A Quality Assurance of NKT’s Temperature Measuring : Evaluation and Uncertainty Analysis of the Temperature Measuring Chain on High Voltage Cables

Abstract: An overall increase of interest and need for high voltage cable have been acknowledge overthe last years. This is mainly due to the effect of the rapid expansion and transition fromfossil-energy sources to a future with non fossil-energy. Good and capable high voltagecables are needed, whether it is to connect countries to a large power grid or transfer a largemagnitude of energy from offshore wind farms. One of the world leading companies thatsupply and develop these types of cable is NKT. During production, several tests are carriedout to ensure that the cables will withstand the stresses they will later be subjected to oncethey are in operation. Majority of these tests focus on evaluating the thermal stress insidethe cables and the importance of measuring the correct temperature becomes crucial. Thisstudy aims to analyze and evaluate the uncertainty within the temperature measurementchain of NKT’s high voltage cables. The study uses the laws of thermodynamics to clarifythe various effects and consequences that can occur when temperature sensors are incorrectly installed, in this study type-T thermocouples. The thermocouples form the first partof the whole temperature chain and its installation will prove to have a significant effect onthe result. The thermocouples were mounted on a HVAC cable that underwent 4, 24 hoursheating cycles. There were 33 thermocouples and they were divided between two differenttypes of logger systems. The data collected during the cycles is used to compare the twosystems. In addition to the experimental evaluation of the two systems, a theoretical evaluation and analysis was also carried out. The results between the two methods showed somedifference but in the end, it was chosen to consider the two systems as equivalent. The experimental method resulted in an uncertainty of the temperature chain regardless of thelogger used to ± 1.6◦C with a conductor temperature of 70◦C, while the theoretical value forthe two logger-system was calculated to ± 1.5◦C up to ± 1.8◦C depending on which calculation method that was used. Unlike the experimental method, the calculations were based ona conductor temperature of 100◦C, this to account for the contributions that were difficultto evaluate.