Subventions et des contributions :

Subvention ou bourse octroyée s'appliquant à plus d'un exercice financier. (2017-2018 à 2022-2023)

Reliable operation of electric power systems is highly dependent on an insulation system that can safely isolate energized electrical components from ground and from each other. Insulation degradation and breakdown is a major root cause of the failure of electric power system equipment. The aged insulation of existing electric power systems operate at high voltages and are now under higher levels of stress than they were designed to tolerate. In addition, emerging renewable electric power sources, e.g. wind and solar energies, use power electronics that introduce high frequency voltages that accelerate the aging of the electric insulation. These stressed and aging electric insulators increase the risk of sudden equipment failure, outage, and disturbance.

Condition monitoring and diagnostics of power systems insulation are required to minimize failures and outages. A key to condition monitoring is the detection of partial discharges (PD): small, localized electrical breakdowns that occur within imperfections and voids in solid and liquid insulation materials and in insulating gases. The presence of PDs indicates that material degradation is accelerating towards a catastrophic failure, which can be avoided if PD activities are monitored. The existing electric power grid does not have sufficient or effective condition monitoring and diagnostics for its electrical insulation system.

The proposed research program is focused on the development of online, autonomous, and smart, PD collection and analysis systems. Together with my HQP, we will develop novel PD sensing hardware and establish time-domain techniques for their characterization. Accurate simulation models for various power system components (such as transformers, transmission lines and cables, and gas-insulated switch gears) will be developed that are capable of simulating the propagation of PD. We will develop PD analysis algorithms based on machine learning techniques for online condition monitoring and diagnostics of high voltage insulation systems that will form the core of a smart PD analysis system. The developed techniques will be capable of identifying the cause of discharge activities that will be an important factor in risk management and decision-making for asset managers with regards to condition-based maintenance.

The development of new and improved techniques for online condition monitoring of high voltage power systems will improve Canadian power security by contributing to the minimization of power outages and disruptions due to sudden equipment failure. These techniques will be essential to the development of a smart grid for future transmission and distribution of electric power, and will support this rapidly growing segment of the economy. The team of 10 HQP trainees of this research will have advanced multidisciplinary training and will be a key support for the future Canadian power system industry.