Steam turbine performance testing
AssigneeGeneral Electric Company
InventorsScott Victor Hannula, Duncan George Watt
Life of this patent
Estimating the term…
Abstract
A steam turbine performance testing system, including at least one computer hardware device, including a neural network created using a dynamic steam turbine thermodynamic model and preliminary data collected from a steam turbine; a network tester for testing the neural network with testing data; a current performance calculator for calculating a current performance of the steam turbine from operation data of the steam turbine; and a projected performance calculator for calculating a projected performance of the steam turbine from the current performance.
Claims7
A steam turbine performance testing system, comprising:
- at least one computer hardware device including:
- a neural network created using a dynamic steam turbine thermodynamic model and preliminary data collected from a steam turbine, wherein the dynamic steam thermodynamic model allows for a variation of a thermodynamic definition of the steam turbine by modifying parameters including a stage efficiency, a section efficiency, a pressure variation, a temperature variation, and a mechanical loss, and wherein the dynamic steam turbine thermodynamic model includes a heat soak rejection model for allowing time dependent changes in the variation of the thermodynamic definition by including an amount of heat from at least one of a heat decrease as the steam turbine cools down and a heat increase as the steam turbine heats up;
- a network tester for testing the neural network with testing data;
- a current performance calculator for calculating a current performance of the steam turbine from operation data of the steam turbine;
- and
- a projected performance calculator for calculating a projected performance of the steam turbine from the current performance, wherein the projected performance is based upon a guarantee point.
A method, comprising:
- collecting, using a computer hardware device, preliminary data from at least one cycle of operation of a steam turbine;
- creating, using the computer hardware device, a neural network from the collected preliminary data using a dynamic steam turbine thermodynamic model, wherein the dynamic steam thermodynamic model allows for a variation of a thermodynamic definition of the steam turbine by modifying parameters including a stage efficiency, a section efficiency, a pressure variation, a temperature variation, and a mechanical loss, and wherein the dynamic steam turbine thermodynamic model includes a heat soak rejection model for allowing time dependent changes in the variation of the thermodynamic definition by including an amount of heat from at least one of a heat decrease as the steam turbine cools down and a heat increase as the steam turbine heats up;
- testing, using the computer hardware device, the neural network using a set of testing data;
- collecting, using the computer hardware device, operational data from the steam turbine;
- and
- using the dynamic steam turbine thermodynamic model on the neural network to:
- calculate, using the computer hardware device, a current performance of the steam turbine;
- and
- calculate, using the computer hardware device, a projected performance of the steam turbine, wherein the projected performance is based upon a guarantee point.
A computer program product comprising program code embodied in at least one non-transitory computer-readable storage medium, which when executed, enables a computer system to implement a method, the method comprising:
- collecting preliminary data from at least one cycle of operation of a steam turbine;
- creating a neural network from the collected preliminary data using a dynamic steam turbine thermodynamic model, wherein the dynamic steam thermodynamic model allows for a variation of a thermodynamic definition of the steam turbine by modifying parameters including a stage efficiency, a section efficiency, a pressure variation, a temperature variation, and a mechanical loss, and wherein the dynamic steam turbine thermodynamic model includes a heat soak rejection model for allowing time dependent changes in the variation of the thermodynamic definition by including an amount of heat from at least one of a heat decrease as the steam turbine cools down and a heat increase as the steam turbine heats up;
- testing the neural network using a set of testing data;
- collecting operational data from the steam turbine;
- and
- using the dynamic steam turbine thermodynamic model on the neural network to:
- calculate a current performance of the steam turbine;
- and
- calculate a projected performance of the steam turbine, wherein the projected performance is based upon a guarantee point.
Dependencies are inferred from the claim text.