Optimisasi Techno Economic Kondisi Operasi Multi-stage Independent Gas Turbine Compressor Dengan Mempertimbangkan Ketidakpastian Proses Menggunakan Algoritma Stokastik

Afandy, Muhammad Arif (2021) Optimisasi Techno Economic Kondisi Operasi Multi-stage Independent Gas Turbine Compressor Dengan Mempertimbangkan Ketidakpastian Proses Menggunakan Algoritma Stokastik. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of Laporan Thesis_02311850017003_Final.pdf] Text
Laporan Thesis_02311850017003_Final.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (5MB) | Request a copy


Centrifugal gas compressors are widely used by the upstream oil and gas industry to deliver hydrocarbon product or gas injections for oil lifting. Saka Indonesia Pangkah Limited (SIPL) is one of the companies that possess Gas Turbine Compressors (GTC) used for Gas Lift and Gas Booster which are configured serially independent through the pipeline’s arrangement. The serial configuration intends to reduce Gas Lift Compressor (GLC) compression ratio in order to increase the injection gas flow rate. The research objective is to determine the optimal operating point of gas compressor, analyze the operating conditions and the economic impact of the compression system by considering process uncertainty. GTC Serial Multistage Independent modeling using Artificial Neural Network (ANN) with Partial Least Square and Pearson methods for pre-processing input/output variable data.
This ANN-based compression system modelling uses Multi Layer Perceptron (MLP) with Finite Impulse Response (FIR) input structure and Lavenberg-Marquardt training algorithm. This model is built on operating data compiled from the actual parameters of the gas turbine and centrifugal compressor unit GLC and 1st Stage Medium Pressure Compressor (MPC-1). The data used for ANN based modeling is 3950, data set is then allocated for training 76% and validation 24%. Analysis and optimization of centrifugal gas compressors serial independent operating conditions using Genetic Algorithm Optimization (GAO) with the objective function of maximum total gas delivery and compressor shaft rotational speed as optimized variables where both compressors operation at stable region as constraint. The best optimization simulation results are obtained in the “Maximum Production and Gas Lift” scenario with the objective function value of total gas delivery 85.10 kg/min and MPC-1 speed setpoints at 11683 RPM and GLC 15160 RPM. Optimization of serial independent compressors operating condition also increasing revenue of natural gas sales in that scenario, where the total income value is 143,434.05 USD/day with a profit against the pre-optimized value of 28,294.89 USD/day.

Item Type: Thesis (Masters)
Uncontrolled Keywords: GTC, Serial Multistage, Artificial Neural Network
Subjects: Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics)
Q Science > QA Mathematics > QA336 Artificial Intelligence
Q Science > QA Mathematics > QA402.5 Genetic algorithms.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QC Physics > QC151 Fluid dynamics
T Technology > T Technology (General) > T57.5 Data Processing
T Technology > T Technology (General) > T57.62 Simulation
T Technology > TJ Mechanical engineering and machinery > TJ778 Gas turbines
T Technology > TJ Mechanical engineering and machinery > TJ990 Compressors
Divisions: Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30101-(S2) Master Thesis
Depositing User: Muhammad Arif Afandy
Date Deposited: 08 Sep 2021 05:07
Last Modified: 08 Sep 2021 05:07
URI: http://repository.its.ac.id/id/eprint/91837

Actions (login required)

View Item View Item