Monitoring

• US 5,309,379 A
• Filed: 03/11/1991
• Issued: 05/03/1994
• Est. Priority Date: 02/07/1989
• Status: Expired due to Term

• Current Assignee: QED Intellectual Property Services Limited
• Original Assignee: Smiths Industries Public Limited Company
• Inventors: Witcomb, Roger C., Rawlings, Keith C.
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): Ramirez, Ellis B.
• Application Number:
  US07/663,855
• Time in Patent Office: 1,149 Days
• Field of Search: 364/578, 364/551.01, 364/508, 364/431.02, 364/424.06, 364/148, 364/149, 364/151, 395/913, 395/914, 395/920
• US Class Current: 703/8
• CPC Class Codes: F02C 9/00 Controlling gas-turbine plants; Controlling fuel supply in air- breathing jet-propulsion plants controlling air intakes F02C7/057; controlling turbines F01D; controlling compressors F04D27/00; controlling in general G05
• CPC Class Codes: F05D 2270/09 to cope with emergencies
• CPC Class Codes: F05D 2270/10 to cope with, or avoid, compressor flow instabilities
• CPC Class Codes: F05D 2270/11 to prolong engine life
• CPC Class Codes: F05D 2270/44 active, predictive, or anticipative
• 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: G07C 3/00 Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
• CPC Class Codes: Y02T 50/60 Efficient propulsion technologies, e.g. for aircraft

Abstract

The operation of an engine or other assembly is monitored by detecting the sound it makes and deriving from this, in conjunction with its speed, temperature and control settings, a multi-dimension condition vector. The vector changes during operation of the engine and is supplied to a processor which continuously builds a model of the engine at different stages of operation such as by a sequence-learning neural net or by hidden markov model when an atypical condition vector is received, the processor generates an alert signal to a display.

Claims

1. A method of monitoring the condition of a sound-generating assembly characterized in that the method includes the steps of deriving signals representative of the condition of the assembly, continuously building a model of the assembly at different stages of use through the life of the assembly such that the model is continuously modified by any ageing of the assembly, and detecting the amount of deviation of the current condition of the assembly from the aged model of the assembly, said step of deriving signals including using a microphone to monitor the acoustic spectrum of the assembly transmitted through air such that the model includes information about the acoustic spectrum of the assembly at different stages of use through the life of the assembly.

2. A method of monitoring the condition of an engine characterized in that the method includes the steps of deriving signals representative of the condition of the engine, continuously building a model of the engine at different stages of use through the life of the engine such that the model is continuously modified by any ageing of the engine, and detecting the amount of deviation of the current condition of the engine from the aged model of the engine.

9. A method according to claim 2 wherein the engine is a gas-turbine engine.

10. A method according to claim 2 or 5 wherein the step of deriving signals representative of the condition of the assembly includes monitoring the speed of the engine.

3. A method of monitoring the condition of an assembly characterized in that the method includes the steps of deriving signals representative of the condition of the assembly, continuously building a model of the assembly at different stages of use through the life of the assembly such that the model is continuously modified by any ageing of the assembly, establishing a multi-dimension vector in respect of the condition of the assembly, and detecting the amount of deviation of the current condition of the assembly from the aged model of the assembly.

7. A method of monitoring the condition of a sound-generating assembly according to any one of claims 3, 4, 5, or 6 wherein the step of deriving signals representative of the condition of the assembly includes monitoring the acoustic spectrum of the assembly such that the model includes information about the acoustic spectrum of the assembly at different stages of use through the life of the assembly.

8. A method according to claim 7 wherein the acoustic spectrum is the acoustic spectrum transmitted through air, and the acoustic spectrum is monitored with a microphone.

4. A method of monitoring the condition of an assembly characterized in that the method includes the steps of deriving signals representative of the condition of the assembly, using a sequence-learning neural net to continuously build a model of the assembly at different stages of use through the life of the assembly such that the model is continuously modified by any ageing of the assembly, and detecting the amount of deviation of the current condition of the assembly from the aged model of the assembly.

5. A method of monitoring the condition of an assembly characterized in that the method includes the steps of deriving signals representative of the condition of the assembly, using a hidden Markov model to continuously build a model of the assembly at different stages of use through the life of the assembly such that the model is continuously modified by any ageing of the assembly, and detecting the amount of deviation of the current condition of the assembly from the aged model of the assembly.

6. A method of monitoring the condition of an assembly characterized in that the method includes the steps of building a model of a first assembly, utilizing said model on a second assembly of the same kind as said first assembly, deriving signals representative of the condition of the second assembly, modifying the model of the assembly at different stages of use through the life of the second assembly such that the model is continuously modified by any ageing of the second assembly, and detecting the amount of deviation of the current condition of the second assembly from the aged model of the assembly.