This paper presents a nonlinear flatness-based control (FBC) approach for a full-order doubly fed induction generator (DFIG) in the wind turbine system. Flat outputs of the DFIG and the FBC controller are derived using differential flatness theory. The proposed approach ensures an efficient decoupled control for both active and reactive powers of the DFIG. Also, it provides a smooth trajectory tracking in the start-up and the rest to rest modes without any saturation. Therefore, the system satisfactory operates at a variable speed of the rotor with an effective active/reactive power tracking. The variable rotor speed represents a perturbation caused by changes in the wind speed or different wind energy capacity. The requirements on the active and the reactive power are converted into system variables using a high-level reference trajectory generator (HLRTG). The effectiveness of the proposed system is verified by simulations.

The aim of this research is to finalize implementation of new method and algorithm of Collaborative and Non-Collaborative Dynamic Path Prediction for Mobile objects Collision Detection with Dynamic Obstacles in 2D and 3D Space. The method is based human behavior in collision detection with vehicles in real-life natural environment. Advantages of proposed method are full decentralization of the system, minimizing network traffic and simplifying inclusion of additional agents in the system. The proposed method is inspired by nature and implemented in mobile robotics. The method decreases uncertainty and increases predictability in collision detection with dynamic obstacles. Method allows implementation of fully functional algorithm which is tested in experimental environment and shows excellent results both in collaborative mode using exchange of coordinates as well as non-collaborative mode using OpenCV library for computer imaging and mobile objects tracking. The proposed algorithm is named Sliding Holt algorithm. This research paper should be considered as a part of series of research papers published earlier.

Single hidden layer feed forward neural networks are widely used for various practical problems. However, the training process for determining synaptic weights of such neural networks can be computationally very expensive. In this paper we propose a new learning algorithm for learning the synaptic weights of the single hidden layer feedforward neural networks in order to reduce the learning time. We propose combining the upgraded bat algorithm with the extreme learning machine. The proposed approach reduces the number of evaluations needed to train a neural network and efficiently finds optimal input weights and the hidden biases. The proposed algorithm was tested on standard benchmark classification problems and functions and compared with other approaches from literature. The results have shown that our approach produces a satisfactory performance in almost all cases and that it can obtains solutions much faster than the traditional learning algorithms.