Hexagon Shape Siw Bandpass Filter With CSRRS Using Artificial Neural Networks Optimization

Abstract

A dual-band hexagon shape substrate integrated waveguide (SIW) based band-pass filter with single loop complementary spilt ring resonators (CSRRs) is introduced in this paper. The design parameters of this filter are optimized by using artificial neural networks (ANNs). Especially an error back propagation multilayer perceptron (EBP-MLP) neural network with Levenberg-Marquart (LM) algorithm is used. A physical prototype of the proposed model is fabricated and tested. In the lower passband from 10.2 to 10.6 GHz, the insertion loss is about −0.8 dB with a fractional bandwidth of 3.85%, and in the upper passband from 12.11 to 13.31 GHz, the insertion loss is about −0.8 dB with a fractional bandwidth of 9.56%. It is observed that the insertion loss is the same in both the passbands. The obtained experimental results are in good agreement with the estimated results using full-wave analysis and ANN optimization.

A Two-Step Learning-by-Examples Method for Photovoltaic Power Forecasting

“…Artificial intelligence (AI ) and machine learning (ML) are promising candidates to effectively address the PV forecasting problem [3]. More specifically, the learning-by-examples (LBE ) paradigm encompasses different supervised learning methodologies [e.g., support vector machines (SVM s), artificial neural networks (ANN s), Gaussian processes (GP s), and deep neural networks (DNN s)] enabling the creation of accurate predictors starting from the information embedded within an off-line generated database of training examples/observations [4][5][6][7][8][9][10][11][12][13][14][15][16][17]. In such a framework, different strategies have been proposed in the state-of-the-art to predict the output power of PV plants [3].…”