Control valve stiction detection by use of AlexNet and transfer learning

Control valve stiction is a common problem faced by the process industries, which can have strong adverse effect on profitable operation of plants. Although various detection methods based neural networks been proposed, few these studies considered performance use 2D representations signals. In this paper, such an approach pretrained convolutional network, AlexNet. The proposed network (CNN-SD) method showed highly satisfactory performance, be further applied real industrial data.