Search

Design of controller for a doubly fed induction generator driven by a variable speed wind turbine employing deep learning neural networks whose weights are tuned by grey artificial bee colony algorithm is developed and simulated in this work. This paper presents the mathematical modelling of the doubly fed induction generator (DFIG) and the controller design is implemented using the third generation deep learning neural network (DLNN). In the proposed work, the variable speed wind turbine generator torque is regulated employing a proportional–integral–derivative (PID) controller. The gains of the PID controller are tuned using DLNN model. The proposed density-based grey artificial bee colony (D-GABC) algorithm provides the optimal dataset required for training DLNN model. As well, the weights of the developed neural network controller are also optimized by D-GABC algorithm to avoid premature convergence and to reduce the incurred computational time of the network model. The effectiveness of the proposed DLNN-based controller for DFIG in wind energy conversion is proved and observed to be better than that of the other methods proposed in the previous literature works in respect of the simulated results obtained.

Object segmentation is a prominent and challenging issue pertaining to image analysis and computer vision applications. This segmentation enables a higher number of applications like image retrieval, object recognition and object reconstruction. Considering this importance of object segmentation, the ultimate aim of the proposed research work in this paper is to focus on color image segmentation that is visually triggered using the GBCPA, popularly known as global biotic cross-pollination algorithm dependent on evolutionary strategy models. Evolutionary techniques imitate the procedure of genetic evolution and make utilization of the operators of recombination, mutation and selection to create the new-generation individuals. The new flower pollination optimization algorithm (FPOA) proposed in this paper aims to improve the objective characteristics of the basic FPOA approach. The significant contribution here mainly focused on preventing the premature convergence for dealing with the complicated constraints of the color image segmentation problem. The performance of GBCPA is tested over a Gould segmentation data set comprising of 715 images, both urban and rural images. The data set is tested under various evaluation techniques, and the proposed algorithm results in better accuracy than the other methods considered in comparison with the existing literature.

This paper focuses on the visual-based colour image segmentation with a global biotic cross pollination algorithm (GBCPA). The global biotic cross pollination algorithm segments the structurally challenging objects based on the colour, edge, entropy and edge information in the CIE L*a*b* colour space. The L*a*b* colour space is a colour-opponent space considered to approximate human vision. L* denotes the luminosity or brightness layer, chromaticity layer a* indicates colour falls along red-green axis and chromaticity layer b* indicates the blue-yellow axis. The FPO algorithm considering the global biotic cross pollination is proposed to improve the quality of the solution and computational speed. GBCPA is first introduced to find the locality of the solution. The performance of GBCPA is tested on a standard Berkeley segmentation dataset with 300 images. The dataset is illustrated under different evaluating strategies.