Tool condition monitoring system

• US 5,587,931 A
• Filed: 10/20/1995
• Issued: 12/24/1996
• Est. Priority Date: 10/20/1995
• Status: Expired due to Fees

• Current Assignee: Tri-Way Manufacturing Technologies
• Original Assignee: Tri-Way Machine Limited
• Inventors: Wu, Ya, Jones, Joel W.
• Primary Examiner(s): Ramirez, Ellis B.
• Assistant Examiner(s): PIPALA, EDWARD J
• Application Number:
  US08/546,361
• Time in Patent Office: 431 Days
• Field of Search: 364/551.02, 364/511, 364/474.12, 364/474.15, 364/474.16, 364/474.17, 364/474.19, 73/660, 73/104, 318/476, 340/660, 340/664, 340/679, 340/680
• US Class Current: 702/34
• CPC Class Codes: G05B 2219/37256 Wear, tool wear
• CPC Class Codes: G05B 2219/37348 Power, wattmeter voltage times current
• CPC Class Codes: G05B 2219/37457 On machine, on workpiece
• CPC Class Codes: G05B 2219/37525 Mean, average values, statistical derived values
• CPC Class Codes: G05B 2219/50197 Signature analysis, store working conditions, compare with actual
• CPC Class Codes: G05B 2219/50203 Tool, monitor condition tool
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
• CPC Class Codes: G05B 23/0235 based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold

Abstract

A tool monitoring system monitors the condition of an electrically powered tool performing a cyclical operation. The tool monitoring system operates generally in two modes: learning mode and monitor mode. In learning mode, the tool monitoring system gathers statistical data on the power consumption of tools of the selected tool type during learning cycles. A power threshold is generated based upon the statistical data. The tool monitoring system then counts the number of crossings by each of the learning cycles of the power threshold and generates statistical data regarding the number of crossings. Preferably, the mathematical operation of wavelet packet transform is used to calculate the power threshold. Feature wavelet packets of the power consumption signal of the tool are calculated. The power consumption signal is then reconstructed from the feature wavelet packets and used to determine the power threshold. In monitor mode, the tool monitoring system counts the number of crossings of the power threshold by the power consumption signal of a tool in operation. The tool monitoring system identifies a worn tool when the number of crossings increases to a certain number relative to the crossings by the learning cycles.

Claims

1. A method for monitoring the condition of an electrically powered tool of a selected tool type performing a cyclical task comprising the steps of:

a. using a tool to perform a cyclical task and measuring a power consumption signal of said tool as said tool performs said cyclical task;

b. decomposing said power consumption signal of said tool into components by converting said power consumption signal into time segments, each time segment having a digital value equivalent to the analog value of said signal at said time, and decomposing said time segments of said power consumption signal into different frequency components using a mathematical algorithm;

c. selecting feature components of said power consumption signal;

d. using said selected components of said power consumption signal to set a power threshold in a learning cycle;

e. using a tool to perform a cyclic task and monitoring its power consumption signal; and

f. identifying a worn condition of the tool of step e. based upon a comparison of said power consumption signal with said predetermined power threshold of step d.

2. The method for monitoring of claim 1, wherein the selected components of said step c. are reconstructed prior to said step d.

3. The method for monitoring of claim 1, wherein said step b. includes the step of applying said power consumption signal of said tool to a plurality of filters.

4. The method for monitoring of claim 1, wherein the decomposition of the signal includes the step of calculating wavelet packet transforms of said power consumption signal.

5. The method for monitoring of claim 1, wherein said power consumption signal of step e. is not decomposed prior to step f.

6. The method for monitoring of claim 1, wherein said step f. includes the steps of counting the number of crossings of said power threshold by said power signal, and comparing the number of crossings to an expected number of crossings.

