The effective use of data analytics in an advanced compressor performance and degradation monitoring system
Radika Lucas A , Andy Jones B C , Wesley Ford A and Matt Doyle BA Origin Energy, 135 Coronation Drive, Milton, Qld 4064, Australia.
B Xodus Group, Level 17, The Forrest Centre, 221 St. Georges Terrace, Perth, WA 6000, Australia.
C Corresponding author. Email: andy.jones@xodusgroup.com
The APPEA Journal 58(2) 723-727 https://doi.org/10.1071/AJ17086
Accepted: 1 March 2018 Published: 28 May 2018
Abstract
Origin is the upstream operator and joint venture partner in Australia Pacific LNG. Origin’s integrated gas operations require reliable, sustainable delivery of gas to the downstream LNG facility on Curtis Island. This scale of operation requires establishing a ‘single source of truth’ regarding compressor condition and performance while achieving maximum and reliable compression capacities. Therefore, capability of monitoring performance of centrifugal compressors across the fleet is considered an essential component of production surveillance.
Xodus leveraged Origin’s OSIsoft PI AF (PI Historian Asset Framework) tool. This system was used to build a compressor performance and degradation monitoring tool to accurately identify early indications of degradation in a multi-stage centrifugal compression train. The tool utilises live data from the PI historian to calculate key performance indicators which define compressor and driver operation.
Dimensionless parameter analysis allows Origin to accurately quantify performance degradation regardless of variations in plant inlet conditions at each gas processing facility. Deviation from baseline performance in dimensionless parameters such as polytropic efficiency, work input number and polytropic head coefficient is used to quantify capacity losses, additional power consumption and increase in suction pressure.
The tool provided the ability to use performance indicators to confidently determine the mode and extent of compressor degradation and prevent accelerated fouling which can lead to premature bundle changes. Also, this information helps streamline and has led to a major step change for the decision-making process concerning maximum production from rotating equipment. Additionally, this allowed operations to be confident on the condition of the compressor bundle, continue operation with higher capacities during high demand periods and ensure compressor bundle changeout is optimised for availability and economic aspects.
Keywords: capacity maximisation, centrifugal, coal seam gas, compression train, compressor, degradation, dimensionless analysis, fouling, gas processing facility, head losses, intelligent monitoring, liquid carryover, monitoring, online real-time monitoring, optimisation, polytropic efficiency, polytropic head.
Radika Lucas is a Senior Optimisation Engineer in the Planning and Optimisation team for Integrated Gas Operations at Origin Energy, Queensland, Australia. She has 13 years of experience in the oil and gas industry in varying locations in Australia. Radika has spent the last 10 years of her career supporting operations and optimising production from upstream gas networks with an emphasis on centrifugal compressors. Radika’s expertise covers design, commissioning and operation of surface facilities from the wellhead through to high-pressure transmission networks. In the last 5 years, she has worked on maximising production, eliminating defects and improving efficiencies from upstream gas processing facilities with a major focus in delivering customised solutions for centrifugal compressor controls, performance and maintenance strategies. Radika graduated from Hope College, Holland, Michigan, USA, in 2003 with a BSc (Cum laude) in Engineering and a BA in Chemistry. Radika also has a MEng from the University of Adelaide, Australia. |
Andrew Jones is Digital Asset Manager for Xodus Group, Perth, Australia and is a process engineer with 23 years of experience in the chemical, nuclear and oil and gas sectors in varying locations worldwide. He spent 19 years of his career in dynamic process simulation modelling and integrated technology-based solutions. His expertise covers both the subsea and topside simulation of fluids from the wellhead, through pipeline, topsides and onwards delivery to downstream operations. Most recently, he has focused his attentions on creating intelligent monitoring/data analytics applications. He graduated from the University of Nottingham, UK, in 1994 with a BSc (Hons) in Chemical Engineering. |
Wesley Ford is an Optimisation Engineer in the Planning and Optimisation team for Integrated Gas Operations at Origin Energy, Queensland, Australia. He has been working in the oil and gas industry for the last 4 years. Wesley’s experience covers optimisation of surface facilities from the wellhead, low-pressure gathering networks, compression and high-pressure pipeline networks, as well as dynamic process simulation modelling. His main emphasis is on centrifugal compression performance monitoring, gathering network optimisation and debottlenecking. Wesley graduated from the University of Queensland, Brisbane, Australia in 2013 with a BE/ME in Chemical Engineering (Hons). |
Matthew Doyle is a Principal Process Engineer for Xodus Group, Perth, Australia with 13 years of experience in the oil and gas industry in varying locations worldwide. Matthew is an expert in dynamic process simulation with a significant emphasis on compressor operation and control. Matthew’s experience covers design, simulation and optimisation of systems from the wellhead, through pipeline and onto the topside process systems. Matthew graduated from Curtin University of Technology, Perth, Australia, in 2004 with a BEng (Hons) in Chemical Engineering and a BSc in Applied Chemistry. |
References
Ludtke, K. H. (2004). ‘Process Centrifugal Compressors: Basics, Function, Operation, Design, Application.’ (Springer: New York.)Wilcox, E. (1999). Practical methods for field performance testing centrifugal compressors. In ‘Proceedings of the 28th Turbomachinery Symposium’. pp. 165–177.
