No abstract

This paper proposes an Energy Management System for the optimal operation of Smart Grids and Microgrids, using Fully Connected Neuron Networks combined with Optimal Power Flow. An adaptive training algorithm based on Genetic Algorithms, Fuzzy Clustering and Neuron-by-Neuron Algorithms is used for generating new clusters and new neural networks. The proposed approach, integrating Demand Side Management and Active Management Schemes, allows significant enhancements in energy saving, customers' active participation in the open market and exploitation of renewable energy resources. The effectiveness of the proposed Energy Management System and adaptive training algorithm is verified on a 23-bus 11 kV microgrid.

Because of their excellent scheduling capabilities,\ud artificial neural networks (ANNs) are becoming\ud popular in short-term electric power system forecasting,\ud which is essential for ensuring both efficient and reliable\ud operations and full exploitation of electrical energy trading\ud as well. For such a reason, this paper investigates the\ud effectiveness of some of the newest designed algorithms in\ud machine learning to train typical radial basis function (RBF)\ud networks for 24-h electric load forecasting: support vector\ud regression (SVR), extreme learning machines (ELMs), decay\ud RBF neural networks (DRNNs), improves second order,\ud and error correction, drawing some conclusions useful for\ud practical implementations

No abstract

Traditional bipolar differential amplifiers have only a ±5mV operational range with nonlinear distortion below 0.1dB. In this paper, a linearization technique based on neural network training algorithm is proposed to expand this 0.1dB linear region to a much wider ±200mV range. Compared with the traditional and recent state-of-the-art techniques for linearization, where gain or noise performance is always being tradeoff for high linearity, the proposed technique leads to the increase of the amplifier's gain with low noise. Another advantage of the proposed approach is that, it is implemented using a simple, paralleled structure that leads to a smaller area and less power consumption than other linearization schemes. The design procedure also enables a completely customizable solution based on required linear range, ripples, and the number of available transistors. The experimental results show that the proposed linearized amplifier has a very good linearity and is capable to wide bandwidth, good gain, and low noise performance.