초록 ▼

A method of preventive detection of a fault in at least one device under surveillance of a group comprising at least two devices, the device under surveillance having at least one first parameter correlated with at least one second parameter of at least one second device in the group, said parameter

A method of preventive detection of a fault in at least one device under surveillance of a group comprising at least two devices, the device under surveillance having at least one first parameter correlated with at least one second parameter of at least one second device in the group, said parameters representing state variables of said devices. The method includes the following steps: predicting a value of the first parameter from a measured value of the second parameter; comparing the predicted value of the first parameter and a measured value of the first parameter; and analyzing the result of the comparison effected in the comparison step to detect a potential fault. The invention also relates to a computer program, an installation, and a module for preventive detection of a fault in a device.

대표청구항 ▼

1. A method of preventive detection of a fault in at least a first device of a group comprising the first device and at least one second device, the first device having at least one first parameter correlated with at least one second parameter of the at least one second device, the first device and

1. A method of preventive detection of a fault in at least a first device of a group comprising the first device and at least one second device, the first device having at least one first parameter correlated with at least one second parameter of the at least one second device, the first device and the at least one second device being redundant devices, said parameters representing state variables of said first device and at least one second device, wherein the method includes the following steps: A) predicting a value of the first parameter from only a measured value of the at least one second parameter;B) comparing the predicted value of the first parameter with a measured value of the first parameter; andC) analyzing the result of the comparison effected in the step B) to detect a correlation error that relates to a potential fault. 2. The method according to claim 1, wherein analyzing the result of the comparison effected in the step B) consists in verifying whether there is a correlation error between the first parameter and the at least one second parameter at a time t. 3. The method according to claim 1, further including a step of diagnosing a type of fault if a potential fault is detected, said step consisting in: making at least one test of the first device,analyzing the result of the test and of the comparison made at step B) to determine the type of the detected fault. 4. The method according to claim 3, wherein diagnosing the type of fault consists in carrying out at least one test of the first device. 5. The method according to claim 1, wherein predicting the value of the first parameter from a measured value of the at least one second parameter is based on a prediction model defined by previously learning a coherence relationship between previously measured values of the first parameter and the at least one second parameter. 6. The method according to claim 5, wherein the learning is effected by a learning method utilizing a neural network system. 7. The method according to claim 5, wherein the learning is effected by a learning method of the statistical type. 8. A method according to claim 2, wherein the steps A), B), and C) are repeated systematically until a correlation error is detected between the first parameter and the at least one second parameter. 9. A method according to claim 5, wherein the prediction model defined for a given time t is corrected regularly as a function of the values of the first parameter and the at least one second parameters measured at a time preceding the time t. 10. The method according to claim 1, wherein at least one device in the group is an oil-insulated power transformer or a power rectifier. 11. A computer program for carrying out the method according to claim 1 when it is executed on a computer, wherein the computer program includes instructions for carrying out the following steps: A) predicting a value of the first parameter from a only a measured value of the at least one second parameter;B) comparing the predicted value of the first parameter with a measured value of the first parameter; andC) analyzing the result of the comparison effected in the step B) to detect a correlation error that relates to a potential fault. 12. The method according to claim 1, wherein the first parameter and the second parameter respectively represent the same state variable of the first device and the second device, respectively. 13. The method according to claim 1, wherein the first parameter and the second parameter represent different state variables, and wherein the first parameter is linked to the second parameter by a relationship, the relationship being the relationship that exists between a temperature of the first device and an electrical power consumed by the second device, or between a pressure in an actuator of the first device and a gas flow rate of the second device. 14. An installation including a group of at least two devices, a first device, having at least one first parameter correlated with at least one second parameter of at least one second device in the group, the first device and the at least one second device being redundant devices, said first parameter and at least one second parameter representing state variables of said first device and the at least one second device, respectively, wherein said installation further includes: an acquisition system adapted to measure the first parameter and the at least one second parameters;a processing module adapted to communicate with the acquisition system and to predict a value of the first parameter from only a measured value of the at least one second parameter and to compare the predicted value of the first parameter with a measured value of the first parameter; anda decision module adapted to analyze the result of comparing the predicted value of the first parameter and the measured value of said first parameter in order to detect correlation error that relates to a potential fault. 15. The installation according to claim 14, further including a diagnostic module adapted to communicate with the processor system and to determine the type of fault. 16. The installation according to claim 14, wherein at least one device in the group is an oil-insulated power transformer or a power rectifier. 17. A module for preventive detection of a fault is adapted to detect a correlation error between at least a first parameter and at least a second parameter representing respective state variables of a first device and a second device both belonging to a group of devices installed in an installation according to claim 14, said module including: prediction means for predicting a value of the first parameter from only a measured value of the at least one second parameter;comparison means for comparing the predicted value of the first parameter and a measured value of that same first parameter; andanalysis means for analyzing the result of the comparison effected by the comparison means in order to detect a correlation error that relates to a potential fault.