The invention is explained in following description in view of the drawings that show:
Contaminant particles entrained in the cooling air of a gas turbine engine can accumulate on surfaces in the cooling air piping and channels. The inventors have recognized that such accumulations can increase maintenance costs and reduce the availability of the engine. The inventors further recognized that the cooling air is representative of the gas turbine working fluid flow at the point of cooling air extraction, and that it contains particles ingested from the environment and those entrained from internal corrosion in the compressor and the cooling system. Predicting particle accumulations using analytical methods alone has been unsatisfactory, because particle concentrations depend upon a variety of unpredictable factors such as changes in environmental conditions. Such factors can cause unexpected acceleration of corrosion or ingestion of particles. The inventors recognized that continuously monitoring particles in the cooling air could offer significant improvements in efficiency and safety of engine operation, allowing early detection of increased particle flow from any cause and facilitating a predictive capability that provides for safe, efficient operation and for condition-based maintenance.
To monitor particles suspended in the cooling air flow 32, one or more particle sensors 61-63 may be mounted at one or more location(s) L1-L3 in the cooling piping 33. An example of such a sensor is the PCME® DT990 electrodynamic sensor, which senses passing particles by detecting their electrostatic charge. The antenna of this sensor may be custom engineered to withstand the temperature and dynamic pressure of the cooling air flow 32. The particle sensors 61-63 provide particle flow data via a data bus 50 and interface circuits 52 to a data acquisition unit DAU. The particle flow data may be in units such as particle counts or particle mass density.
To convert such particle flow data to particle mass flow values, the particles may be sampled and analyzed to determine their composition and size distributions. This may be done by installing particle sampling devices 65-67 at one or more locations L1-L3 of the respective particle sensor(s) 61-63 or at other locations. The sampling devices 65-67 may have collection probes in the piping 33. The probes may be tubes or channels with isokinetic tips or openings as known in the art for diverting a portion of the cooling air to a filter, such as thimble filters S1-S3 for example. A needle valve and a differential pressure gauge across each thimble filter S1-S3 may be used to set an air flow rate through the thimble filter. This differential pressure configuration or other configurations known in the art may be used for monitoring the condition of the filter. Each thimble filter S1-S3 collects particles from the cooling air stream. Tubing upstream of the filter will be relatively hot, and may be made of heat resistant material such as Inconel® 600 to minimize oxidation inside the sampling tubing. Other tubing may be made of less expensive material, such as stainless steel. The thimble filters may be changed periodically when the differential pressure gage indicates that a significant amount of particle matter has collected. The filters S1-S3 collect samples of the particles for analysis by a laboratory. Samples may be taken under each of several stages of engine operation.
Alternate sampling strategies may be used. For example, particle sampling and analysis may be done in a given engine, and the resulting particle size and composition data may be used to calibrate particle mass flow measurements in other engines of the same type. In another strategy, a continuous particle size distribution sensor may be installed in the piping based on laser diffraction or other known techniques.
A differential pressure sensor 74 may be mounted in the piping 33. This may include a Pitot device that senses dynamic pressure and static pressure of the cooling air flow 32, and calculates a flow velocity from these measurements. It may also include a temperature sensor 72 to derive density and mass flow measurements of the cooling air flow 32. An example of this type of sensor is a Rosemount® integrated Annubar® system.
A data acquisition unit DAU collects data from the various sensors 61-63, 65-67, 69-72, and 74 of the invention via one or more electronic data buses 50. Interface circuits 52 may be provided on the buses and/or in the sensors to condition electronic signals from the sensors for communication with the data acquisition unit. The data acquisition unit DAU compiles and stores data for analysis and inquiry. It may have a data communications link 54 for off-site monitoring and analysis of the sensor data. The data acquisition unit DAU may include a processing capability and algorithms for computing derived parameters from the sensor data. Alternately, one or more of the sensors 61-63, 65-67, 69-72, and 74 may include electronics that compute some or all derived parameters.
Analysis may include comparing a parameter such as an instantaneous particle mass flow rate to a predetermined value such as a normal upper limit of particle mass flow rates for a given engine operating condition. Analysis may further include determining a maintenance requirement as a function of an integrated particle mass flow, with or without consideration of an instantaneous particle mass flow. If multiple locations are monitored, the parameters may be compiled in a matrix with a cell for each parameter per location per engine operating condition. The data and analyses may be consolidated into reports and graphs by section being monitored. These presentations may highlight exceptions such as accelerated corrosion. Analyses may be done by algorithms in the data access unit DAU or in an attached on-site computer, or they may be done remotely using a remote client computer via such algorithms and/or graphical displays for interactive operator analysis.
The present invention contemplates real time measurement of particle flow information during various phases of operation of a gas turbine engine. In a diagnostic mode, the invention may include the installation of an on-line particle detection system effective to provide a particle flow value associated with a portion of the working fluid flowing through a gas turbine engine, such as the portion diverted for cooling flow purposes. Particle flow information is then collected and analyzed for various modes of operation of the engine. Grab samples of the particles are collected as necessary to calibrate and characterize the on-line particle flow value information. Data may be collected from various locations in the fluid flow path, thereby facilitating an analysis of the various sources of particles. Similar data may be collected on a plurality of engines and statistics may be generated to represent a fleet of engines. In a monitoring mode, on-line particle data may be compared to predetermined set points for the generation of alarms for the engine operator and for the recognition of an action, such as adjusting an operating parameter of the engine, shutting down the engine, or performing a maintenance activity on the engine, for examples. Such set points may vary in response to an operating condition of the engine, such as having a higher expected particle flow rate during startup after the plant has been idle for a period of time. Such set points may also be varied as a function of an integrated value of the particle flow, such as lowering an alarm, maintenance or shutdown value as the total mass flow of particles accumulates over time. Particle flow information integrated over time may be used to predict a need for a maintenance service, such as cleaning of cooling flow passages or other surfaces, in order to prevent particle-induced changes in the engine performance. For example, the prediction of an engine shutdown date for a cleaning activity may be responsive to an instantaneous value of the particle flow rate. Changes in particle flow values may be correlated to other engine operating conditions for diagnosis and maintenance decisions. Plant design decisions may reflect the availability of such on-line particle flow information. For example, because actual, real-time particle flow data is available, a decision may be made to construct at least a portion of a cooling air system with less-expensive carbon steels that are more subject to corrosion and the generation of particles than the alternative more-expensive stainless steels or internally coated materials. Alternatively, a decision may be made to invest in corrosion-resistant materials for other portions of a plant, such as the compressor casing for example.
While various embodiments of the present invention have been shown and described herein, it will be obvious that such embodiments are provided by way of example only. Numerous variations, changes and substitutions may be made without departing from the invention herein. Accordingly, it is intended that the invention be limited only by the spirit and scope of the appended claims.