國立虎尾科技大學 |

Fractional-order activation functions for neural networks = case studies on forecasting wind turbines' generated power /

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Fractional-order activation functions for neural networks/ by Kishore Bingi, Ramadevi Bhukya, Venkata Ramana Kasi.
其他題名:	case studies on forecasting wind turbines' generated power /
作者:	Bingi, Kishore.
其他作者:	Kasi, Venkata Ramana.
出版者:	Cham :Springer Nature Switzerland : : 2025.,
面頁冊數:	xvii, 238 p. :ill. (some col.), digital ; : 24 cm.;
Contained By:	Springer Nature eBook
標題:	Process Engineering. -
電子資源:	https://doi.org/10.1007/978-3-031-88091-9
ISBN:	9783031880919

Fractional-order activation functions for neural networks = case studies on forecasting wind turbines' generated power /
Bingi, Kishore.

Fractional-order activation functions for neural networkscase studies on forecasting wind turbines' generated power /[electronic resource] :by Kishore Bingi, Ramadevi Bhukya, Venkata Ramana Kasi. - Cham :Springer Nature Switzerland :2025. - xvii, 238 p. :ill. (some col.), digital ;24 cm. - Studies in systems, decision and control,v. 5882198-4190 ;. - Studies in systems, decision and control ;v. 2. .

Introduction -- Fractional-order Activation Functions -- Fractional-order Neural Networks -- Forecasting of Texas Wind Turbines' Generated Power -- Forecasting of Jeju Islands Wind Turbines' Generated Power -- Forecasting of Renewable Energy Using Fractional-Order Neural Networks -- Fractional Feedforward Neural Network-Based Smart Grid Stability Prediction Model.

This book suggests the development of single and multi-layer fractional-order neural networks that incorporate fractional-order activation functions derived using fractional-order derivatives. Activation functions are essential in neural networks as they introduce nonlinearity, enabling the models to learn complex patterns in data. However, traditional activation functions have limitations such as non-differentiability, vanishing gradient problems, and inactive neurons at negative inputs, which can affect the performance of neural networks, especially for tasks involving intricate nonlinear dynamics. To address these issues, fractional-order derivatives from fractional calculus have been proposed. These derivatives can model complex systems with non-local or non-Markovian behavior. The aim is to improve wind power prediction accuracy using datasets from the Texas wind turbine and Jeju Island wind farm under various scenarios. The book explores the advantages of fractional-order activation functions in terms of robustness, faster convergence, and greater flexibility in hyper-parameter tuning. It includes a comparative analysis of single and multi-layer fractional-order neural networks versus conventional neural networks, assessing their performance based on metrics such as mean square error and coefficient of determination. The impact of using machine learning models to impute missing data on the performance of networks is also discussed. This book demonstrates the potential of fractional-order activation functions to enhance neural network models, particularly in predicting chaotic time series. The findings suggest that fractional-order activation functions can significantly improve accuracy and performance, emphasizing the importance of advancing activation function design in neural network analysis. Additionally, the book is a valuable teaching and learning resource for undergraduate and postgraduate students conducting research in this field.

ISBN: 9783031880919

Standard No.: 10.1007/978-3-031-88091-9doiSubjects--Topical Terms:

1366385
Process Engineering.

LC Class. No.: QA76.87

Dewey Class. No.: 006.32

Fractional-order activation functions for neural networks = case studies on forecasting wind turbines' generated power /
LDR:03494nam a2200337 a 4500 001 1162290
003 DE-He213
005 20250523130345.0
006 m d
007 cr nn 008maaau
008 251029s2025 sz s 0 eng d
020 $a 9783031880919 $q (electronic bk.)
020 $a 9783031880902 $q (paper)
024 7 $a 10.1007/978-3-031-88091-9 $2 doi
035 $a 978-3-031-88091-9
040 $a GP $c GP
041 0 $a eng
050 4 $a QA76.87
072 7 $a TH $2 bicssc
072 7 $a TEC031000 $2 bisacsh
072 7 $a TH $2 thema
082 0 4 $a 006.32 $2 23
090 $a QA76.87 $b .B613 2025
100 1 $a Bingi, Kishore. $e author. $3 1313935
245 1 0 $a Fractional-order activation functions for neural networks $h [electronic resource] : $b case studies on forecasting wind turbines' generated power / $c by Kishore Bingi, Ramadevi Bhukya, Venkata Ramana Kasi.
260 $a Cham : $c 2025. $b Springer Nature Switzerland : $b Imprint: Springer,
300 $a xvii, 238 p. : $b ill. (some col.), digital ; $c 24 cm.
490 1 $a Studies in systems, decision and control, $x 2198-4190 ; $v v. 588
505 0 $a Introduction -- Fractional-order Activation Functions -- Fractional-order Neural Networks -- Forecasting of Texas Wind Turbines' Generated Power -- Forecasting of Jeju Islands Wind Turbines' Generated Power -- Forecasting of Renewable Energy Using Fractional-Order Neural Networks -- Fractional Feedforward Neural Network-Based Smart Grid Stability Prediction Model.
520 $a This book suggests the development of single and multi-layer fractional-order neural networks that incorporate fractional-order activation functions derived using fractional-order derivatives. Activation functions are essential in neural networks as they introduce nonlinearity, enabling the models to learn complex patterns in data. However, traditional activation functions have limitations such as non-differentiability, vanishing gradient problems, and inactive neurons at negative inputs, which can affect the performance of neural networks, especially for tasks involving intricate nonlinear dynamics. To address these issues, fractional-order derivatives from fractional calculus have been proposed. These derivatives can model complex systems with non-local or non-Markovian behavior. The aim is to improve wind power prediction accuracy using datasets from the Texas wind turbine and Jeju Island wind farm under various scenarios. The book explores the advantages of fractional-order activation functions in terms of robustness, faster convergence, and greater flexibility in hyper-parameter tuning. It includes a comparative analysis of single and multi-layer fractional-order neural networks versus conventional neural networks, assessing their performance based on metrics such as mean square error and coefficient of determination. The impact of using machine learning models to impute missing data on the performance of networks is also discussed. This book demonstrates the potential of fractional-order activation functions to enhance neural network models, particularly in predicting chaotic time series. The findings suggest that fractional-order activation functions can significantly improve accuracy and performance, emphasizing the importance of advancing activation function design in neural network analysis. Additionally, the book is a valuable teaching and learning resource for undergraduate and postgraduate students conducting research in this field.
650 2 4 $a Process Engineering. $3 1366385
650 2 4 $a Mathematical and Computational Engineering Applications. $3 1387767
650 1 4 $a Mechanical Power Engineering. $3 1365892
650 0 $a Wind turbines $x Mathematical models. $3 768832
650 0 $a Fractional programming. $3 1489166
650 0 $a Neural networks (Computer science) $x Mathematics. $3 528378
700 1 $a Kasi, Venkata Ramana. $3 1489165
700 1 $a Bhukya, Ramadevi. $3 1489164
710 2 $a SpringerLink (Online service) $3 593884
773 0 $t Springer Nature eBook
830 0 $a Studies in systems, decision and control ; $v v. 2. $3 975418
856 4 0 $u https://doi.org/10.1007/978-3-031-88091-9
950 $a Engineering (SpringerNature-11647)