Proceedings of the Soedirman International Conference on Mathematics and Applied Sciences (SICOMAS 2021)

<Previous Article In Volume
Next Article In Volume>

Development and Testing of Artificial Neural Network with Backpropagation Algorithm to Predict the Power Ratio of Savonius Wind Turbine

Authors
Jamrud Aminuddin1, *, Bilalodin Bilalodin1
1Department of Physics, Faculty of Mathematics and Natural Sciences, Universitas Jenderal Soedirman, Jl. dr. Suparno, 61, Grendeng, Purwokerto Utara, 53122, Jawa Tengah, Indonesia.
*Corresponding author. Email: jamrud.aminuddin@unsoed.ac.id
Corresponding Author
Jamrud Aminuddin
Available Online 25 May 2022.
DOI
10.2991/apr.k.220503.018How to use a DOI?
Keywords
ANN; rotor; Savonius; wind
Abstract

A power ratio of Savonius as a vertical wind type has been predicted using an artificial neural network (ANN) with a backpropagation algorithm. The ANN is a method for processing information that is adapted from a biological neuron. This method is developed from the principle of a human being’s brain which is consist of input and output neurons. The procedure in this method is conducted using the initial input as the power ratio target to predict the next value as the output target. This prediction is employed to know the blade characteristic and its effect on the power. The ANN architecture is multi-layers with a backpropagation algorithm. The layers are input, hidden, and output. The learning rule is consisting of the forward-propagation, backward-propagation, and weight-update with the number of neurons being 2-5-1, respectively. The result of this development and testing shows that the optimum values for momentum and learning rate are 0.90 and 0.01, respectively. The result has been tested by comparing the output and input with an error of approximately 0.65%. This result indicates that the ANN method with backpropagation algorithm is prospective to predict the power ratio of various blades of Savonius wind turbine.

Copyright
© 2022 The Authors. Published by Atlantis Press International B.V.
Open Access
This is an open access article distributed under the CC BY-NC 4.0 license.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the Soedirman International Conference on Mathematics and Applied Sciences (SICOMAS 2021)
Series
Advances in Physics Research
Publication Date
25 May 2022
ISBN
978-94-6239-579-4
ISSN
2352-541X
DOI
10.2991/apr.k.220503.018How to use a DOI?
Copyright
© 2022 The Authors. Published by Atlantis Press International B.V.
Open Access
This is an open access article distributed under the CC BY-NC 4.0 license.

Cite this article

risenwbib
TY  - CONF
AU  - Jamrud Aminuddin
AU  - Bilalodin Bilalodin
PY  - 2022
DA  - 2022/05/25
TI  - Development and Testing of Artificial Neural Network with Backpropagation Algorithm to Predict the Power Ratio of Savonius Wind Turbine
BT  - Proceedings of the Soedirman International Conference on Mathematics and Applied Sciences (SICOMAS 2021)
PB  - Atlantis Press
SP  - 84
EP  - 90
SN  - 2352-541X
UR  - https://doi.org/10.2991/apr.k.220503.018
DO  - 10.2991/apr.k.220503.018
ID  - Aminuddin2022
ER  -