With evolving maintenance strategies in the electricity industry internationally, there has been increasing pressure to develop improved techniques for condition monitoring. Specifically there has been a trade off between the speed and accuracy of testing. Traditionally, transformer condition monitoring involved high accuracy tests, which due to their duration, could only be performed on a discrete periodic basis. ElectraNet SA has experienced many limitations associated with this form of condition monitoring, and recently there has been a trend towards high speed on-line monitoring techniques for power transformers. Though these new techniques do not provide the level of accuracy found in traditional forms of testing, they overcome many of their limitations. This paper, describes ElectraNet SA's techniques and experience with power transformer monitoring.
This paper describes the results of a study focused on enhancing the performance of a non linear dynamic inversion scheme augmented with a neural network to cancell the dynamic inversion error. The approach is based on adding a pre-trained neural network providing the values of the aerodynamic stability and control derivatives required by the dynamic inversion calculations, as the aircraft moves throughout its flight envelope. Additionally, a comparison is performed using two different classes of neural networks (Sigma-Pi and EMRAN algorithms) for the cancellation of the dynamic inversion errors. The study is performed using the WVU IFCS F-15 simulation environment. The results show that the updating of the aerodynamic derivatives reduces the error compensating activity of the neural network. Performance improvements in terms of tracking error are observed for some maneuvers; however, a significant sensitivity to the update rate has been noticed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.