Model-based method and system to detect heat exchanger fouling

• US 10,295,284 B2
• Filed: 08/18/2016
• Issued: 05/21/2019
• Est. Priority Date: 08/18/2016
• Status: Active Grant

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Hanov, Nail, Nicks, Eric L.
• Primary Examiner(s): Beaulieu, Yonel
• Application Number:
  US15/240,350
• Publication Number:
  US 20180051945A1
• Time in Patent Office: 1,006 Days
• CPC Class Codes: B64D 13/06 the air being conditioned pressurising B64D13/02
• CPC Class Codes: B64D 2013/0603 Environmental Control Systems
• CPC Class Codes: B64D 2013/0618 with arrangements for reducing or managing bleed air, using another air source, e.g. ram air
• CPC Class Codes: B64D 2045/0085 Devices for aircraft health monitoring, e.g. monitoring flutter or vibration
• CPC Class Codes: B64D 45/00 Aircraft indicators or protectors not otherwise provided for camouflage F41H3/00
• CPC Class Codes: F28F 17/00 Removing ice or water from heat-exchange apparatus
• CPC Class Codes: F28F 19/00 Preventing the formation of deposits or corrosion, e.g. by using filters or scrapers
• CPC Class Codes: F28F 2200/00 Prediction; Simulation; Testing measuring quantity of heat conveyed by flowing mediums G01K17/06
• CPC Class Codes: F28F 27/00 Control arrangements or safety devices specially adapted for heat-exchange or heat-transfer apparatus control arrangements in general G05
• CPC Class Codes: F28F 27/02 for controlling the distribution of heat-exchange media between different channels static flow control means in header boxes F28F9/026; arrangements of guide plates or guide vanes F28F9/22, F28F25/12
• CPC Class Codes: F28G 15/003 Control arrangements
• CPC Class Codes: Y02T 50/50 On board measures aiming to increase energy efficiency

Abstract

A computer-implemented method, environmental control system (ECS) for an aircraft, and system are provided. Air temperatures at the inlet and outlet of a ram air duct for an ECS are measured. Also, air temperatures at outlets from one or more heat exchangers arranged in the ram air duct are measured. The temperatures are typically measured when the aircraft is on the ground and when aircraft engines supplying air to the ECS are operating at idle. Shortly thereafter, the heat exchangers are manually inspected to determine a level of actual fouling (e.g., dirt, dust, or other debris accumulated on heat exchanger surfaces). A supervised machine learning or other statistical analysis is performed on the measured temperatures and associated actual fouling levels to develop a model that predicts an amount of heat exchanger fouling, based on the temperature measurements. An alert can be output if the predicted fouling exceeds a fouling threshold.

Claims

1. A computer-implemented method of detecting fouling of a heat exchanger system in a vehicle, the computer-implemented method comprising:

receiving a first temperature measurement from an upstream end of an air duct supplying air to a first heat exchanger and a second heat exchanger, wherein the first heat exchanger comprises a first heat exchanger inlet and a first heat exchanger outlet, wherein the first heat exchanger inlet is configured to receive pressurized air from a gas turbine engine operable to provide power to the vehicle;

receiving a second temperature measurement from a downstream end of the air duct;

receiving a third temperature measurement of air exiting the first heat exchanger;

receiving a fourth temperature measurement of air exiting the second heat exchanger, wherein the second heat exchanger is downstream of, and in serial fluid communication with, the first heat exchanger;

calculating a first heat exchanger fouling indication and a second heat exchanger fouling indication, by operation of one or more computer processors and based on a heat exchanger fouling model; and

upon at least one of the first heat exchanger fouling indication and the second heat exchanger fouling indication exceeding a heat exchanger fouling threshold, causing maintenance to be scheduled for the heat exchanger system.

2. The computer-implemented method of claim 1, further comprising generating the heat exchanger fouling model, wherein generating the heat exchanger fouling model comprises:

storing past instances of the first temperature measurement, the second temperature measurement, the third temperature measurement, and the fourth temperature measurement as past instances of operation of the first heat exchanger;

associating each of the stored past instances of temperature measurements with a first heat exchanger fouling indication and a second heat exchanger fouling indication; and

calculating a regression model based on a supervised machine learning algorithm applied to the stored past instances of temperature measurements and associated first and second heat exchanger fouling indications, wherein the calculated regression model is the heat exchanger fouling model.

