Airborne particles ingested in aircraft engines deposit on compressor blading and end walls. Aerodynamic surfaces degrade on a microscopic and macroscopic scale. Blade row, compressor, and engine performance deteriorate. Optimization of maintenance scheduling to mitigate these effects requires modeling of the deterioration process. This work provides a deterioration model on blade row level and the experimental validation of this model in a newly designed deposition test rig. When reviewing previously published work, a clear focus on deposition effects in industrial gas turbines becomes evident. The present work focuses on quantifying magnitudes and timescales of deposition effects in aircraft engines and the adaptation of the generalized Kern and Seaton deposition model for application in axial compressor blade rows. The test rig's cascade was designed to be representative of aircraft engine compressor blading. The cascade was exposed to an accelerated deposition process. Reproducible deposition patterns were identified. Results showed an asymptotic progression of blade row performance deterioration. A significant increase in total pressure loss and decrease in static pressure rise were measured. Application of the validated model using existing particle concentration and flight cycle data showed that more than 95% of the performance deterioration due to deposition occurs within the first 1000 flight cycles.

No abstract

The performance of gas turbines degrades over time due to deterioration mechanisms and single fault events. While deterioration mechanisms occur gradually, single fault events are characterized by occurring accidentally. In the case of single events, abrupt changes in the engine parameters are expected. Identifying these changes as soon as possible is referred to as detection. State-of-the-art detection algorithms are based on expert systems, neural networks, special filters, or fuzzy logic. This paper presents a novel detection technique, which is based on Bayesian forecasting and dynamic linear models (DLMs). Bayesian forecasting enables the calculation of conditional probabilities, whereas DLMs are a mathematical tool for time series analysis. The combination of the two methods can be used to calculate probability density functions prior to the next observation, or the so called forecast distributions. The change detection is carried out by comparing the current model with an alternative model, where the mean value is shifted by a prescribed offset. If the forecast distribution of the alternative model better fits the actual observation, a potential change is detected. To determine whether the respective observation is a single outlier or the first observation of a significant change, a special logic is developed. In addition to change detection, the proposed technique has the ability to perform a prognosis of measurement values. The developed method was run through an extensive test program. Detection rates >92% have been achieved for changed heights, as small as 1.5 times the standard deviation of the observed signal (sigma). For changed heights greater than 2 sigma, the detection rates have proven to be 100%. It could also be shown that a high detection rate is gained by a high false detection rate (∼2%). An optimum must be chosen between a high detection rate and a low false detection rate, by choosing an appropriate uncertainty limit for the detection. Increasing the uncertainty limit decreases both detection rate and false detection rate. In terms of prognostic abilities, the proposed technique not only estimates the point of time of a potential limit exceedance of respective parameters, but also calculates confidence bounds, as well as probability density and cumulative distribution functions for the prognosis. The conflictive requirements of a high degree of smoothing and a quick reaction to changes are fulfilled in parallel by combining two different detection conditions.

Atmospheric air is always contaminated by liquid or solid particles of different size, concentration and chemical composition. This leads to performance degradation during the operation of stationary or flying gas turbines. Erosion and the deposition of particles along the flow path are of particular importance. Multiple numerical studies investigated the influences of these phenomena. However, the basic challenge of modelling the particle wall interaction and its data spread with sufficient accuracy remains. In this work a model that estimates the statistical spread of rebound data due to target surface roughness through analytical considerations is presented. The model predicts the local impact angle of an individual particle by evaluating how deep a particle can theoretically penetrate the target surface with respect to its size. Based on roughness profiles which have been found to be characteristic for performance deterioration in compressor application a sensitivity study is conducted. A dimensionless roughness parameter Φ R was found that characterizes the effect of target surface roughness on rebound spread data. The spread model is connected with a quasi-physical model, to evaluate the effect of surface roughness for a particle's individual rebound behaviour. The synthesized data is discussed by taking into account measurement data reported in literature.

Recoverable and non-recoverable performance degradation has a significant impact on power plant revenues. A more in depth understanding and quantification of recoverable degradation enables operators to optimize plant operation. OEM degradation curves represent usually non-recoverable degradation, but actual power output and heat rate is affected by both, recoverable and non-recoverable degradation. This paper presents an empirical method to correct longterm performance data of gas turbine and combined cycle power plants for recoverable degradation. Performance degradation can be assessed with standard plant instrumentation data, which has to be systematically stored, reduced, corrected and analyzed. Recoverable degradation includes mainly compressor and air inlet filter fouling, but also instrumentation degradation such as condensate in pressure sensing lines, condenser or bypass valve leakages. The presented correction method includes corrections of these effects for gas turbine and water steam cycle components. Applying the corrections on longterm operating data enables staff to assess the non-recoverable performance degradation any time. It can also be used to predict recovery potential of maintenance activities like compressor washings, instrumentation calibration or leakage repair. The presented correction methods are validated with long-term performance data of several power plants. It is shown that the degradation rate is site-specific and influenced by boundary conditions, which have to be considered for degradation assessments.