Classification of deviations in a process

• US 6,999,898 B2
• Filed: 12/31/2003
• Issued: 02/14/2006
• Est. Priority Date: 01/07/2003
• Status: Active Grant

• Current Assignee: Hach Company (Veralto Corp.)
• Original Assignee: Hach Company (Veralto Corp.)
• Inventors: Wang, Ziyi, Kroll, Dan J., King, Karl L.
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): SUAREZ, FELIX E
• Application Number:
  US10/751,168
• Publication Number:
  US 20040158432A1
• Time in Patent Office: 776 Days
• Field of Search: 702/27, 702/38, 702/58, 702/60, 702/182, 702/183, 702/187, 702/188, 324/693, 700/28, 700/108, 704/209, 205/742, 205/743
• US Class Current: 702/183
• CPC Class Codes: G01N 33/18 Water
• CPC Class Codes: G05B 23/0281 Quantitative, e.g. mathematical distance; Clustering; Neural networks; Statistical analysis

Abstract

A process analysis system includes sensors and a processing system. The sensors monitor the process to generate sensor signals. The processing system processes the sensor signals to detect a deviation from a baseline for the process. The processing system generates a process vector for the deviation in response to detecting the deviation. The processing system compares the process vector to a plurality of library vectors to classify the deviation. In some examples, the process comprises a system that supplies water.

Claims

1. A method of operating a process analysis system to analyze a process, the method comprising:

in a plurality of sensors, monitoring the process to generate sensor signals that indicate sensor data for the process;

in a processing system, processing the sensor signals to determine a deviation between the sensor data and a baseline for the process;

in the processing system, comparing the deviation to a threshold and generating a trigger if the deviation exceeds the threshold;

in the processing system, generating a process vector representing the deviation; and

in the processing system, in response to the trigger, comparing the process vector representing the deviation to a plurality of library vectors representing abnormal operating conditions to classify the deviation.

2. The method of claim 1 wherein the process comprises a system that supplies water.

3. The method of claim 2 wherein the classified deviation comprises a contaminant in the water.

4. The method of claim 2 wherein the sensor signals indicate pH, conductivity, turbidity, chlorine, and total organic carbon of the water.

5. The method of claim 2 further comprising signaling a control system to operate a valve in response to classifying the deviation as a contaminant in the water.

6. The method of claim 2 further comprising signaling a control system to add a marker to the water in response to classifying the deviation as a contaminant in the water.

7. The method of claim 6 wherein the marker comprises a colorant.

8. The method of claim 2 further comprising signaling a control system to perform a treatment on the water in response to classifying the deviation as a contaminant in the water.

9. The method of claim 8 wherein the treatment comprises adding a disinfectant to the water.

10. The method of claim 8 wherein the treatment comprises adding chlorine to the water.

11. The method of claim 8 wherein the treatment comprises exposing the water to ultraviolet radiation.

12. The method of claim 1 wherein processing the sensor signals to detect the deviation and comparing the deviation to the threshold comprises processing the sensor signals to produce a single variable and comparing the single variable to the threshold.

13. The method of claim 1 wherein generating the process vector representing the deviation comprises generating a unit vector.

14. The method of claim 1 wherein comparing the process vector representing the deviation to the library vectors representing the abnormal operating conditions comprises comparing an angle between the process vector and one of the library vectors to another threshold.

15. The method of claim 1 wherein classifying the deviation comprises identifying one of the abnormal operating conditions represented by one of the library vectors that matches the process vector.

16. The method of claim 1 further comprising, in response to an unknown classification, storing the process vector as a new one of the library vectors and associating a new abnormal operating condition with the new library vector.

17. A process analysis system comprising:

a plurality of sensors configured to monitor a process to generate sensor signals that indicate sensor data for the process; and

a processing system configured to process the sensor signals to determine a deviation between the sensor data indicated by the sensor signals and a baseline for the process, compare the deviation to a threshold and generate a trigger if the deviation exceeds the threshold, generate a process vector representing the deviation, and in response to the trigger, compare the process vector representing the deviation to a plurality of library vectors representing abnormal operating conditions to classify the deviation.

18. The process analysis system of claim 17 wherein the processing system is configured to process the sensor signals to produce a single variable and compare the single variable to the threshold to detect the deviation and compare the deviation to the threshold.

19. The process analysis system of claim 17 wherein the process vector representing the deviation comprises a unit vector.

20. The process analysis system of claim 17 wherein the processing system is configured to compare an angle between the process vector representing the deviation and one of the library vectors representing one of the abnormal operating conditions to another threshold.

21. The process analysis system of claim 17 wherein the processing system is configured to identify one of the abnormal operating conditions represented by one of the library vectors that matches the process vector to classify the deviation.

22. The process analysis system of claim 17 wherein the processing system is configured to store the process vector representing the deviation as a new one of the library vectors and associate a new abnormal operating condition with the new library vector in response to an unknown classification.

23. The process analysis system of claim 17 wherein the process comprises a system that supplies water.

24. The process analysis system of claim 23 wherein the classified deviation comprises a containment in the water.

25. The process analysis system of claim 24 wherein the processing system is configured to signal a control system to perform a treatment on the water in response to classifying the deviation as a contaminant in the water.

26. The process analysis system of claim 23 wherein the sensor signals indicate pH, conductivity, turbidity, chlorine, and total organic carbon of the water.

27. The process analysis system of claim 23 wherein the processing system is configured to signal a control system to operate a valve in response to classifying the deviation as a contaminant in the water.

28. The process analysis system of claim 23 wherein the processing system is configured to signal the control system to add a marker to the water in response to classifying the deviation as a contaminant in the water.

29. The process analysis system of claim 28 wherein the marker comprises a colorant.

30. The process analysis system of claim 25 wherein the treatment comprises adding a disinfectant to the water.

31. The process analysis system of claim 25 wherein the treatment comprises adding chlorine to the water.

32. The process analysis system of claim 25 wherein the treatment comprises exposing the water to ultraviolet radiation.