Method for detecting presence of an electric arc and associated device

A computer-implemented process for detecting presence of an electric arc on an electric line using a machine learning method, the learning being performed using:

• a plurality of measurements relating to the formation of an electric arc, each measurement of the plurality of measurements comprising a time-correlated measurement of the arc voltage Varc, the current I and the source voltage Vsource;
• a knowledge model relating the arc voltage Varc and the current I to constants depending on the knowledge model;
• the process being implemented using a learning model derived from a learning phase during which, using the plurality of measurements and the knowledge model, a step of adjusting the learning model is implemented, the learning model taking as inputs the current I and the source voltage Vsource and providing as outputs the arc voltage Varc and the constants of the knowledge model, the learning being performed using a cost function L given by the following relationship L=L1+L2 where L1 is a first cost sub-function measuring a first difference between the arc voltage
• V
• arc
• predit
• predicted by the learning model and the arc voltage Varc measured:
• L
• 1
• =
• 
• V
• arc
• predit
• -
• V
• arc
• 
• .
• and where L2 is a second cost sub-function measuring a second difference between the prediction of the knowledge model and the values measured;
• the method comprising a detection phase including:
• a step of continuously measuring the current and the source voltage Vsource on the electric line, so as to acquire input data relating to the current I and the source voltage Vsource over a measurement window
• {
• I
• ⁡
• (
• t
• m
• )
• ,
• V
• source
• }
• m
• =
• 1
• M
• where M is the number of measurement points in the measurement window l is the current at time tm and Vsource (tm) is the source voltage at time tm;
• a step of continuously determining the presence of an arc, using a decision function for detecting, as a function of the M last input data relating to the current/and the source voltage Vsource measured on the electric line during the measurement step, presence of an arc, the decision function being a function of the second cost sub-function.