#TITLE_ALTERNATIVE#

In the Hyl III Direct Reduction Plant, PT. XYZ, the reduction process to transform iron ore into sponge iron is performed. This process needs the process gas (H2 and CO) at 930 degrees C. Therefore, the gas heater system is required. Its purpose is to heat the process gas (H2 and CO) until it reache...

Full description

Saved in:
Bibliographic Details
Main Author: DIPAYANA (NIM 13304086), PRATAMA
Format: Final Project
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/11149
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:In the Hyl III Direct Reduction Plant, PT. XYZ, the reduction process to transform iron ore into sponge iron is performed. This process needs the process gas (H2 and CO) at 930 degrees C. Therefore, the gas heater system is required. Its purpose is to heat the process gas (H2 and CO) until it reaches the temperature at 930 degrees C. The process that occur in gas heater is combustion which use 2 kind of fuel, that are natural gas and tail gas.<p> <br /> <br /> <br /> <br /> <br /> Since the gas heater system hold an important role, a control system was designed to keep the desired process gas temperature which flow out from the gas heater system. Moreover, the control system design is included to save the natural gas used, to increase energy efficiency of the gas heater, and to reduce the environmental aspect from flue gas. The designed control system is PI controller because of its simplicity, and it can be implemented directly into Honeywell DCS TDC 3000, which currently is used by PT. XYZ. However, because of the nonlinearity and time-varying behaviour of the process, the PI controller requires a correct parameter tuning. The genetic algorithm was used to optimize the PI parameters.<p> <br /> <br /> <br /> <br /> <br /> The Gas Heater system was identified before the simulation and the controller design. The system identification was done by 2 steps, the Recursive Least Square and the Neural Network based on ADALINE structure with Least Mean Square algorithm, whereas the chosen model, for both, was the ARX. From the simulation, the designed control system keeps the process gas temperature at the desired set point and has a well enough disturbance rejection. Furthermore, the natural gas use was conserved up to 21.7%, the efficiency was improved from 79% to 82%, and the temperature at the stack became lower from 189 degrees C to 116 degrees C, so the polution from calor energy could be reduced.