Detection of change in physicochemical composition of a liquid

Patent metadata

Field	Value
Publication number	US-12498359-B2
Application number	US-202117927320-A
Country	US
Kind code	B2
Filing date	May 24, 2021
Priority date	May 25, 2020
Publication date	Dec 16, 2025
Grant date	Dec 16, 2025

Abstract

A field of measuring and monitoring liquids is provided. More specifically, the improved detection of a change in physicochemical composition of a liquid is provided, based on a set of measurements from sensors collocated in the liquid, and applying data partitioning to the set of measurements. If at least two measurement clusters are identified, an abnormal change in the composition of the liquid is detected.

First claim

The invention claimed is:

1 . A computer-implemented method for detecting a change in physicochemical composition of a liquid, comprising:
receiving time series of measurements from a plurality of collocated sensors for sensing physicochemical variables of said liquid by means of at least one communication link; applying data partitioning to the time series of measurements; checking for a presence of at least two measurement clusters; if at least two measurement clusters are present, detecting the presence of an abnormal change in the physicochemical composition of the liquid; otherwise, detecting an absence of an abnormal change in the physicochemical composition of the liquid; the above-mentioned steps being performed with a processor; and if an abnormal change is detected, undertaking additional analysis and/or corrective actions regarding this change.

2 . The computer-implemented method as claimed in claim 1 , comprising smoothing the measurements.

3 . The computer-implemented method as claimed in claim 1 , comprising principal component analysis of the measurements.

4 . The computer-implemented method as claimed in claim 3 , wherein the principal component analysis comprises multiplying a measurement matrix by a centering-reduction matrix, and then by a transformation matrix, and wherein at least one of the centering-reduction and transformation matrices was computed in a training phase based on measurements from said plurality of sensors.

5 . The computer-implemented method as claimed in claim 1 , wherein applying the data partitioning comprises applying a density-based partition.

6 . The computer-implemented method as claimed in claim 1 , wherein applying the data partitioning comprises applying a first data partitioning method and, if the first method does not manage to partition the measurements into at least two clusters, applying a second data partitioning method different from the first one.

7 . The computer-implemented method as claimed in claim 6 , wherein the first method is a local density-based data partitioning method, and the second method is centroid-based.

8 . The computer-implemented method as claimed in claim 1 , comprising, if the presence of an abnormal change is detected, categorizing the abnormal change before undertaking the additional analysis and/or corrective actions.

9 . The computer-implemented method as claimed in claim 8 , wherein the category of the abnormal change is determined on the basis of a variation direction of the values of the measurements between each of the at least two measurement clusters.

10 . The computer-implemented method as claimed in claim 1 , comprising, if the presence of an abnormal change is detected, locating the abnormal change before undertaking the additional analysis and/or corrective actions.

11 . The computer-implemented method as claimed in claim 10 , wherein the location of the abnormal change is determined on the basis of a location of at least one measurement positioned, by the data partitioning, at the limit between two clusters.

12 . The computer-implemented method as claimed in claim 11 , wherein the plurality of sensors are located in a mobile probe, the measurements are timestamped, and said location of a measurement is obtained on the basis of the path taken by the probe in the liquid and the timestamp of the measurement.

13 . The computer-implemented method as claimed in claim 12 , wherein the location of the measurement is obtained through linear extrapolation, over the journey of the probe, of a ratio between the difference between the timestamp of the measurement and the starting time, and the journey time.

14 . The computer-implemented method as claimed in claim 1 , wherein the plurality of sensors comprises at least one sensor chosen from a group comprising:
a conductivity sensor; a temperature sensor; a pH sensor; and an oxidoreduction potential sensor.

15 . The computer-implemented method as claimed in claim 14 , wherein the plurality of sensors comprises:
a conductivity sensor; and a temperature sensor.

16 . The computer-implemented method as claimed in claim 1 , wherein the plurality of sensors are collocated in a mobile probe.

17 . The computer-implemented method as claimed in claim 16 , wherein the mobile probe is a ball float.

18 . A device for detecting a change in physicochemical composition of a liquid, comprising:
at least one communication link to a plurality of collocated sensors for sensing physicochemical variables of said liquid; and a processor configured so as to execute the respective steps of a method as claimed in claim 1 .

19 . A probe for detecting a change in physicochemical composition of a liquid, said probe comprising:
a plurality of sensors for sensing physicochemical variables of said liquid; and at least one communication link to a device as claimed in claim 18 .

20 . The probe as claimed in claim 19 , said probe being designed to float in the liquid.

21 . A system for detecting a change in physicochemical composition of a liquid, comprising:
a device for detecting the change in physicochemical composition of the liquid, comprising at least one communication link to a plurality of collocated sensors for sensing physicochemical variables of said liquid, and a processor configured so as to execute the respective steps of a method as claimed in claim 1 ; and at least one probe comprising the plurality of sensors and a communication link to said device.

22 . The system as claimed in claim 21 , comprising a graphical interface allowing a user to define at least one parameter for detecting the change in physicochemical composition of the liquid.

23 . A computer program product for detecting a change in physicochemical composition of a liquid, comprising program code instructions recorded on a computer-readable medium for executing a method as claimed in claim 1 when said program product is executed on a computer.

Assignees

• Suez Int

Inventors

• Maruejouls Thibaud

Classifications

• C02F2209/008

  comprising telecommunication features, e.g. modems or antennas · CPC title

• C02F2209/055

  Hardness · CPC title

• C02F2209/02

  Temperature · CPC title

• C02F2209/06

  pH · CPC title

• C02F2209/05

  Conductivity or salinity · CPC title