3. The computer-implemented method of claim 2, wherein the vehicle comprises an aircraft, wherein the heat exchanger system is on-board the aircraft, wherein the first and second heat exchangers are supplied with air from a gas turbine engine, wherein the computer-implemented method includes operating the gas turbine engine at an idle condition during receipt of the first temperature measurement, the second temperature measurement, the third temperature measurement, and the fourth temperature measurement.

4. The computer-implemented method of claim 2, wherein the past instances of the first temperature measurement, the second temperature measurement, the third temperature measurement, and the fourth temperature measurement for past instances of operation of the first heat exchanger are from multiple vehicles of the same type.

5. The computer-implemented method of claim 2, wherein generating the fouling model further comprises calculating a rate of change of the heat exchanger fouling indication based on the calculated regression model, and wherein the computer-implemented method further comprises:

upon the heat exchanger fouling indication not exceeding the heat exchanger fouling threshold;

predicting a first amount time before the first heat exchanger fouling indication exceeds the heat exchanger fouling threshold and a second amount of time before the second heat exchanger fouling indication exceeds the heat exchanger fouling threshold, based on a most-recent set of temperature measurements received from the vehicle applied to the regression model and the calculated rate of change; and

outputting an alert indicating the predicted first and second amounts of time.

6. The computer-implemented method of claim 1, wherein the first heat exchanger fouling indication and the heat exchanger fouling threshold are expressed as percentages.

7. The computer-implemented method of claim 1, further comprising outputting an alert upon the at least one of the first heat exchanger fouling indication and the second heat exchanger fouling indication exceeding the heat exchanger fouling threshold;

wherein the gas turbine engine is further operable to provide pressurized air for use in a passenger cabin.

21. The computer-implemented method of claim 1, wherein the scheduled maintenance is performed for the vehicle.

8. An environmental control system (ECS) for a vehicle, the ECS comprising:

a gas turbine engine operable to provide power to the vehicle and to provide pressurized air for use in a passenger cabin;

an air duct that includes an inlet and an exhaust configured to pass air from the atmosphere therethrough;

a first heat exchanger arranged in the air duct, wherein the first heat exchanger comprises a first heat exchanger inlet and a first heat exchanger outlet, the first heat exchanger inlet configured to receive pressurized air from the gas turbine engine;

a second heat exchanger arranged in the air duct, wherein the second heat exchanger comprises a second heat exchanger inlet and a second heat exchanger outlet, the second heat exchanger inlet configured to receive air from the first heat exchanger outlet;

a first temperature sensor operable to measure a first air temperature, the first temperature sensor arranged in the air duct between the inlet and the first and second heat exchangers;

a second temperature sensor operable to measure a second air temperature, the second temperature sensor arranged in the air duct between the first and second heat exchangers and the exhaust;

a third temperature sensor operable to measure a third air temperature, the third temperature sensor arranged in the first heat exchanger outlet of the first heat exchanger;

a fourth temperature sensor operable to measure a fourth air temperature, the fourth temperature sensor arranged in the second heat exchanger outlet of the second heat exchanger; and

a controller comprising;

a computer memory operable to store;

a heat exchanger fouling model; and

received values for the measured first air temperature, second air temperature, third air temperature, and fourth air temperature; and

a computer processor operable to;

calculate a first heat exchanger fouling indication and a second heat exchange fouling indication, based on the heat exchanger fouling model; and

output a first alert upon at least one of the first heat exchanger fouling indication and the second heat exchanger fouling indication exceeding a heat exchanger fouling threshold.

9. The ECS of claim 8, further comprising a fan arranged in the air duct and operable to force air through the air duct in a first direction from the inlet to the exhaust, and wherein the fan is further operable to force air through the air duct in a second direction from the exhaust to the inlet upon at least one of the first heat exchanger fouling indication and the second heat exchanger fouling indication exceeding the heat exchanger fouling threshold.

10. The ECS of claim 8, wherein the first heat exchanger fouling indication and the second heat exchanger fouling indication are expressed as percentages.

11. The ECS of claim 8, wherein the computer memory of the controller further stores:

past instances of the measured first air temperature, second air temperature, third air temperature, and fourth air temperature;

first heat exchanger fouling indications and second heat exchanger fouling indications associated with each of the past instances of measured temperatures; and

wherein the computer processor further calculates a regression model based on a supervised learning algorithm applied to the past instances of temperature measurements and associated heat exchanger fouling indications, wherein the calculated regression model is the heat exchanger fouling model.

