A comparison of regression algorithms for wind speed forecasting at Alexander Bay

Walt, Christiaan M Van der; Botha, Nicolene

doi:10.1109/robomech.2016.7813147

Cited by 7 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reference [32] proposes an algorithm based on Lasso for forecasting solar intensity. Reference [33] uses Bayesian Ridge Regression for forecasting wind speed and direction 1 to 24 hours ahead. Reference [34] proposes the group method of data handling based on the Elastic Net in order to forecasting the load of a power grid.…”

Section: Relationship To the State Of The Artmentioning

confidence: 99%

An End-to-End Trainable Feature Selection-Forecasting Architecture Targeted at the Internet of Things

et al. 2021

View full text Add to dashboard Cite

We develop a novel end-to-end trainable feature selection-forecasting (FSF) architecture for predictive networks targeted at the Internet of Things (IoT). In contrast with the existing filter-based, wrapper-based and embedded feature selection methods, our architecture enables the automatic selection of features dynamically based on feature importance score calculation and gamma-gated feature selection units that are trained jointly and end-to-end with the forecaster. We compare the performance of our FSF architecture on the problem of forecasting IoT device traffic against the following existing (feature selection, forecasting) technique pairs: Autocorrelation Function (ACF), Analysis of Variance (ANOVA), Recurrent Feature Elimination (RFE) and Ridge Regression methods for feature selection, and Linear Regression, Multi-Layer Perceptron (MLP), Long Short Term Memory (LSTM), 1 Dimensional Convolutional Neural Network (1D CNN), Autoregressive Integrated Moving Average (ARIMA), and Logistic Regression for forecasting. We show that our FSF architecture achieves either the best or close to the best performance among all of the competing techniques by virtue of its dynamic, automatic feature selection capability. In addition, we demonstrate that both the training time and the execution time of FSF are reasonable for IoT applications. This work represents a milestone for the development of predictive networks for IoT in smart cities of the near future.

show abstract

Section: Relationship To the State Of The Artmentioning

confidence: 99%

An End-to-End Trainable Feature Selection-Forecasting Architecture Targeted at the Internet of Things

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The real-time wind speed data can be provided from weather observatories or using the predicted wind speeds methods like regression technique and moving-average process (ARMA) [31]. Christiaan et al presented a comparison of regression algorithms for wind speed forecasting [32]. In this paper, we used real-time wind speed data to do simulation.…”

Section: Introductionmentioning

confidence: 99%

Achieving Low Carbon Emission for Dynamically Charging Electric Vehicles Through Renewable Energy Integration

Mou

Ying-ji²,

Jiang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Dynamic wireless charging for Electric Vehicles (EVs) can promote the take-up of EVs due to its potential of extending the driving range and reducing the size and cost of batteries of EVs. However, its dynamic charging demand and rigorous operation requirements may stress the power grid and increase carbon emissions. A novel adaptive dynamic wireless charging system is proposed that enables mobile EVs to be powered by renewable wind energy by taking advantages of our proposed traffic flow-based charging demand prediction programme. The aim is to cut down the system cost and carbon emissions at the same time, whilst realising fast demand prediction and supply response as well as relieving the peak demand on the power grid. Simulation results show that the proposed system can adaptively adjust the demand side energy response according to customers' welfare analysis and charging price, thereby to determine the power supply method. Moreover, due to the prioritised use of renewable energy, EV charging system requires less electricity from the power grid and thus the overall carbon emissions are reduced by 63.7%. INDEX TERMS Dynamic wireless charging, electric vehicles, renewable energy, demand response.

show abstract

COVID-19: Automated Detection and Monitoring of Patients Worldwide Using Machine Learning

Gupta

Salehi

Sharma

et al. 2021

Modeling, Control and Drug Development for COVID-19 Outbreak Prevention

View full text Add to dashboard Cite

A comparison of regression algorithms for wind speed forecasting at Alexander Bay

Cited by 7 publications

References 11 publications

An End-to-End Trainable Feature Selection-Forecasting Architecture Targeted at the Internet of Things

An End-to-End Trainable Feature Selection-Forecasting Architecture Targeted at the Internet of Things

Achieving Low Carbon Emission for Dynamically Charging Electric Vehicles Through Renewable Energy Integration

COVID-19: Automated Detection and Monitoring of Patients Worldwide Using Machine Learning

Contact Info

Product

Resources

About