This paper presents a data-driven approach to short-term wind turbine fault prediction and condition monitoring based on a hybrid architecture of recurrent neural network (RNN) and long short-term memory (LSTM). The proposed architecture is established by utilizing time series data from SCADA and a Bladed model of a 5 MW wind turbine to predict faults occurring to the wind generator. The RNN-LSTM training procedure is enhanced with self-organizing maps and LSTM auto encoder so as to describe the complex interaction between the mechanical system and unpredictable wind speed. To verify the performance of the proposed scheme, we conduct in-depth numerical experiments by applying the hybrid architecture to the Bladed 5 MW wind turbine model with rated wind speed of 11.8 m/s. Experiment results confirm that the proposed scheme has superior accuracy and practicality of fault prediction compared with eminent existing machine learning algorithms such as extreme gradient boost and random forest regressor.

Global stabilization of complex biological networks means controlling the dynamic state of the network to converge to a desired stable state, or an attractor, irrespective of initial states by perturbing a set of control target nodes. Global stabilization is of paramount importance in systems biology since attractors represent key phenotypes of biological systems. Most of the previous attempts to solve the global stabilization problem resort to either near-brute force or analytical searching with huge computational burden. In this paper, we propose a novel control scheme that can achieve global stabilization of complex biological networks by avoiding such issues. For this purpose, a Boolean network model of biological networks is considered. We reduce the Boolean network using a simple coordinate transformation with respect to the desired attractor without loss of the information on connectivity that is needed for global stabilization. We then identify control inputs by searching for a minimum set of nodes the perturbation of which makes the reduced Boolean network acyclic. Our control scheme ensures global stabilization with modest complexity and scalability.

A common approach to controller synthesis for hybrid systems is to first establish a discrete-event abstraction and then to use methods from supervisory control theory to synthesise a controller. In this paper, we consider behavioural abstractions of hybrid systems with a prescribed discrete-event input/output interface. We discuss a family of abstractions based on so called experiments which consist of samples from the external behaviour of the hybrid system. The special feature of our setting is that the accuracy of the abstraction can be carefully adapted to suit the particular control problem at hand. Technically, this is implemented as an iteration in which we alternate trial control synthesis with abstraction refinement. While localising refinement to where it is intuitively needed, we can still formally establish that the overall iteration will solve the control problem, provided that an abstraction-based solution exists at all.

While type-2 fuzzy control has gained a lot of attention in the recent years, several challenges still remain with respect to controller design. The footprint of uncertainty (FoU) plays a very important role in designing type-2 fuzzy controllers, since the performance of type-2 fuzzy controllers largely depends on the choice of FoU. In this paper, we propose a simplified model of interval type-2 (IT2) fuzzy two-term controllers of Takagi–Sugeno (TS) type with only two rules in the rule base. The controller model is derived with varying FoUs based on only two type-2 input fuzzy sets (the simplest case). An extension to multiple input fuzzy sets is also presented. We investigate the variation in control surface and computational aspects of the IT2 fuzzy two-term controller as the FoU is varied. The performance of the proposed IT2 fuzzy controller with respect to varying FoUs is evaluated in the simulation study.