12. The ECS of claim 11, wherein the computer memory stores an instance of the measured first air temperature, second air temperature, third air temperature, and fourth air temperature when the gas turbine engine is at an idle condition.

13. The ECS of claim 8, wherein the controller is operable to:

transmit the received values for the measured first air temperature, second air temperature, third air temperature, and fourth air temperature, and the first and second heat exchanger fouling indications to a remote computer that generates the heat exchanger fouling model;

receive the heat exchanger fouling model from the remote computer; and

store the received heat exchanger fouling model in the computer memory.

14. The ECS of claim 13, wherein the heat exchanger fouling model is based on past instances of the measured first air temperature, second air temperature, third air temperature, and fourth air temperature for multiple ECSs for a fleet of vehicles.

15. A system, comprising:

a data input operable to receive;

past instances of temperature measurements for a fleet of vehicles, the temperature measurements comprising;

a first temperature measurement from an upstream end of an air duct supplying air to a first heat exchanger and a second heat exchanger, wherein the first heat exchanger comprises a first heat exchanger inlet and a first heat exchanger outlet, wherein the first heat exchanger inlet is configured to receive pressurized air from a gas turbine engine operable to provide power to at least one vehicle of the fleet of vehicles;

a second temperature measurement from a downstream end of the air duct;

a third temperature measurement of air exiting the first heat exchanger; and

a fourth temperature measurement of air exiting the second heat exchanger; and

a first past heat exchanger fouling indication and a second past heat exchanger fouling indication associated with each past instance;

a computer memory operable to store the received past instances of temperature measurements and associated first and second past heat exchanger fouling indications;

a computer processor operable to generate a heat exchanger fouling model by calculating a regression model based on a supervised machine learning algorithm applied to the stored measurements and associated first and second past heat exchanger fouling indications; and

a data output operable to transmit the generated heat exchanger fouling model to the fleet of vehicles;

wherein a first present exchanger fouling indication and a second present heat exchanger fouling indication for a particular vehicle are calculated based on the generated heat exchanger fouling model and based further on a current instance of temperature measurements for the particular vehicle;

wherein at least one of the first present heat exchanger fouling indication and the second present heat exchanger fouling indication exceeding a heat exchanger fouling threshold causes maintenance to be scheduled for at least one of a first heat exchanger and a second heat exchanger of the particular vehicle.

16. The system of claim 15, wherein the computer processor is further operable to calculate a rate of change of a fouling indication for the particular vehicle among the fleet of vehicles based on the calculated regression model and a sequence of past instances of temperature measurements for the particular vehicle.

17. The system of claim 16, wherein the computer processor:

predicts an amount of time before at least one of the first present heat exchanger fouling indication and the second present heat exchanger fouling indication exceeds a heat exchanger fouling threshold based on a most-recent set of temperature measurements received from the particular vehicle applied to the regression model and the calculated rate of change; and

outputs the predicted amount of time to the particular vehicle.

18. The system of claim 15, wherein the data input is further operable to receive the current instance of temperature measurements for the particular vehicle;

and

wherein the computer processor is further operable to;

calculate the first present heat exchanger fouling indication and the second present heat exchanger fouling indication based on the received current instance of temperature measurements and the generated heat exchanger fouling model; and

output, via the data output, the calculated first present heat exchanger fouling indication and the second present heat exchanger fouling indication to the particular vehicle.

19. The system of claim 18, wherein the computer processor is further operable to schedule maintenance for the at least one of the first heat exchanger and the second heat exchanger of the particular vehicle upon the at least one of the first present heat exchanger fouling indication and the second present heat exchanger fouling indication exceeding the heat exchanger fouling threshold, wherein the scheduled maintenance is performed for the vehicle;

wherein the second heat exchanger is downstream of, and in serial fluid communication with, the first heat exchanger;

wherein the gas turbine engine is further operable to provide pressurized air for use in a passenger cabin, wherein the particular vehicle comprises one of;

(i) a vehicle of the fleet of vehicles and (ii) a vehicle other than the fleet of vehicles.

20. The system of claim 18, wherein the first present heat exchanger fouling indication and the second present heat exchanger fouling indication are expressed as percentages.

22. The system of claim 15, wherein the scheduled maintenance is performed for the particular vehicle.