A comparative study of wavelet-based ANN and classical techniques for geophysical time-series forecasting

Bhardwaj, Shivam; Chandrasekhar, E.; Padiyar, Priyanka; Gadre, Vikram M.

doi:10.1016/j.cageo.2020.104461

Cited by 39 publications

(23 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…., n) are scale and wavelet coefficients at the resolution level. For detailed information regarding wavelet transformation, readers can refer to [52][53][54][55][56].…”

Section: Wavelet Transformsmentioning

confidence: 99%

Estimation of Daily Stage–Discharge Relationship by Using Data-Driven Techniques of a Perennial River, India

et al. 2020

View full text Add to dashboard Cite

Modeling the stage-discharge relationship in river flow is crucial in controlling floods, planning sustainable development, managing water resources and economic development, and sustaining the ecosystem. In the present study, two data-driven techniques, namely wavelet-based artificial neural networks (WANN) and a support vector machine with linear and radial basis kernel functions (SVM-LF and SVM-RF), were employed for daily discharge (Q) estimation. The hydrological data of daily stage (H) and discharge (Q) from June to October for 10 years (2004–2013) at the Govindpur station, situated in the Burhabalang river basin, Orissa, were considered for analysis. For model construction, an optimum number of inputs (lags) was extracted using the partial autocorrelation function (PACF) at a 5% level of significance. The outcomes of the WANN, SVM-LF, and SVM-RF models were appraised over the observed value of Q based on performance indicators, viz., root mean square error (RMSE), Nash–Sutcliffe efficiency (NSE), Pearson’s correlation coefficient (PCC), and Willmott index (WI), and through visual inspection (time variation, scatter plot, and Taylor diagram). Results of the evaluation showed that the SVM-RF model (RMSE = 104.426 m3/s, NSE = 0.925, PCC = 0.964, WI = 0.979) outperformed the WANN and SVM-LF models with the combination of three inputs, i.e., current stage, one-day antecedent stage, and discharge, during the testing period. In addition, the SVM-RF model was found to be more reliable and robust than the other models and having important implications for water resources management at the study site.

show abstract

“…., n) are scale and wavelet coefficients at the resolution level. For detailed information regarding wavelet transformation, readers can refer to [52][53][54][55][56].…”

Section: Wavelet Transformsmentioning

confidence: 99%

Estimation of Daily Stage–Discharge Relationship by Using Data-Driven Techniques of a Perennial River, India

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Optimal Network Structure were first subjected to denormalization process by using the network model forecasted, then 300 the MAPE value of the forecasting was determined (Table 5). 301 Renewable energy source data are not generally stationary, but series may be rendered stationary with very little difference elicit Bhardwaj et al (2020). As seen in the cologram in Figure 5, it is not stationary at serial level in 24 delayed stationary test.…”

Section: Figurementioning

confidence: 97%

“…......... Time series forecasting is a significant research field in which previous data are used to estimate future values by developing a statistical model, facilitating to develop a statistical framework to forecast future values of the system with the least predictable error(Bhardwaj et al 2020). ARIMA method, the most popular forecast method, is used for stationary time series due to its flexibility and simplicity (Kazemzadeh, Amjadian and Amraee, 2020).…”

mentioning

confidence: 99%

The Forecasting of Renewable Energy Generation for Turkey by Artificial Neural Networks and a Auto-Regressive Integrated Moving-Average Model -2023 Generation Targets by Renewable Energy Resources

Albayrak

2021

Preprint

View full text Add to dashboard Cite

Turkey attaches particular importance to energy generation by renewable energy sources in order to remove negative economic, environmental and social effects caused by fossil resources in energy generation. Renewable energy sources are domestic and do not have any negative effect, such as external dependence in energy and greenhouse gas, caused by fossil resources and which constitute a threat for sustainable economic development. In this respect, the prediction of energy amount to be generated by Renewable Energy (RES) is highly important for Turkey. In this study, a generation forecasting was carried out by Artificial Neural Networks (ANN) and Autoregressive Integrated Moving Average (ARIMA) methods by utilising the renewable energy generation data between 1965-2019. While it was predicted by ANN that 127.516 TWh energy would be generated in 2023, this amount was estimated to be 45.457 TeraWatt Hour (TWh) by ARIMA (1.1.6) model. The Mean Absolute Percentage Error (MAPE) was calculated in order to specify the error margin of the forecasting models. This value was determined to be 13.1% by ANN model and 21.9% by ARIMA model. These results suggested that the ANN model provided a more accurate result. It is considered that the conclusions achieved in this study will be useful in energy planning and management.

show abstract

“…Meanwhile, the ANN is recommended to handle timeseries data and nonlinear patterns [18], [19]. It is also proven to be the most common stochastic learning method for predicting [14], [15].…”

Section: A Prediction Techniquementioning

confidence: 99%

Modeling Orbital Propagation Using Regression Technique and Artificial Neural Network

Salleh¹,

Yuhaniz²,

Azmi³

2022

International Journal on Advanced Science, Engineering and Information Technology

View full text Add to dashboard Cite

Orbital propagation models are used to predict the position and velocity of natural and artificial objects orbiting the Earth. It is crucial to get accurate predictions to ensure proper satellite operational planning and early detection of possible disasters. It became critical as the number of space objects grew due to many countries scrambling to explore space for various purposes such as communications, remote sensing, scientific mission, and many more. Physical-based and mathematical expression approaches provide orbital propagation with high accuracy. However, these approaches require substantial expenditure to provide suitable facilities and are complicated for those with no expertise in this field. The orbital propagation model is developed using regression techniques and artificial neural networks in this study. The aim is to have a reliable and precise orbital propagation model with minimal computational and cost savings. The past orbital data is used instead of complicated numerical equations and expensive tools. As a result, the trained orbital propagation model with accuracy up to 99.49% with a distance error of 18.73km per minute is achievable. The trained model can be improved further by modifying the network model and various input data. This model is also expected to provide vital information for organizations and anyone interested. Finally, this research can help organizations with insufficient resources to have their orbit propagation model without special tools or rely on other countries with satellite data at a lower cost.

show abstract

A comparative study of wavelet-based ANN and classical techniques for geophysical time-series forecasting

Cited by 39 publications

References 26 publications

Estimation of Daily Stage–Discharge Relationship by Using Data-Driven Techniques of a Perennial River, India

Estimation of Daily Stage–Discharge Relationship by Using Data-Driven Techniques of a Perennial River, India

The Forecasting of Renewable Energy Generation for Turkey by Artificial Neural Networks and a Auto-Regressive Integrated Moving-Average Model -2023 Generation Targets by Renewable Energy Resources

Modeling Orbital Propagation Using Regression Technique and Artificial Neural Network

Contact Info

Product

Resources

About