Mass impregnated non-draining HVDC submarine cables

The permissible current loading for high voltage power cables is, among other things, determined by the maximum temperature the insulation can be exposed to without becoming damaged. For mass impregnated direct current cables, the upper limit is usually set to some 50-55 degrees Centigrade. Long term experiments where aging of the paper in this type of cable insulation is studied at different temperatures have been done. The results show that the paper can withstand significantly higher temperatures for decades without any notable degradation. Hence, thermal aging of the paper does not seem to be a critical degradation mechanism for such cables. Measurements of internal pressure and pressure changes in the insulation of 4 meter lengths of a 525 kV mass impregnated subsea cable show that load changes can cause rapid and large changes in the pressure inside the approximately 20 mm thick insulation. Minutes after a full load current was applied, pressure differences exceeding 30 bar were recorded between the inner and outer layers of the insulation. It is reasonable to assume that this causes some of the impregnation (high viscosity) to be pushed outwards. This may lead to a poorer impregnation and a risk that hazardous cavities are formed in the inner layers of the insulation when the load is turned off and the temperature drops. Experiments have also shown that the internal pressure in the cables changes slowly with time (weeks and months), even at constant ambient temperature and pressure. Strain gauge measurements on the outer layers of the cable indicate that this is due to slow plastic deformation in the lead and polyethylene layers. The internal pressure is an important parameter, as the dielectric strength of the insulation varies with the pressure. Fast load changes can give large pressure changes, which should be considered during operation of such cables. Detailed knowledge on the mechanisms governing this may, in certain cases, allow for ?rougher? load patterns than currently used without increasing the risk of failure. A numerical model has been developed that includes flow of impregnation, pressure distribution, temperature distribution and electrical field distribution in the cable under load. Results from the numerical model supports the assumption that impregnation flow due to pressure gradients is an important process for mass impregnated cables. The long time required to reach a new equilibrium of the mass distribution in the cable insulation after a change in temperature found experimentally, can be explained by the numerical model through a low permeability of the paper and the mechanical properties of the cable casing (primarily the lead, polyethylene and the steel bands). The permeability of the paper is a measure of the resistance to oil flow through the paper, a low permeability means that there is high resistance towards oil flow. Comparison of modeling results with oil flow measurements through a stack of papers, and internal pressure measurements on MIND cables under load, shows that a lower permeability than previously assumed for such cables must be used for modern MIND cables. This is probably due to high paper tension during lapping, forcing the oil to flow through the paper instead of in between the paper sheets. The pressure in the oil in the outer regions of the insulation is determined through the equilibrium between the pressure in the oil under the lead and the resistance towards expansion of the insulation exerted by the cable casing. The force form the cable casing on the insulation changes slowly with time due to slow plastic deformation in the lead and polyethylene layers. Further investigations and development of the model will be carried out to clarify the mechanisms involved and the consequences this may have regarding the load patterns that are permissible for such cables. Type tests have previously been performed at a voltage up to 525 kV which currently corresponds to the maximum operating voltage for installed mass impregnated HVDC cable systems. In this project, testing of the first 600 kV mass impregnated HVDC cable was carried out. In addition to increased voltage, the test was also performed with a higher conductor current. Correspondingly, the power transmission capacity is increased to about 1200 MW per cable resulting in a potential power rating of 2000 MW in bi-pole configuration (two cables of 1200 MW). Unfortunately, the test failed due to the increased current. The most probable reason for the failure is the test set-up where the heated oil experience low pressure in the interface between the termination and the cable. This is not a relevant scenario for operating installed cables. Simulations and testing performed in the project show that HVDC massimpregnated cables can withstand 600 kV, but the increased conductor temperature causes problems due to the test set-up.

Prosjektet har bidratt til å øke forståelsen betydelig for mekansimene som begrenser overføringseffekten i MIND-kabler. Arbeid med eksperimenter, småskalaforsøk, utvikling av en numerisk modell og fullskalatesting har økt innsikten i aldringsmekanismer, feil som kan oppstå i kabelisolasjonen og effekten av trykkvariasjoner ved hurtige lastendringer. Det er påvist at hurtige lastendringer kan gi store trykkendringer i isolasjonen som igjen kan føre til radiell strømning av impregneringsoljen og hulromsdannelse. Dette utgjør en betydelig fare for elektrisk gjennomslag. Resultatene viser også at sannsynligheten for hulromdannelse øker ytterligere ved lave temperaturer og lavt eksternt trykk. Erfarningene fra prosjektet har stor verdi for fremtidig utvikling og testing av MIND-kabler. Videre utvikling av pålitelige kraftkabler med stor overføringseffekt, MIND-kabler, vil i årene framover være svært viktig for å sikre enegiutveksling mellom land og fra sjø til land.

While extruded HVDC cables are increasingly being utilized at voltages up to 320 kV (up to 600 MW per cable), mass impregnated non-draining (MIND) cables is still the preferred option for the highest voltage/power levels. This is a proven technology that has been used for decades, but still the mechanisms limiting the utilization of the existing cables, and the current and voltage ratings of new cables, are not fully understood. In this project material characterization, numerical simulations and laboratory investigations on MIND cables will provide the foundation for improving the design and operation of such cables. Scaled models and full size prototypes will be tested to validate these designs. Increasing the power transmission rating without increasing the dimensions of the cable is only possible by raising the conductor current and/or voltage level. Raising the current will result in a higher conductor temperature and increased temperature gradient across the insulation, while higher voltage results in a higher maximum electric field in the insulation than in today's design. This project addresses how this can be achieved by modifying the materials used, the cable design and the manufacturing process with basis in a fundamental understanding of the degradation and (dielectric) breakdown mechanisms and causes for MIND cables. The planned innovations involve both improving future MIND cables and improving the operation of existing cables. HVDC subsea cables facilitate the transition to more renewable energy sources by making power transmission over large distances and to remote locations possible. An example is the possible utilization of hydro power from the Nordic region of Europe as balancing power for Europe, as these facilities can be used to store energy. In order to realize this new HVDC subsea cables with higher power ratings are needed.