Influence of Pollution Sources on Insulation Performance


The quality of external insulation, whether in a substation or on an overhead line, is one of the key factors determining reliability of a power grid. In this regard, choice and dimensioning of suitable insulation is closely linked to the pollution level that will be experienced by the equipment at any specific location. Inappropriate choice of insulation can lead to technical problems in the short and medium term. Over-dimensioning increases cost of structures while under-dimensioning can lead to penalties related to failures and ultimately to the need to upgrade structures.

Assessing the level of pollution affecting an electrical network and consequently deciding on required dimensioning of insulation is necessary whenever a new grid is defined or when there is a modification to operating voltage. In cases where an overhead line is already being operated, it is current practice to carry out a study of past failure incidents. Then, current dimensioning is considered according to some acceptable rate of such incidents.

Another approach is to review past or interim pollution maps. In this case, two issues have to be addressed: If the map was established years ago, are the sources of pollution still present and in the same quantities? And, if the map is recent, how was it constructed and what parameters were taken into account? Given the answer to these questions, the confidence level in available information can be challenged.

Electrical activity on the surface of an insulator depends on the nature and state of the pollution deposit on its surface, on the atmospheric conditions and also on type of insulator material and profile. For example, a pollution deposit can be intrinsically conductive or become conductive if sufficiently wetted. In both cases, criticality of surface pollution is influenced by its quantity and/or nature. Identifying the principal pollution source or sources is therefore essential. Moreover, since wind is the main vector of pollution between source and insulator surface, its impact can be differentiated according to whether the scale being used is local, regional, continental or global.

Setting aside exceptional pollution episodes, such as exceptional salt storm events, aerosol density of pollution deposits tends to stabilize over time as a result of factors such as natural washing from rain or due to abrasion of the insulator surface by sand. For the same amount of pollution affecting a grid, the frequency of such natural washing will determine criticality of the pollution layer.

In summary, the level of pollution at any given location in the grid depends on identification of its source, nature of the specific pollutants, the way they are deposited on insulators and the frequency of natural washing.

Fig. 1: Schematic history of polluted insulator.

The Köppen-Geiger classification is a three-letter code used to identify an area according to its climate for any point on the earth. As example, the climate in Martigues, the location of a large outdoor insulator test station on the Mediterranean, is classified CSA, namely temperate and warm. Fig. 2 shows the diversity of climates observed in Asia using this classification.

Table 1: Köppen Climate Classification Scheme
Fig. 2: Köppen-Geiger climate type map of Asia.

Obtaining knowledge on climate provides information on rainfall nature, intensity and frequency and is therefore valuable in anticipating behaviour of external insulation exposed to pollution. For more regional and local approaches, using measurements from weather stations enables also taking into account the specificities of local topography. Mesh size of the study area is important as well to further refine the quality of such information.

Attend the 2022 INMR WORLD CONGRESS where research specialist and insulation testing expert, Christian Pons of EDF Lab les Renardières in France, will deliver a lecture on characterization of pollution level using classical tools and complex modelling. He will also highlight limitations of each alternative method as well as associated level of confidence.