Device and method for monitoring a gas turbine

• US 8,831,911 B2
• Filed: 04/30/2009
• Issued: 09/09/2014
• Est. Priority Date: 05/08/2008
• Status: Active Grant

• Current Assignee: MTU Aero Engines GmbH (MTU)
• Original Assignee: MTU Aero Engines GmbH (MTU)
• Inventors: Lipowsky, Holger, Staudacher, Stephan
• Primary Examiner(s): COSIMANO, EDWARD R
• Application Number:
  US12/991,322
• Publication Number:
  US 20110153273A1
• Time in Patent Office: 1,958 Days
• Field of Search: 73/86, 73/87, 731/120.1, 734/321, 738/658, 738/659, 702/1, 702/33, 702/34, 702/127, 702/179, 702/181, 702/182, 702/183, 702/187, 702/189, 708/100, 708/105, 708/200
• US Class Current: 702/182
• CPC Class Codes: G01M 15/14 Testing gas-turbine engines or jet-propulsion engines
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G06F 11/30 Monitoring
• CPC Class Codes: G06F 17/40 Data acquisition and logging for input to computer G06F3/00
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A device and method for monitoring a gas turbine is disclosed. The device includes a receiver for receiving condition values of the gas turbine and an analytical device. The analytical device is designed to determine condition information from the condition values of the gas turbine using Bayes'"'"' Prediction.

Claims

1. A device for monitoring a gas turbine, comprising:

a receiver, wherein the receiver receives condition values of the gas turbine; and

an analytical device, wherein the analytical device determines condition information from the condition values of the gas turbine using Bayes'"'"' Prediction and wherein the analytical device identifies a rapid change in the condition values as condition information.

2. The device according to claim 1, wherein the analytical device identifies a prognosis about a future course of the condition values as condition information.

3. The device according to claim 1, wherein the analytical device determines the condition information by a comparison of a cumulative Bayes'"'"' factor with a threshold value.

4. The device according to claim 1, wherein the condition values are measured values of the gas turbine.

5. The device according to claim 1, wherein the condition values are residua which identify a deviation of a measured value from an expected value assigned to the measured value.

6. The device according to claim 1, wherein the Bayes'"'"' Prediction includes a description of the condition values by at least a dynamic linear model.

7. The device according to claim 6, wherein the dynamic linear model describes at least one linear trend of the condition values.

8. The device according to claim 6, wherein the dynamic linear model describes at least one higher order polynomial trend of the condition values.

9. A device for monitoring a gas turbine, comprising:

a receiver, wherein the receiver receives condition values of the gas turbine; and

an analytical device, wherein the analytical device determines condition information from the condition values of the gas turbine using Bayes'"'"' Prediction;

wherein the Bayes'"'"' Prediction includes a description of the condition values by at least a dynamic linear model;

and wherein the Bayes'"'"' Prediction includes a determination of a probability density of a current dynamic linear model and a determination of a probability density of an alternative dynamic linear model, wherein a mean value of the alternative dynamic linear model is offset by a specific value with respect to the current dynamic linear model.

10. The device according to claim 9, wherein the Bayes'"'"' Prediction includes a comparison of the probability density of the current dynamic linear model with the probability density of the alternative dynamic linear model.

11. The device according to claim 9, wherein the Bayes'"'"' Prediction includes a determination of Bayes'"'"' factors from the probability density of the current dynamic linear model and the probability density of the alternative dynamic linear model.