Temperature rise testing of current transformers: improvement in test method

The nature of electric power generation, transmission and distribution systems frequently faces high voltages, high currents, or a combination of both. Electric phenomena as well as material robustness requirements often mean that direct measurements using conventional voltmeters and ammeters are not cost effective, or even possible. For this reason, indirect measuring devices such as instrument transformers are employed in order to operate protective relays, thus preserving system stability and, as consequence, system reliability. Instrument transformers provide the only interface between high-power electric circuits and control electronics, thus becoming a key component of any power system. In order to verify that Current Transformers will be able to handle the stresses in the networks in which they will be installed, a series of socalled type tests are performed. Among these type tests is the Temperature Rise Testing, which is used to validate the Current Transformer’s ability to withstand a rated temperature within an operating timeframe. The temperature rise test is a standardized test. The aim of this thesis work is to evaluate the methodology being currently used in the industry and to propose improvements by applying concise analysis, electrical power engineering principles and optimization theories in order to increase measurement accuracy while also reducing test performance times. It has been suggested that the developed method should also ideally comprise modules for easy mounting, as well as flexibility to be used in other type of temperature rise tests, for an increased amount of CT Channels. The ultimate goal is to get familiarized with the development of type testing procedures in order to provide insightful feedback and thus real improvement in the quality of the Temperature Rise Test, while at the same time developing a methodology which should be both simple and straightforward. Throughout the testing development, one of the key findings was that a big percentage of the calculations were being performed by hand because of software incompatibilities. It was then proposed that time could be reduced significantly by creating a standardized software platform to collect data and perform automated calculations. Additionally, the process’ predictability was also increased by means of the establishment of a baseline, calculated from experimental data. While this last feature did not reduce the time required to reach a steady state, it improved the system predictability and thus reliability. Developments in reducing hardware setting times for testing were also explored, although some of them were ultimately deemed as not having a reasonable cost-benefit ratio, as well as providing a reduced overall test reliability