A new approach for ground resistance measurements in onshore wind farms based on clamp-on meters and artificial neural network

Abstract

Monitoring the grounding resistance of wind turbine generators (WTG) over the lifespan of a wind farm is important to ensure the safety of the personnel and equipment on the facilities. Conventional methods used to measure grounding resistance are not feasible in periodic maintenance activities of onshore wind farms due to test setup complexity and associated high costs. The clamp-on meter is an alternative method proposed to measure ground resistance in multi-grounded electrical systems. However, a significant error occurs in clamp-on meters on large grounding systems, such as those found in wind farms. Hence, we propose a solution to minimize methodological errors resulting from applying the clamp-on method in wind farms based on a neural network. Besides presenting lower errors than compared methods for a real Brazilian wind farm complex, it offers even greater practicality and safety to the workers involved in the measurement, as well as less test execution time.