Model-based method to detect abnormal operations in heat exchangers

Patent metadata

Field	Value
Publication number	US-9724979-B1
Application number	US-201615140046-A
Country	US
Kind code	B1
Filing date	Apr 27, 2016
Priority date	Apr 27, 2016
Publication date	Aug 8, 2017
Grant date	Aug 8, 2017

Abstract

A computer-implemented method, environmental control system, and aircraft are provided. Environmental operating conditions data for a heat exchanger is received. Measured values from sensors measuring parameters of the heat exchanger are also received. An operating model for the heat exchanger based on the received environmental operating conditions is retrieved from a first data structure. The retrieved model includes estimated values for the sensors. A difference between the estimated values and the measured values is calculated. The calculated differences are compared to at least one threshold value. Upon the differences exceeding the at least one threshold value, the calculated differences are compared to abnormal operation models in an abnormal operation library data structure. Upon the calculated difference matching an abnormal operation model in the abnormal operation library data structure, outputting an abnormal operation identity associated with the abnormal operation model. If there is no match, a learning process develops a new model for a new abnormal operation.

First claim

What is claimed is:

1. An aircraft, comprising:
an aircraft cabin adapted to be pressurized;
a gas turbine engine, wherein the gas turbine engine includes at least one engine bleed;
an environmental control system (ECS) operable to receive bleed air from the at least one engine bleed and to provide conditioned air to the aircraft cabin, the ECS including:
a heat exchanger operable to remove heat from an airflow therethrough;
first sensors operable to detect environmental operating conditions of the heat exchanger;
second sensors operable to detect airflow conditions in the heat exchanger;
a controller operable to:
receive environmental operating conditions data for the heat exchanger from the first sensors;
receive measured values from sensors measuring parameters of the heat exchanger from the second sensors;
retrieve from a first data structure an operating model for the heat exchanger based on the received environmental operating conditions, wherein the retrieved operating model includes estimated values for the sensors;
calculate differences between the estimated values and the measured values;
compare the calculated differences to at least one threshold value;
upon the differences exceeding the at least one threshold value:
compare the calculated differences to abnormal operation models in an abnormal operation library data structure; and
upon the calculated differences matching an abnormal operation model in the abnormal operation library data structure, output an abnormal operation identity associated with the matched abnormal operation model.

2. The aircraft of claim 1 , wherein the controller is further operable to:
upon the calculated differences not matching an abnormal operation model in the abnormal operation library data structure, store the calculated differences in a new entry in the abnormal operation library data structure;
calculate an abnormal operation model for the new entry; and
store the calculated abnormal operation model in the new entry in the abnormal operation library data structure.

3. The aircraft of claim 2 , wherein the controller is further operable to:
determine an abnormal operation identity associated with the calculated differences stored in the new entry; and
associate the determined abnormal operation identity with the new entry.

4. The aircraft of claim 2 , wherein the controller calculates the abnormal operation model for the new entry by determining weights of a neural network for a basis function, wherein the basis function is based on at least the received measured values from sensors measuring parameters of the heat exchanger.

5. The aircraft of claim 4 , wherein the basis function is further based on the received environmental operating conditions data for a heat exchanger.

6. The aircraft of claim 2 , wherein the controller calculates the abnormal operation model for the new entry by determining coefficients for a quadratic equation, wherein at least the received measured values from sensors measuring parameters of the heat exchanger are inputs for the quadratic equation.

7. A computer-implemented method, comprising:
receiving environmental operating conditions data for a heat exchanger;
receiving measured values from sensors measuring parameters of the heat exchanger;
retrieving from a first data structure an operating model for the heat exchanger based on the received environmental operating conditions, wherein the retrieved operating model includes estimated values for the sensors;
calculating differences between the estimated values and the measured values;
comparing the calculated differences to at least one threshold value;
upon the differences exceeding the at least one threshold value:
comparing the calculated differences to abnormal operation models in an abnormal operation library data structure; and
upon the calculated difference matching an abnormal operation model in the abnormal operation library data structure, outputting an abnormal operation identity associated with the matched abnormal operation model.

8. The computer-implemented method of claim 7 , further comprising:
upon the calculated differences not matching an abnormal operation model in the abnormal operation library data structure, storing the calculated differences in a new entry in the abnormal operation library data structure;
calculating an abnormal operation model for the new entry; and
storing the calculated abnormal operation model in the new entry in the abnormal operation library data structure.

9. The computer-implemented method of claim 8 , further comprising:
determining an abnormal operation identity associated with the calculated differences stored in the new entry; and
associating the determined abnormal operation identity with the new entry.

10. The computer-implemented method of claim 8 , wherein calculating the abnormal operation model for the new entry comprises determining weights of a neural network for a basis function, wherein the basis function is based on at least the received measured values from sensors measuring parameters of the heat exchanger.

11. The computer-implemented method of claim 10 , wherein the basis function is further based on the received environmental operating conditions data for a heat exchanger.

12. The computer-implemented method of claim 8 , wherein calculating the abnormal operation model for the new entry comprises determining coefficients for a quadratic equation, wherein at least the received measured values from sensors measuring parameters of the heat exchanger are inputs for the quadratic equation.

13. An environmental control system (ECS) for a vehicle, comprising:
a heat exchanger operable to remove heat from an airflow therethrough;
first sensors operable to detect environmental operating conditions of the heat exchanger;
second sensors operable to detect airflow conditions in the heat exchanger;
a controller operable to:
receive environmental operating conditions data for the heat exchanger from the first sensors;
receive measured values from sensors measuring parameters of the heat exchanger from the second sensors;
retrieve from a first data structure an operating model for the heat exchanger based on the received environmental operating conditions, wherein the retrieved operating model includes estimated values for the sensors;
calculate differences between the estimated values and the measured values;
compare the calculated differences to at least one threshold value;
upon the differences exceeding the at least one threshold value:
compare the calculated differences to abnormal operation models in an abnormal operation library data structure; and
upon the calculated differences matching an abnormal operation model in the abnormal operation library data structure, output an abnormal operation identity associated with the matched abnormal operation model.

14. The ECS of claim 13 , wherein the controller is further operable to:
upon the calculated differences not matching an abnormal operation model in the abnormal operation library data structure, store the calculated differences in a new entry in the abnormal operation library data structure;
calculate an abnormal operation model for the new entry; and
store the calculated abnormal operation model in the new entry in the abnormal operation library data structure.

15. The ECS of claim 14 , wherein the controller is further operable to:
determine an abnormal operation identity associated with the calculated differences stored in the new entry; and
associate the determined abnormal operation identity with t

Assignees

• Boeing Co

Inventors

• Thumati Balaje
• Hanov Nail
• Nicks Eric L

Classifications

• B64D13/06

  the air being conditioned (pressurising B64D13/02) · CPC title

• B64D2045/0085

  Devices for aircraft health monitoring, e.g. monitoring flutter or vibration · CPC title

• B64D2013/0618

  with arrangements for reducing or managing bleed air, using another air source, e.g. ram air · CPC title

• B64D13/08

  the air being heated or cooled · CPC title

• G05B23/0254

  based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks · CPC title