Search

Energy demand management is particularly important for developing and emerging economies. Their energy consumptions increase significantly, depending on their growing economies. As a result of Turkey’s rapid economic and population growth, electricity consumption is increasing. Electricity consumption forecasting plays an essential role for energy suppliers, consumers, and policy makers. Therefore, using models to accurately and reliably forecast future electricity consumption trends is a key issue for the planning and operation of electric power systems. This paper focused on forecasting electrical energy consumption by utilizing deep learning methods, i.e., Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models, for time series data. One-hour and three-hour ahead forecasting are accomplished by using a historical dataset of electrical energy consumption in Turkey. The comparison results show that the GRU model is slightly better than that of the LSTM. Our study also reveals that one-hour ahead predictions are more accurate than three-hour ahead predictions.

The novel Coronavirus (COVID-19) has significantly affected millions of people around the world since the first notification until nowadays. The rapid spread of the virus has dramatically increased the workload of healthcare systems in many countries. Therefore, the need for efficient use of the healthcare system leads researchers to forecast the trend of virus spread. For this purpose, Machine Learning (ML) and Artificial Intelligence (AI) applications have intensively used to struggle against the coronavirus outbreak. In this study, Temporal Convolutional Network (TCN) is applied for modeling the cumulative confirmed COVID-19 cases and forecasting the spread of it in various European countries using time series data. It is also presented that numerical examples for comparing performances of TCN against Long-Short Term Memory (LSTM) and Gates Recurrent Units(GRU) in terms of computation time, root-mean-square error (RMSE), normalized root-mean-square error (NRMSE), root mean squared log error (RMSLE), mean absolute percentage error (MAPE), and symmetric mean absolute percentage error (SMAPE). Simulation results indicate that the Temporal Convolutional Networks used in this manuscript performs better than other models for forecasting the cumulative confirmed COVID-19 cases.

Due to their nice properties, unimodular sequences possessing low autocorrelation sidelobes have already been employed as transmit waveforms in numerous radar systems. It could also be desired to have a transmit waveform which avoids certain frequency bands allocated for specific communication applications. Hence, adaptation and specific design of the waveform spectrum might also be essential. Accordingly, in this manuscript, we propose two alternative algorithms for designing radar transmit sequences satisfying simultaneous temporal integrated sidelobe level (ISL) and spectral stopband constraints. The proposed algorithms utilize the majorization-minimization (MM) method for directly minimizing ISL by first proposing a majorizing function in place of the objective function of the minimization problem. Then, solution of the minimization problem is obtained in an iterative manner. Accelerated versions of the newly proposed algorithms are also derived to facilitate their fast convergence. We also present numerical examples for comparing performances of our newly proposed methods against some existing methods from the literature in terms of computation time, number of iterations, and level of suppression in stopbands. (C) 2020 Elsevier Inc. All rights reserved.