SD Project: Dielectric Spectroscopy for Insulation Diagnosis

SD Project: Dielectric Spectroscopy for Insulation Diagnosis

Project director: Eng. Dorin Popa (ICMET Craiova)

The electrical insulation of power equipment undergoes a continuous aging process under normal conditions. It is severely degraded as a result of thermal and electrical stresses in the network. Aging directly leads to a drastic reduction in service life. In many cases, defects occur that affect the safe operation of the power system. This system ensures the vital transmission and distribution of electricity.

Current research is directed towards a much better understanding of the aging phenomenon. The goal is to find new advanced insulation diagnostics technologies. We want a correct estimate of the remaining life of the equipment. Several types of preventive maintenance are already applied worldwide. They ensure the security of the system operation with the lowest possible financial costs.

A first step was to select methods for investigating the condition of the electrical insulation. Methods were chosen that provide completely objective and extremely precise information. We selected the analysis of dissolved gases in oil for the vital prevention stage. We chose the measurement of partial discharges to highlight serious defects in the system.

The project approaches a completely new and revolutionary technological method in the field. We use frequency domain analysis of the response of the studied insulation. This perfectly highlights incipient defects and accurately assesses the remaining service life. The method is much less influenced by the variation of the working temperature. It is also not affected by the complex geometry of the equipment being examined.

We will use computer-aided modeling and physical experiments in the laboratory. The tests are performed on insulation mock-ups with preset degrees of severe infestation. The dielectric losses will be precisely determined as a function of the applied frequency. We evaluate the functional state of the insulation by direct comparison during diagnosis. The procedure will be validated by actual measurements on power and instrument transformers. The tests are performed both in the laboratory and on-site.

Influence of temperature and humidity

The study analyzed the known techniques for an objective diagnosis of the equipment. The current methods are strongly influenced by environmental and aging factors. The temperature of the insulation radically changes the calculated moisture content in the system. The aging products dissolved in the oil visibly increase the conductivity of the entire insulation. They practically simulate the undesirable role of water in the degradation process.

Equipment geometry and dielectric response

When the insulation has an unusual geometry, the accuracy of the moisture calculation drops sharply. The dielectric response is strongly influenced by oil barriers and channels. This response accurately indicates the moisture exclusively in the main insulation of the equipment. It is not very sensitive to the water content around the electrical conductors.

SDS, PDC and RVM analysis methods

The RVM analysis software is heavily dependent on the conductivity of the transformer oil. The PDC analysis is slightly influenced by the insulation geometry and temperature dependence. The frequency domain analysis (FDS) ensures optimal compensation for the insulation geometry. According to the FDS, the cellulose apparently dries faster with increasing measured temperature. The project established strict criteria for the selection of modeling software programs.

Moisture distribution in solid insulation

The stage fully evaluated the water content of the paper-oil insulation of transformers. It was found that the relationship between water in the oil and solid depends on the temperature. Three types of insulating structures were identified inside the transformer. The thick structures represent half of the total mass of the existing cellulose insulation.

Cold and warm thin structures

Cold thin structures operate at the temperature of the oil inside the device enclosure. They represent almost a third of the total mass of cellulosic materials. They retain a huge amount of water, available for diurnal thermal migration. Hot thin structures operate at temperatures very close to those of the conductors.

Thermodynamic equilibrium and water diffusion

The equilibrium state depends on the temperature, geometry and humidity of the insulation studied. With increasing temperature, the solubility of water directly in the oil increases significantly. The absorption capacity of cellulose decreases, forcing water molecules to migrate. When the temperature decreases, cellulose materials immediately absorb water from the transformer oil.

Mathematical model and experimental procedures

An advanced mathematical model for moisture diffusion in cellulose was presented. Strict procedures were developed for the preparation of each type of sample studied. The working procedures were optimized until the exact reproducibility of the results was obtained. The final conclusion was that all the optimal conditions for starting laboratory experiments were met.