Multi-agent architecture for smart micro-grid optimal control using a hybrid BP-PSO algorithm for wind power prediction

Sigmoidal and Gaussian functions based neural neuron technique for grid interactive solar energy system enabling power quality improvement

“…It is used to filter the load currents for higher order harmonics. The output of the summation neurons ( U P∑ and U q∑ ) is given as [26] where w ∑ bias presents initial bias of ∑ and ∑ 1 . These summation neurons transform load currents into internal signals as U p∑ and U q∑ .…”

“…The signals, U p∑ and U q∑ , are transformed by the sigmoidal functions ( f s and f s 1 ) into signals U p∑S and U q∑S within unity amplitude. These signals, U p∑S and U q∑S , are estimated as [26] where λ ∑ is a sigmoidal gain factor. It is taken in this ANN‐based control, as unity.…”

“…The substantial literature survey shows the control algorithms, such as radial based network functions [24], least mean square‐based control, and a training function [25], can be executed to control the PCC voltage distortions. The control structures like back‐propagation [26], ADALINE [27], an adaptive filter‐based control [28], are used to extract active load component. An application of damped SOGI controller is reported in [29].…”