ADAPTIVE METHOD AND SYSTEM OF MONITORING SIGNALS FOR DETECTING ANOMALIES

Abstract

There is described a method for detecting anomalies in a set of signals that normally exhibit analogous behaviour. The method comprises: acquiring a set of signals from at least one measurement source; designating a reference signal Sref from the acquired set of signals; computing a deviation Devi between the reference signal Sref at a reference point in time tref(Sref(tref)) and at least one of the acquired signals Si over time; and generating a deviation alarm trigger when the deviation meets a given deviation criteria.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

These and other features and aspects will become better understood with regard to the following description and accompanying drawings wherein:

FIG. 1 is a typical prior art membrane cell used in the chlor-alkali industry;

FIG. 2 is a flow chart of a method for detecting an anomaly using a reference signal according to a first embodiment;

FIG. 3 is a flow chart of a method for detecting an anomaly using a reference signal and a confirmation signal according to a second embodiment;

FIG. 4 is a flow chart of a method for detecting an anomaly according to a third embodiment;

FIG. 5 is a block diagram of a system for detecting an anomaly according to an embodiment; and

FIG. 6 is a screen shot of data signals that are anticipated to have similar behavioral characteristics, with a deviating signal.

Claims

1. A method for detecting anomalies in a set of signals that normally exhibit analogous behaviour, the method comprising: acquiring a set of signals from at least one measurement source;designating a reference signal Sref from the acquired set of signals;computing a deviation Devi between the reference signal Sref at a reference point in time tref(Sref(tref)) and at least one of the acquired signals Si over time; andgenerating a deviation alarm trigger when the deviation meets a given deviation criteria.

2. The method according to claim 1, wherein the deviation criteria comprises at least one of: a reached or exceeded threshold value for a specified duration of time; changes in value distribution, the value distribution comprising at least one of mean and standard deviation; and evolution of the slope/curve fitting parameters of the Devi over time.

3. The method according to claim 1, wherein a signal from the set of signals is defined as a series of discrete values transmitted over time from at least one measurement source, the values being representative of the process parameters over time.

4. The method according to claim 3, wherein said process parameters comprise at least one of voltage, temperature, and pressure.

5. The method according to claim 1, further comprising calculating a difference Deltai between at least one of the acquired signals Si at a reference point in time tref(Si(tref)) and the reference signal Sref at a reference point in time tref(Sref(tref)), hence expressed as follows: Deltai=Si(tref)−Sref(tref).

6. The method according to claim 5, further comprising comparing the difference Deltai to a given delta criteria; and generating a delta alarm trigger when the given delta criteria is met.

7. The method according to claim 6, wherein when the given delta criteria is not met, performing the deviation computing (Devi) at time t by subtracting, from at least one of the acquired signals at time t(Si(t)), the reference signal Sref at time t (Sref(t)) and the difference Deltai, hence expressed as follows: Devi=(Si(t)−Sref(t)−Deltai)

8. The method according to claim 7, wherein the deviation criteria comprises at least one of: a reached or exceeded threshold value for a specified duration of time; changes in value distribution, the value distribution comprising at least one of mean and standard deviation; and evolution of the slope/curve fitting parameters of the Devi over time.

9. The method according to claim 1, further comprising designating a confirmation signal Sconf from the acquired set of signals for monitoring the behaviour of the reference signal Sref.

10. The method according to claim 9, further comprising calculating a reference deviation RDevi as follows: RDevi=(Sconf(t)−Sref(t))−(Sconf(tref)−Sref(tref));

11. The method according to claim 10, further comprising comparing the reference deviation RDevi to a given confirmation criteria; and generating a confirmation alarm trigger when the given confirmation criteria is met.

12. The method according to claim 11, further comprising activating and deactivating at least one of the deviation alarm trigger, delta alarm trigger and confirmation alarm trigger according to at least one of an external action, and a process operation condition.

13. The method according to claim 11, further comprising manually or automatically setting at least one of the deviation alarm criteria, delta alarm criteria and confirmation alarm criteria based on the shape of measured signals.

14. The method according to claim 11, further comprising using at least one of the deviation alarm trigger, delta alarm trigger and confirmation alarm trigger to perform at least one of shutting down an industrial process, characterizing the performance of an electrolyzer having elementary cells in the context of an industrial electrolysis process, and detecting pinholes in cell membranes for an electrolysis chlor-alkali process.

15. The method according to claim 10, further comprising automatically determining at least one of the reference or confirmation signals by choosing the best centered values in the set of signals at the reference point in time tref.

16. The method according to claim 10, further comprising setting the reference point in time tref according to at least one of an external action, and a process operation condition.

17. The method according to claim 1, further comprising scaling the signals from the set of signals prior to their acquisition.

18. The method according to claim 1, further comprising integrating the steps of acquiring a set of signals, designating a reference signal, computing a deviation, and generating a deviation alarm trigger in an autonomous voltage measurement system.

19. A system for detecting anomalies in a set of signals that normally exhibit analogous behaviour, the system comprising: a signal measurement module for acquiring a set of signals from at least one measurement source and designating a reference signal Sref from the acquired set of signals;a deviation computing module for computing a deviation Devi between the reference signal Sref at a reference point in time tref(Sref(tref)) and at least one of the acquired signals Si over time; andan anomaly detector for generating a deviation alarm trigger when the deviation meets a given deviation criteria.

20. The system according to claim 20, further comprising a delta computing module for calculating a difference Deltai between at least one of the acquired signals Si at a reference point in time tref(Si(tref)) and the reference signal Sref at a reference point in time tref(Sref(tref)), hence expressed as follows: Deltai=Si(tref)−Sref(tref).

21. The system according to claim 21, wherein said anomaly detector is further for comparing the difference Deltai to a given delta criteria; and generating a delta alarm trigger when the given delta criteria is met.

22. The system according to claim 22, wherein said anomaly detector is first for verifying if the given delta criteria is met, if not, said anomaly then performs the deviation computing (Devi) at time t by subtracting, from at least one of the acquired signals at time t(Si(t)), the reference signal Sref at time t(Sref(t)) and the difference Deltai, hence expressed as follows: Devi=(Si(t)−Sref(t)−Deltai)