flow rate (SA flow) have a big role in changing operating conditions. Based on the sensitivity test using the ANN model, it can be seen that the most optimum PA and SA flow configuration settings for steam production and boiler operating conditions. The configuration at low load is the PA flow setting of 145000 Nm3/h and SA flow of 190000 Nm3/h, the PA:SA ratio is 43:57 with a bed temperature of 900 degC and a boiler efficiency of 83%. Meanwhile, at high loads, the PA:SA ratio is 45:51, namely 180000 Nm3/h and SA 190000 Nm3/h with a bed temperature of 905,01 degC and a boiler efficiency of 75%.

Item Type:	Thesis (Master)
Contributors:	Contribution Contributors NIP/NIM Thesis advisor IRAWAN, ANTON 197510012008011007 Thesis advisor KURNIAWAN, TEGUH 198305062006041002
Additional Information:	Circulated Fluidized Bed (CFB) boiler merupakan salah satu jenis boiler yang digunakan pada beberapa pembangkit listrik tenaga uap (PLTU) dengan kapasitas rendah di negara kepulauan Indonesia. CFB boiler merupakan jenis boiler yang memiliki pola operasi kompleks karena melibatkan dua fenomena yaitu hidrodinamika dan pembakaran sekaligus dalam furnace. Permasalahan kehandalan timbul akibat kompleksitas pola operasi CFB boiler. Temperature dan tekanan sepanjang furnace merupakan parameter penting untuk dijaga karena berkaitan dengan proses pembakaran batubara. Analisis kondisi operasi pada CFB boiler secara akurat diperlukan untuk dapat mengetahui parameter yang paling berpengaruh terhadap kondisi CFB boiler. Tujuan penelitian ini adalah mengetahui pola operasi tekanan dan temperature CFB boiler sehingga dapat digunakan untuk memprediksi kondisi optimal dalam meningkatkan kehandalan dan efisiensi boiler. Metode yang digunakan pada penelitian ini adalah Artificial Neural Network (ANN) yang merupakan salah satu metode pada machine learning yang dapat digunakan untuk melakukan prediksi berbasiskan data. Kondisi operasi tekanan dan temperature beberapa titik serta laju produksi steam pada CFB Boiler dapat diprediksi dengan melalui tahapan processing dataset yaitu training, validating, dan testing terhadap parameter data operasi yang ditentukan. Hasil penelitian ini menunjukkan model machine learning dapat memberikan prediksi kondisi temperature CFB Boiler secara akurat dengan error R 0,9993. Parameter input laju alir batubara (Coal feed), laju alir udara primer (PA flow) dan laju alir udara sekunder (SA flow) memiliki peran besar dalam perubahan kondisi operasi. Berdasarkan uji sensitivitas menggunakan model ANN dapat diketahui setting konfigurasi PA dan SA flow yang paling optimum terhadap produksi steam dan kondisi operasi boiler. Konfigurasi pada beban rendah adalah pada setting PA flow 145000 Nm3/h dan SA flow 190000 Nm3/h ratio PA:SA yaitu 43:57 dengan bed temperature 900 degC dan efisiensi boiler 83%. Sedangkan pada beban tinggi ratio PA:SA yaitu 45:51 yaitu 180000 Nm3/h dan SA 190000 Nm3/h dengan bed temperature 905,01 degC dan efisiensi boiler 75% .
Uncontrolled Keywords:	Keywords : CFB Boiler, Neural Network, Efficiency. Kata kunci : CFB Boiler, Neural Network, Efisiensi.
Subjects:	T Technology > T Technology (General)
Divisions:	08-Pascasarjana 08-Pascasarjana > 84104-Magister Teknik Kimia
Depositing User:	Mr Asep Kurniawan
Date Deposited:	24 May 2023 15:30
Last Modified:	24 May 2023 15:30
URI:	http://eprints.untirta.ac.id/id/eprint/24317

Actions (login required)

View Item