Scientists from Novosibirsk State Technical University (NETI) have proposed using mathematical models to monitor dangerous developing defects in power transformers.
Methods of non-destructive testing and technical diagnostics of electrical equipment are the least expensive today, because their use directly contributes to reducing the number of equipment failures and accidents at electric power facilities. Power transformers with a voltage of 35-220 kV are critically important equipment in operation at electric power system facilities. The cost of the failure risk of such equipment is many times higher than the cost of its maintenance and repair. This determines the need for the early detection of developing defects leading to failure and their timely elimination by the introduction of adequate corrective actions.
With regular diagnostics, the offline operator comes to the substation with a certain frequency to take the transformer oil for analysis. The results of the analysis indicate how things are with the equipment: are there any prerequisites for defects or is everything normal.
According to Vladimir Levin, Head of the Department of Automated Electric Power Systems at NSTU-NETI, Doctor of Technical Sciences, the main task of online monitoring of transformers is to monitor the dynamics of changes in the controlled parameters of the equipment condition with the ability to predict the development of the situation, and if one of the parameters reaches (or exceeds) the set maximum permissible value— it is to fix a fault sign.
"Based on the processing of statistical data, we have obtained models that will allow us to quickly draw a conclusion about the technical condition of the transformer, determine the type and nature of the malfunction, the causes of its occurrence and the degree of danger to further operation of the equipment. Next, one of the possible operational solutions is formed to eliminate this cause using appropriate corrective action on transformer equipment. In addition, recommendations are being developed for maintenance personnel prescribing how to properly operate the transformer further," Vladimir Levin said.
The next stage of work on the project will be the refinement of the information support for the remote monitoring system.
The proposed approach can be used for the remote monitoring of a group of transformers of the same voltage class operated at substations in the area of the distribution network. Automation of remote diagnostic monitoring and distributed control of the operational status of equipment based on industrial Internet of Things technology is a promising direction in improving the reliability of transformer equipment operation.