Method for monitoring an electrical installation comprises

Patent metadata

Field	Value
Publication number	US-12152941-B2
Application number	US-202017117202-A
Country	US
Kind code	B2
Filing date	Dec 10, 2020
Priority date	Dec 12, 2019
Publication date	Nov 26, 2024
Grant date	Nov 26, 2024

Abstract

A method for monitoring an electrical installation comprises:
while the installation is in operation, measuring electrical currents that flow through the installation and temperature values at predefined locations in the installation; using a numerical model acquired beforehand, computing a numerical index representative of thermal properties of the installation, from differences between the measured temperature values and corresponding temperature values estimated by means of said model from the measured current values; and detecting a thermal anomaly when the computed numerical index differs from a reference value.

First claim

The invention claimed is:

1. A method for continuously monitoring an electrical installation comprising:
while the installation is in operation, measuring on a continuous basis over time electrical currents that flow through the installation and temperature values at predefined locations in the installation by means, respectively, of current sensors and temperature sensors disposed in the installation at the predefined locations, wherein the predefined locations are one or more upstream and/or downstream locations relative to one or more points of connection of conductors of the installation through which the electrical currents flow;
by an electronic data processing device and using a numerical model acquired beforehand, automatically computing, on a continuous basis, a numerical index representative of thermal properties of the installation, from differences between the measured temperature values and corresponding temperature values estimated by means of said model from the measured current values;
by the electronic processing device, detecting, on a continuous basis, a thermal anomaly when the computed numerical index differs from a reference value;
based on the detected thermal anomaly and using the predefined locations of the current sensors and temperature sensors, identifying an occurrence and a location of a defect or of a situation likely to lead to a defect at the identified location that can compromise safety of the installation; and
responding to the identification of the occurrence of the defect or situation by providing the identified location for correcting the defect or situation using the identified location, to mitigate a compromise of safety,
wherein the numerical model is representative of thermal properties of the electrical installation and is configured to associate current values measured by the current sensors with temperature values estimated for the locations at which the temperature sensors are positioned, the numerical model being previously parameterized by learning performed on the electrical installation; and
wherein the relationship between the temperature estimated for one of the locations and the current values measured in the installation is given by the following formula:

θ
t
i

=

α
i

+

∑

k
=
1

L

⁢

∑

l
=
1

m

⁢

α

k
,
l

⁢

θ

t
-
k

l

+

∑

k
=
1

q

⁢

∑

l
=
1

m

⁢

β

k
,
l

⁢

P

t
-
k

l

+

n
t

in which θ t i designates the temperature estimated for this location at a given instant, θ t-1 i designates the temperature estimated for this same location at a preceding instant, “n” is a statistical noise associated with the temperature sensor for this instant, “m” is the number of current sensors, “L” is the duration of the measurement window, “P” is the thermal power which depends on the measured currents and “α” and “β” are parameters of the numerical model.

2. The method according to claim 1 , which comprises a preliminary step of parameterizing of the numerical model, the preliminary step comprising:
while the installation is in operation, measuring, over time, the electrical currents that flow through the conductors of the installation and temperature values at the predefined locations in the installation by means, respectively, of the current sensors and the temperature sensors disposed in the installation;
computing parameters of the model from the measured current and temperature values.

3. The method according to claim 1 , which comprises a preliminary step of parameterizing of the numerical model, the preliminary step comprising:
while the installation is in operation, measuring, over time, electrical currents that flow through conductors of the installation and temperature values at the predefined locations in the installation by means, respectively, of the current sensors and the temperature sensors disposed in the installation;
computing parameters of the model from the measured current and temperature values; and in which computing the parameters of the model comprises minimizing a mean square error given by the following formula, for each pair of current and voltage sensors, from a training dataset derived from the installation and for which the installation has experienced no anomaly:

α
ˆ

k
,
l

⁢

β
ˆ

k
,
l

=

arg

min

α
,
β

∑

t

⁢

∑

i
=
0

m

⁢

Assignees

• Schneider Electric Ind Sas

Inventors

• Youssef Wedian
• Schuster Philippe
• Kilindjian Christophe
• Reymond Bruno
• Marie Sylvain
• Rilling Gabriel

Classifications

• H02J13/10

  characterised by displaying of information or by user interaction, e.g. supervisory control and data acquisition [SCADA] systems · CPC title

• H02J13/12

  Monitoring network conditions, e.g. electrical magnitudes or operational status · CPC title

• H02J13/16

  the power network being controlled at grid-level, e.g. using aggregators · CPC title

• G01R19/0092

  Measuring current only · CPC title

• G01R15/181

  using coils without a magnetic core, e.g. Rogowski coils · CPC title