Maximization of generated power from wind energy conversion system using a new evolutionary algorithm T.A. Boghdady, M.M. Sayed * , E.E. Abu Elzahab Electrical Power and Machines Dept., Faculty of Engineering, Cairo University, Giza, Egypt article info Article history: Received 9 July 2015 Received in revised form 30 June 2016 Accepted 18 July 2016 Keywords: Biogeography-Based Optimization Differential evolution Sliding mode control Wind energy conversion system abstract In this paper, a grid-connected Doubly Fed Induction Generator controlled by a Sliding Mode Controller (SMC) is used to maximize the Wind Energy Conversion System (WECS) output power. A SMC is implemented using a PID controller that is tuned using a new algorithm based on hybrid Differential Evolution with a Linearized Biogeography-Based Optimization (LBBO-DE). Biogeography-Based Optimi- zation (BBO) is an evolutionary optimization algorithm based on a mathematical model of organism distribution. BBO permits a recombination of the solutions features by migration. A new migration model based on the sigmoid function is proposed. An analysis of the LBBO-DE is conducted using six different models, including the sigmoid model. Their performance were tested with 23 benchmark functions. The comparison reveals that the sigmoid model has the best performance. Therefore, the LBBO-DE with a sigmoid model is used to optimize the controller parameters to maximize the WECS output power. The LBBO-DE with the sigmoid model is compared with the Tyreus-Luyben tuning method, Genetic Algo- rithm (GA) and Linearized BBO (LBBO). The results showed that the LBBO-DE has the best performance. The proposed algorithm is verified using an experimental setup for the maximization of the generated power from the WECS and reducing power loss. © 2016 Elsevier Ltd. All rights reserved. 1. Introduction Focus on developing the productivity of wind power has increased in the past decade. By using an electrical controller, many goals can be achieved especially in the variable speed processes [1e3]. Classical controllers can be replaced by modern controllers, such as a fuzzy controller [4], robust controller [5], or adaptive controller [6] due to the development and cost reduction of microprocessor based controllers. The Sliding Mode Controller (SMC) is one of the modern controllers that is suitable when dealing with variable speed processes. The SMC has advantages of reduced order and robustness against system parameter variations and disturbances, although it has an undesirable oscillations, as known “chattering” [7e9]. The SMC is implemented in this article by using two Proportional Integral Derivative (PID) controllers. Tuning the PID controller parameters cannot be achieved optimally by conventional techniques, such as the frequency response. Ziegler-Nichols rules, based on open and closed loop testing, were frequently used in the past [10,11]. On the other hand, many papers have recently focused on intelligent controllers, such as the Artifi- cial Neural Network (ANN) controller, fuzzy control, and evolu- tionary algorithms-based controller [12]. In the last few decades, Evolutionary Algorithms (EAs) have proved their effectiveness as an optimization tool. EAs are often based on mathematics of a natural process in which the EA at- tempts to emulate the nature of some organisms in its method of selection, such as GA [13], Ant Colony Optimization (ACO) [14], Differential Evolution (DE) [15] and Particle Swarm Optimization (PSO) [16]. An EA usually consists of a set of random solutions for some optimization problems. These solutions interact with each other and they are subject to random changes. The random changes, to which the solutions are subjected, are called mutations while the interaction between the solutions, such as the crossover process in GA, is called recombination. Both mutation and recom- bination processes produce a new generation of solutions and thus the EA is transferred from one generation to another in its way to obtain the best-ever solution. Biogeography-Based Optimization (BBO) depends on the mathematics of biogeography. Biogeography is a science that deals with the migration of plants and animals between their habitats (islands). BBO had been applied to various applications such as * Corresponding author. E-mail addresses: engtarek82@gmail.com (T.A. Boghdady), Fecu.MSayed@Gmail. com (M.M. Sayed), zahab0@yahoo.com (E.E. Abu Elzahab). Contents lists available at ScienceDirect Renewable Energy journal homepage: www.elsevier.com/locate/renene http://dx.doi.org/10.1016/j.renene.2016.07.045 0960-1481/© 2016 Elsevier Ltd. All rights reserved. Renewable Energy 99 (2016) 631e646