Gas path analysis for the MTT micro turbine

More Info
expand_more

Abstract

Gas turbine diagnostics is as old as the gas turbine itself. Over the years, performance based diagnostics allowed for a shift from time-based maintenance to more economical condition based maintenance playing a fundamental role in enhancing the availability and reliability of gas turbines. By monitoring the condition of the engine over time, maintenance actions can be taken based on information collected from the field. MTT (Micro Turbine Technology) is currently developing a low cost 3kWe micro-turbine CHP (Combined Heat and Power)-system by using off-the-shelf technologies. Once the system will be launched on the market an organized, cost-effective maintenance procedure will be required. The objective of this M. Sc. thesis project was to develop and demonstrate a Gas Path Analysis diagnostic concept for the micro-turbine. Gas Path Analysis (GPA) is a method to assess the condition of the gas turbine by using performance measurements from the gas path. The feasibility of the diagnostic concept was demonstrated by some case studies using data from the first generation field test units. After reviewing a number of gas turbine diagnostic techniques, a non-linear model based gas path analysis approach was chosen. For the development of the diagnostic concept, a non-linear model of a healthy reference engine was used to simulate the off-design behaviour of the engine and derive healthy performance parameter baselines. These baselines are used to compare the performance of field engines against. A component based modelling environment called GSP or the Gas turbine Simulation Program was used to simulate the effect of ambient conditions and deterioration on performance. The diagnostic concept relies on the principle that deterioration causes corrected measurement parameters to shift from the healthy reference baselines. Measurement performance parameters are first corrected to standard ISA conditions before being compared against the healthy baselines. By modelling specific types of deterioration in GSP, signature parameter shifts could be recorded for each of the deterioration modes. These signature parameter shifts are used to compare shifts in performance parameters against and determine the closest pattern-match which can be used to identify the most probable cause of deterioration. The proposed concept is capable of performing engine level diagnostics and partially component level diagnostics. Multiple fault diagnostics and quantifying the level of deterioration are more difficult due to the limited number of sensors and the relatively large impact of second-order effects such as heat-loss, auxiliary power take-off, mechanical losses, etc. The performance parameter baselines together with the derived rulesets can easily be implemented in a maintenance tool making the concept flexible and easy to use.

Files