Improved description of partial discharges on high-voltage components
Partial discharges on parts of the power grid system under tensile stress cause considerable problems for resources such as generators, transformers, power converters and cables. If the intensity of such discharges exceeds a certain level, this causes initial damage and, in extreme cases, component failure.
As part of the MOMOS project, partial discharge phenomena occurring in equipment in electricity grids shall be analysed using a multiphysics approach. This shall also extensively incorporate the knowledge and experience gained from plasma technology, which among other things is concerned with gas discharges in a similar parameter range. It is intended that the method can be used both during operation and when the equipment is switched off. The project manager is Dr Sergey Gorchakov.
The main objectives can be divided into scientific and technical aims:
Describing individual partial discharge phenomena in actual electrical equipment using plasma diagnostics and modelling methods
Identifying the correlations from the development of partial discharges with electrical detection methods
Evaluating main stress mechanisms on insulation materials that are caused by partial discharges
Correlating the properties of individual partial discharges with the condition of the electricity grid components (evaluating the condition, ageing progress)
Developing a sensitive and specific detection method for partial discharges by combining complementary methods and approaches
Characterising the role of individual stress factors that are responsible for component ageing and the partial discharge activity
Designing a partial discharge monitoring system for analysing the component condition in real time so as to ensure the sustainable use of electricity grid components and the greater reliability of electricity grids
The starting point of the project is to classify partial discharge phenomena in various high-voltage components in electricity grids in accordance with the types of defects and their potential hazards. Components include, for example, generators, transformers and high-performance cables. Based on the classification, reference models will be derived in which the most important fault types can be reconstructed and investigated in an appropriately reproducible manner.
Electrically, acoustically and optically based plasma diagnostics methods will then be used to analyse the chemical and physical processes in the partial discharges and their effects on the ageing of the components. To this end, selected reference samples will be developed with defined positions and visually accessible partial discharges.
Parallel to this, plasma models based on models for dielectric barrier discharges and electrical simulations will be adapted and applied to characterise the observed partial discharge types and the resulting fault types. The juxtaposition and mutual comparison of the experimental and theoretical results provide the essential working steps for gaining a more in-depth physical understanding of the ageing processes.
In addition, it is also intended to use nonlinear dynamics methods to analyse time series for the partial discharge characteristics.
Typical partial discharge phenomena
Corona discharges occur on high-voltage overhead lines and on the surfaces of bare metal electrodes. Corona discharges are particularly known for their characteristic glow.
Water treeing refers to discharges creeping along soiled surfaces on the high-voltage insulators and bushings.
With electrical treeing, tree-like conductive channels form in the dielectric. Electrical treeing is an undesirable effect in insulated high-voltage cables.
Partial discharges can cause gaseous or liquid insulating media to decompose. With solid insulators, conductive channels on the surface and within the insulating materials reduce the dielectric strength of the components such as cables, high-voltage bushings and insulators. Partial discharges often lead to electrical breakdowns, i.e. complete failure of the insulation and, consequently, to short circuiting and the thermal destruction of the components.
Plasma research findings as a success factor
Diagnostic methods specially developed for micro plasma make it possible to determine the properties of partial discharge plasmas. It is to be expected that these properties are largely dependent on the condition of the object (e.g. the degree of ageing). Such differences cannot be derived using conventional methods. The project also benefits from extensive experience in modelling and simulating gas discharges, whose properties depend on, among others, the gas composition, temperature, geometry and material.
The main difference is the inclusion of the analysis of the individual results (individual discharges) for evaluating and diagnosing the electricity grid components. Another innovative approach is the use of plasma modelling that includes the description of plasma-chemical processes in combination with appropriate thermal models and which can be used for predicting the plasma properties. Basic knowledge acquired in this manner about the parameters of individual partial discharges enables the characterisation of electricity grid components at a new level. Other findings, such as whether the method can be used in combination with existing methods, will emerge during the course of the project.
Applicability of analytical methods requires investigations
Despite the diverse range of existing measurement methods for high-voltage devices, questions still remain unanswered regarding condition assessments and the service life forecasting. The superposition of multiple effects and external conditions make it difficult to draw clear conclusions.
Detailed experimental investigations of partial discharges have focussed primarily on the determination and analysis of phase-resolved partial discharge patterns (PRPD patterns), acoustic phenomena and localisation methods. The PRPD signals are recorded using standardised high-voltage pulses that are suitable for machines with a frequency of 50/60 Hz.
The transition to alternative energy sources also means, however, that other frequencies will be used that in some cases are also time-dependent. Although analysis methods used in nonlinear dynamics can be applied to electric variables, thus offering an additional diagnostic approach to the experimental investigation, the applicability of such methods for electricity grid components requires, however, detailed studies. Furthermore, several aspects such as the DC voltage behaviour, alternative grid frequencies, chemical processes and the interaction of stress mechanisms have not yet been studied in detail.
Clear project goals defined
Determining the properties of individual partial discharges will establish unambiguous measurement parameters that can be used for condition assessments and electricity grid component forecasts. Complementary research methods are intended to identify signal patterns that clearly indicate equipment faults and conditions. In addition, the engineers envisage that the developed measurement method will encompass several types and defects, and will also be applicable to different equipment. The application of such systems will substantially increase the service lives of the electricity grid components, improve the ability to plan maintenance work, and thus ensure a more secure and resource-saving operation.
The findings shall be validated under real conditions, in particular to make it possible to specify prospective application and utilisation aspects. At the end of the project, concepts for these monitoring methods will be available that will be subsequently tested in close cooperation with users in terms of their implementation and then transferred to development.
Five milestones leading to the project’s success
- Data on partial discharges in electricity grid components
- Classification of the experimental defects
- Development of reference samples
- Model development and the main part of the parametric studies
- Transferability of the methods from the reference samples to the electricity grid components is confirmed; MOMOS concept can be produced
09/2014 – 08/2017
Leibniz-Institut für Plasmaforschung und Technologie e.V.
17489 Greifswald, Germany