Compressor bleed valve health assessment systems and methods
In one example embodiment, a gas turbine system includes a gas turbine, a compressor bleed valve, various sensors, and a valve health assessment system. The compressor bleed valve bleeds a portion of air received by the compressor during certain operating modes. The valve health assessment system includes a processor that uses a neural network model to generate a first failure mode predictor of the compressor bleed valve by processing operational data provided by the various sensors. The processor also generates a second failure mode predictor of the compressor bleed valve by using empirical data and a valve transfer function. The processor further generates a third failure mode predictor of the compressor bleed valve by using predictive data and a prediction model. The processor then applies a holistic procedure to the three failure mode predictors to derive a probability of occurrence of a failure in the compressor bleed valve.
1. A method comprising:
receiving in a processor, from at least one sensor, operational data associated with at least one component of a gas turbine;
generating, by the processor, a first failure mode predictor of a compressor bleed valve, based at least in part on the operational data, wherein generating the first failure mode predictor comprises using a model to process the operational data associated with the at least one component of the gas turbine;
generating, by the processor, a second failure mode predictor of the compressor bleed valve based at least in part on modeling at least one state of the compressor bleed valve under at least one operating condition, wherein generating the second failure mode predictor comprises using a valve transfer function indicative of at least one state of the compressor bleed valve under at least one operating condition;
generating, by the processor, a third failure mode predictor of the compressor bleed valve based at least in part on a probability of occurrence of at least one failure mode of the compressor bleed valve, wherein generating the third failure mode predictor comprises using a prediction model that is indicative of a probability of occurrence of at least one failure mode of the compressor bleed valve;
generating, by the processor, a control signal by executing a decision tree procedure that is based at least in part on the first failure mode predictor, the second failure mode predictor, and the third failure mode predictor; and
transmitting, by the processor, the control signal to a controller, the control signal modifying an operation of at least one component of the gas turbine.
2. The method of claim 1 , wherein generating the first failure mode predictor comprises using a neural network model to process the operational data associated with the at least one component of the gas turbine.
3. The method of claim 1 , wherein the control signal is further operable to improve a health status of the compressor bleed valve, pre-emptively address a projected failure condition, change an airflow, change an air pressure, or change a timing parameter.
4. The method of claim 3 , wherein the projected failure condition comprises a catastrophic failure of the compressor bleed valve, a potential failure of the compressor bleed valve, a partial failure of the compressor bleed valve, or a deteriorated performance of the compressor bleed valve.
5. The method of claim 1 , further comprising:
generating a predicted failure signal based at least in part on the first failure mode predictor, the second failure mode predictor, and the third failure mode predictor; and
displaying an advisory on the basis of the predicted failure signal to enable an operator to pre-empt a failure or improve a performance of the gas turbine.