Numerical Simulation of Aging by Water-Trees of XPLE Insulator Used in a Single Hi-Voltage Phase of Smart Composite Power Cables for Offshore Farms

Submarine power cables are expected to last 20 years without maintenance to be considered technologically reliable enough and economically beneficial. One of the main issues facing this target is the development of what is called commonly water-trees (nanometer-sized flaws filled with residual humidity), that form within XLPE (cross-linked Polyethylene) insulators and then migrate towards copper, thus leading to its corrosion and further to possible shut-down. Water trees are resulting from the coalescence of nanovoids filled with residual humidity that migrate towards copper under the combined effects of electrical forces and plastic deformation. The nanovoids are originated during manufacturing, shipping, handling and embedding in deep seas. The formation of these nanovoids leads to the degradation of the service lifetime of submarine power cables. Current research is intended to come up with a way to go a little further towards the generalization of coalescence of n nanovoids. In the perspective of multi-physics modeling, a preliminary 3D finite element model was built. Although water voids are distributed randomly inside XLPE, in this study, two extreme cases where the voids are present parallel and perpendicular to the copper surface, were considered for simplification. This will enable checking the electric field effect on neighbouring voids, in both cases as well as the influence of the proximity of the conductor on the plasticity of voids, that further leads to their coalescence. It is worthwhile to note that assessing water-trees formation and propagation through an experimental campaign of ageing tests may extend over decades. It would therefore be an exceptional opportunity to be able to get insight into this mechanism through numerical modeling that needs a much shorter time. The premilinary model suggested is expected to be extended in the future so that to include more variables (distribution and shapes of nano-voids, water pressure, molecular modeling, electric discharge.