7. A method for monitoring the condition of an electric motor powered tool of a selected tool type comprising the steps of:

a. using a tool to perform a cyclical task, and measuring a power consumption signal of said tool as said tool performs a cyclical task;

b. generating a power threshold based upon said power consumption signal of said step a.;

c. using a tool to perform a cyclical task, and measuring the power consumption signal of said tool said tool performs said cyclical task;

d. comparing said power consumption signal of step c. to said power threshold by counting the number of crossings of said power threshold by said power consumption signal of said tool in step a., and counting the number of crossings of said power threshold by said power consumption signal of said tool in step c; and

e. signalling a worn condition of said tool of step c when said power consumption signal of step c. reaches a predetermined level relative to said power threshold, said predetermined level being defined by when said crossings of said tool of step c. exceed said crossings of said tool of step a. by a predetermined amount.

8. The method for monitoring of claim 7, wherein both upper and lower thresholds are generated in said step b.

9. The method for monitoring of claim 8, wherein extreme thresholds are also set outside of said upper and lower thresholds, and signalling a failure should said power consumption signal of step c. exceed said extreme thresholds once.

10. The method for monitoring of claim 7, wherein said tool signals of said step b. are generated using a plurality of sample cycles.

11. The method of monitoring of claim 7, wherein said tools of steps a. and c. are the same tool.

12. The method for monitoring of claim 7, wherein said step b. includes the steps of:

converting said power consumption signal of said tool into time segments, each time segment having a digital value equivalent to the analog value of said signal at said time;

calculating wavelet packets of said power consumption signal of said tool; and

reconstructing said power consumption signal of said tool by calculating inverse wavelet packet transform based on selected wavelet packets.

13. A method for monitoring the condition of an electric motor powered tool of a selected tool type comprising the steps of:

a. selecting a power threshold;

b. using a tool to perform a cyclical task, and measuring a power consumption signal of said tool as said tool performs said cyclical task;

c. counting the number of crossings of said power threshold by said power consumption signal of a learning cycle tool of said selected tool type, counting the number of crossings of said power threshold by said power consumption signal of said tool; and

d. signalling a worn condition of said tool when said crossings of said tool exceed said crossings of said learning cycle tool by a predetermined amount in a predetermined time frame.

14. The method for monitoring of claim 13, wherein said step a. includes the steps of:

measuring a power consumption signal of said learning cycle tool as it performs a cyclical task; and

generating a power threshold based upon said power consumption signal of said learning cycle tool.

15. The method for monitoring of claim 14, wherein said tools of steps a. and b. are the same tool.

16. A method for monitoring the condition of an electric motor powered tool of a selected tool type comprising the steps of:

a. using a first tool to perform a plurality of cyclical tasks, and measuring a power consumption signal of said first tool as said first tool performs said plurality of cycles of cyclical task;

b. decomposing said power consumption signal of each said cycle of said first tool into components;

c. selecting feature components of said components of said power consumption signal;

d. reconstructing said power consumption signal from said feature components;

e. generating a power threshold based upon said reconstructed power consumption signal of said first tool, said power threshold including an upper limit for said power consumption signal;

f. counting the number of crossings of said power threshold by said power consumption signal of said first tool while performing a plurality of cycles of said cyclical task;

g. using a second tool to perform said cyclical task, and measuring the power consumption signal of said second tool as said second tool performs said cyclical task;

h. counting the number of crossings of said power threshold by said power consumption signal of said second tool;

i. comparing the number of crossings of said power threshold by said second tool with the number of crossings of said power threshold by said first tool; and

j. signalling a worn condition of said second tool when the number of crossings of said power threshold by said second tool exceeds the number of crossings of said power threshold of said first tool by a predetermined amount in a predetermined time frame.

17. The method for monitoring the condition of an electrically powered tool of claim 16, wherein said steps b-d include the steps of:

calculating feature wavelet packets of said power consumption signal of said first tool; and

reconstructing said power consumption signal from the inverse wavelet packet transform of the feature wavelet packets.

18. The method for monitoring the condition of an electrical power tool of claim 16, wherein said first and second tools are the same tool.