Apparatus and method for monitoring a running process

• US 7,013,237 B2
• Filed: 06/26/2003
• Issued: 03/14/2006
• Est. Priority Date: 06/27/2002
• Status: Expired due to Fees

• Current Assignee: Fraunhofer Gesellschaft Zur Foerderung Der Angewandten Forsching E.V.
• Original Assignee: Fraunhofer Gesellschaft Zur Foerderung Der Angewandten Forsching E.V.
• Inventors: Endres, Hanns-Erik
• Primary Examiner(s): Nghiem, Michael
• Assistant Examiner(s): LE, JOHN H
• Application Number:
  US10/606,912
• Publication Number:
  US 20040059516A1
• Time in Patent Office: 992 Days
• Field of Search: 702 81- 85, 702/116, 702/182, 324/633, 324/637, 324/639, 324/207.25, 376/216, 376/259, 700108-111, 700/121, 700/4.7, 700/83, 438/14
• US Class Current: 702/182
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks

Abstract

A method for monitoring a process consists of repeatedly acquiring at least two different pieces of information of the process, performing a component transformation using the acquired information to calculate a main component measurement vector in a main component space, calculating a process indicator quantity by using the calculated main component measurement vector with one or several main component measurement vectors of a previous acquisition, and detecting a process end by using the process indicator quantity. Apparatus for monitoring a process is also disclosed.

Claims

1. Method for monitoring a process, comprising:

repeatedly acquiring at least two different pieces of information of the process;

performing a main component transformation due to the acquired information without using information acquired prior to the monitoring method, for calculating a main component measurement vector in a main component space;

calculating a process indicator quantity by using the calculated main component measurement vector and one or several previously calculated main component measurement vectors; and

detecting an end of the process by using the process indicator quantity.

2. Method according to claim 1, where the process is a discontinually running process.

3. Method according to claim 1, where the step of repeatedly acquiring comprises acquiring of at least two different measurement data of the process.

4. Method according to claim 1, where the step of performing a main component transformation further comprises the step of selecting the acquired information to perform a main component transformation merely for the selected information.

5. Method according to claim 1, where the step of performing a main component transformation further comprises averaging the acquired information for generating average value information, which is used in the main component transformation.

6. Method according to claim 1, where the step of calculating a process indicator quantity comprises calculating the quantity and/or direction of a difference vector, which is formed from the calculated main component measurement vector and a previous main component measurement vector.

7. Method according to claim 1, where in the step of detecting an end of the process, the end of the process is indicated when a predetermined number of process indicator quantities are in a predetermined indicator area.

8. Method according to claim 1, further comprising the step of intervening into a running process when an end of the running process is detected in the step of detecting a process end.

9. Apparatus for monitoring a running process, comprising:

acquisition means for repeatedly acquiring at least two different pieces of information of the process;

means for performing a main component transformation due to the acquired information, whereby a main component measurement vector is calculated in a main component space; and

evaluation means for calculating a process indicator quantity by using the calculated main component measurement vector and one or several previously calculated main component measurement vectors, wherein the evaluation means is further formed to detect an end of the process by using the process indicator quantity.

10. Apparatus according to claim 9, where the acquisition means comprises one or several sensors, which are selected from the group comprising an optical sensor, a capacitive sensor, an amperometrical sensor, a temperature sensor and a chemical sensor.

11. Apparatus according to claim 9, where the means for performing a main component transformation is further formed to make a selection from the acquired information of the process, so that a main component transformation is performed merely for the selected information of the process.

12. Apparatus according to claim 9, where the means for performing a main component transformation is further formed to perform a main component transformation by using average value information, which is formed by averaging the acquired information of the process.

13. Apparatus according to claim 9, where the evaluation means is formed to calculate the process indicator quantity by using a spacing of the calculated main component measurement vector from a previous main component measurement vector and/or a direction of the difference vector from the calculated vector and a previous vector.

14. Apparatus according to claim 13, where the process indicator quantity is calculated from the absolute values of the spacing and the direction of the difference vector.

15. Apparatus according to claim 9, where the acquisition means is formed to acquire transient measurement signals.

16. Apparatus according to claim 15, where the transient measurement signals comprise a spectrum or a temperature signal.

17. Apparatus according to claim 9, where the acquisition means is formed to acquire time series information.

18. Apparatus according to claim 9, where the acquisition means is formed to acquire at least two different parameters of different dimensional characteristics, and wherein the means for performing a main component transformation is performed to perform a multi-stage main component transformation due to the at least two different parameters of different dimensional characteristics.

19. Apparatus according to claim 9, which further comprises a control means to intervene in the running process when an end of the process is detected by the evaluation means.