Abstract:
This work presents the results of the application to four hydropower plants in Europe, with a total power of 1.4GW, of a recently developed monitoring and early diagnosti...Show MoreMetadata
Abstract:
This work presents the results of the application to four hydropower plants in Europe, with a total power of 1.4GW, of a recently developed monitoring and early diagnostic methodology. The innovative approach is based on data-driven and machine learning tools, such as Self-Organizing Maps, allowing an unsupervised learning of the global health state of the plant, and, at the same time, allowing to discriminate the plant variables involved in a faulty behaviour. A number of relevant incipient malfunctions were detected in early stage by our approach, during one year of operation in four plants, which are of different size and use different technologies. The feedback from the plant operators was very positive, with respect to the capacity of the system to reveal incipient faults, which were, in most cases, not properly detected by the traditional monitoring systems installed in the plants.
Date of Conference: 26-28 October 2020
Date Added to IEEE Xplore: 10 November 2020
ISBN Information: