Optimization of IBOM Power Plant (PG9171) using Fault Prediction on Gas Path Analysis
Abstract
Almost from the inception of the gas turbine engine (GT), users and engine manufacturers have sought an effective technique to determine the health of the gas-path components (fan, compressors, combustor, turbines) based on available gas-path measurements. The potential of such tools to save money by anticipating the need for overhaul and providing help in work scope definition is substantial, provided they produce reliable results. It furthermore therefore became desirable to monitor the engine performance and diagnose the fault even before the damage is done since the fault can cause permanent damage to the components. Preventive maintenance proves to be a better way considering the longer run. This project thus work describes how modern gas-path analysis can be used as a tool for gas turbine diagnosis. Gas path analysis is studied with the aid of fault predictions obtained from using fuzzy logic was found to be a more suitable method for gas turbine diagnosis because the set of fuzzy rules are described using common language. MATLAB Simulink environment is also used to predict the degree of fault in the gas turbine through its Gas path analysis. The linguistic variables used as inputs are temperature, pressure and speed while the linguistic variable used as output is failure. The universe of discourse for temperature is between [0, 55], pressure is [0, 1000], shaft speed is [0, 5000] and failure which is the fault is [0, 1]. A type 1 fuzzy logic model and the center of gravity method are used as the defuzzification module. From the results it is seen that the value for the highest possible fault is 0.909 and the lowest is 0.217 at 6.3 and 52.7 OC respectively. This research shows that the fault prediction probability increases at higher operating conditions of the gas turbine.
Keywords
Download Options
Introduction
The gas turbine is the most versatile item of turbo machinery today [1]. It can be used in several different modes in critical industries such as power generation, aviation, marine propulsion etc. However, the need to develop accurate and reliable models of GT for different objectives and applications has been a strong motivation for researchers to continue to work in this fascinating area.
This lends credence to the fact that models and control methodologies, based on white-box approaches rely on thermodynamic and energy balance equations, which are coupled and have a high degree of nonlinearity [2]. Consequently, models and control systems that are built with such simplified and/or non linearized equations are not accurate enough to capture system dynamics precisely. These models cannot account for the individual nuances of operating equipment and are not able to accommodate changes as the equipment ages [1]. Therefore, considering assumptions and using linearization methods for simplification and solving these complex dynamics are unavoidable.
The basic operation of a GT is Brayton cycle with air as the working fluid. The machine has three main components. The compressor which draws air into the turbine and pressurizes it, the combustion system typically made up of fuel injectors that injects a steady stream of fuel into the combustion chamber where it mixes with the air and burned at a temperature of more than 1093oC producing a high temperature- high pressure gas stream and the turbine made up of an intricate array of alternate stationary and rotating aerofoil section blades where the high temperature-high pressure gas expands and spins the rotating blade of the turbine. The rotating blade of the turbine performs a dual function. They drive the compressor to draw more compressed air into the combustion chamber and they spin a generator to produce power. A fourth component is often used to increase efficiency (turboprop, turbofan) to convert power into mechanical or electrical form (turbo shaft, electric generator) or to achieve greater power to mass/volume ratio (after burner).
Conclusion
Gas turbines are fielded as an attractive alternate power source. Although its dominance in industry and military application continues to be strong, the risk of running each unit within their safety margin has caused many eyebrows to be raised. The risk in this context is maintainability and reliability. That is why it is important to prevent to machine from break down through preventive maintenance as this will go a long way to reduce unintentional failure when the machine is needed the most. To this effect, GPA and fault prediction is the way forward for better operation of a gas turbine both in power generation and other applications.
The significant conclusions made from the project are listed below.
A detailed study of types of gas turbine was carried out.
Working principles of a GT cycle and their performance are well understood.
A detailed study on fault prediction was undertaken.
PG9171 GT of the Ibom Power Plant Company at Ikot Abasi was used as a case study.