There are many different sources of loss in gas turbines. The turbine tip clearance loss is the focus of this work. In gas turbine components such as compressor and turbine the presence of rotating blades necessitates a small annular tip clearance between the rotor blade tip and the outer casing. This clearance, although mechanically necessary, may represent a source of large loss in a turbine. The gap height can be a fraction of a millimeter but can have a disproportionately high influence on the stage efficiency. A large space between the blades and the outer casing results in detrimental leakages, while contact between them can damage the blades. Therefore, the evaluation of the sources of the performance degradation independently presents useful information that can aid in the maintenance action. As part of the overall blade loss the turbine tip clearance loss arises because at the blade tip the gas does not follow the intended path and therefore does not contribute to the turbine power output and interacts with the outer wall boundary layer. Increasing turbine tip clearance causes performance deterioration of the gas turbine and therefore increases fuel consumption. The increase in turbine tip clearance may as a result of rubs during engine transients and the interaction between the blades and the outer casing. This work deals with the study of the influence of the turbine tip clearance on a gas turbine engine, using a turbine tip clearance model incorporated to an engine deck. Actual data of an existing engine were used to check the validity of the procedure. This paper refers to a single shaft turbojet engine under development, operating under steady state condition. Different compressor maps were used to study the influence of the curve shapes on the engine performance. Two cases were considered for the performance simulation: constant corrected speed and constant maximum cycle temperature.

Poor part-load performance is a well-known undesirable characteristic of gas turbines. Running off-design, both compressor and turbine lose performance. Flow misalignment at the various rows causes losses to increase sharply, thereby decreasing net output faster than decreasing fuel consumption. To bring the flow to alignment with the blade passages, it is required to restagger the blades both at the compressor and at the turbine. To avoid mechanical complexities, it is generally accepted to restagger only the stators. This work deals with a numerical approach to the simulation of a gas turbine equipped with variable stators at the compressor and at the turbine, enabling the search for better-performance operation. A computer program has been developed to simulate virtually any gas turbine having variable stators at the compressor stages and turbine nozzle guide vanes. Variable-inlet guide vanes (VIGVs), variable-stator vanes (VSVs), variable-nozzle guide vanes (VNGVs), variable-geometry compressors (VGCs) and variable-geometry turbines (VGTs) are the focus in this work, which analyses a one-shaft free power turbine for power generation in the search for performance improvement at part load.

This paper describes a procedure to measure the performance of detection and isolation of multiple faults in gas turbines using artificial neural network and optimization techniques. It is on a particular form of artificial neural networks, the traditional multi-layer perceptron (MLP). Error back-propagation and different activation functions are used. The main goal is to recognize single, double and triple faults in a turboshaft engine, whose performance data were output from a gas turbine simulator program, tuned to represent the engine running at an existing power station. MLP network is a nonlinear interpolation function usually made of input layer, hidden-layer and output-layer, with different neuronal units, but in this work, only one hidden-layer was used. Weights were altered by error back-propagation from the initial values established from a seed fixed between 0 and 1. The activation function in the MLP algorithm is the sigmoid function. The best moment to stop the training process and avoid the over fitting problem was chosen by cross-validation. Optimization of convergence error was achieved using the momentum criteria and reducing the oscillation problem in all nets trained. Several configurations of the neural network have been compared and evaluated, using several noise graduations incorporated to the data, aiming at finding the network most suitable to detect and isolate multiple faults in gas turbines. Based on the results obtained it is inferred that the procedure reported herein may be applied to actual systems in order to assist in maintenance programs, at least.

No abstract

No abstract