Classifying a work machine operation

• US 7,487,066 B2
• Filed: 09/16/2005
• Issued: 02/03/2009
• Est. Priority Date: 04/28/2005
• Status: Active Grant

• Current Assignee: Caterpillar Incorporated
• Original Assignee: Caterpillar Incorporated
• Inventors: Baskett, Jeffrey Dale, Patel, Nitin R., Sundermeyer, Jeffry Neil, Allgaier, Ryan Paul, Clarke, Burton Roland, Vik, Timothy Allen, Dunn, Daniel Kimsey III, Sit, Don, Kyuba, Hiroko, Truax, Byron Edwin
• Primary Examiner(s): Lau; Tung S
• Application Number:
  US11/227,155
• Publication Number:
  US 20060243180A1
• Time in Patent Office: 1,236 Days
• Field of Search: 702/182
• US Class Current: 702/182
• CPC Class Codes: A01B 63/00 Lifting or adjusting devices or arrangements for agricultural machines or implements lifting mechanisms for the cutter-bar of a mower A01D34/24; adjusting devices for the cutter-bar of a mower A01D34/28; constructional features of lifting devices per se B66F
• CPC Class Codes: E02F 9/0841 Articulated frame, i.e. having at least one pivot point between two travelling gear units tractor-trailer combinations B62D53/00
• CPC Class Codes: E02F 9/267 Diagnosing or detecting failure of vehicles
• CPC Class Codes: G06F 11/2263 using neural networks

Abstract

A method for analyzing the use of a work machine is disclosed. In one embodiment, the method may include providing a computer with a neural network on the work machine. Further, the method may include inputting data to the computer, at least a portion of the data associated with a load experienced by one of the components of the work machine. The neural network, when executed by the computer may then classify a current operation of the work machine into one of a plurality of types of operations.

Claims

1. A method of analyzing the use of a work machine comprising:

providing a computer with a neural network on the machine;

inputting data to the computer, at least a portion of the data associated with a load experienced by one of the components of the machine; and

using the neural network to classify a current physical operation of the machine as one of a plurality of types of operations performable by the machine, based on the inputted data associated with the load experienced by one of the components of the machine.

2. The method of claim 1, further including:

generating a histogram of the different types of physical operations of the machine.

3. The method of claim 1, including:

determining fatigue of the one component over a first time period of operation of the machine for at least one of the classified types of operations.

4. The method of claim 1, wherein using the neural network to classify the current physical operation is performed in real-time by the computer during operation of the machine.

5. The method of claim 1, wherein using the neural network to classifying the physical operation is performed by logic-based software executed by the computer.

6. The method of claim 1, wherein using the neural network to classify the physical operation includes:

comparing the inputted data to stored data corresponding to physical operations performable by the machine; and

identifying one of the different types of operations as the current physical operation of the machine based on the comparison.

7. The method of claim 1, wherein inputting the data includes:

collecting strain data associated with the component from one or more strain gauges.

8. The method of claim 7, including:

communicating the collected strain data to the computer over a wireless network.

9. A system for classifying a physical operation performed by a machine, comprising:

sensors disposed about the machine, each of the sensors being configured to detect one or more parameters associated with the machine;

a first memory storing classification data associated with different types of physical operations performable by the machine; and

a processor including a neural network;

wherein the processor is configured to receive a signal indicative of at least one of the detected parameters and use the neural network to classify a current physical operation of the machine as at least one of the different types of operations performable by the machine, based upon the received signal and the stored classification data.

10. The system of claim 9, including:

a second memory, included on the machine, that stores data reflecting the classified operation.

11. The system of claim 9, wherein the processor is configured to identify a level of fatigue associated with the machine during the classified operation over a selected time period.

12. The system of claim 9, further including an off-board system configured to receive data associated with the classified operation.

13. The system of claim 9, wherein the processor is configured to classify the physical operation in real-time.

14. The system of claim 9, wherein the neural network includes stored parameters for each of the plurality of different types of operations.

15. The system of claim 14, wherein the processor is configured to:

compare the one or more detected parameters to stored parameters and determine which physical operation includes data that most closely matches the detected parameters; and

classify the physical operation based on the determination.

16. A method of analyzing the use of a machine comprising:

providing a computer with a neural network on the machine;

inputting data to the computer, at least a portion of the data associated with a load experienced by one of the components of the machine;

using the neural network to classify a current physical operation of the machine as one of a plurality of types of operations performable by the machine, based on the inputted data associated with the load experienced by one of the components of the machine; and

generating a histogram of the different types of physical operations performed by the machine.

17. The method of claim 16, wherein using the neural network to classify the current physical operation is performed in real-time by the computer during operation of the machine.

18. The method of claim 16, wherein inputting the data includes:

collecting strain data associated with the component from one or more strain gauges; and

communicating the collected strain data to the computer over a wireless network.

19. The method of claim 17, wherein using the neural network to classify the physical operation includes:

comparing the inputted data to stored data corresponding to physical operations performable by the machine; and

identifying one of the different types of operations as the current physical operation of the machine based on the comparison.