Mobileception-ResNet for transient stability prediction of novel power systems

Linfei Yin and Wei Ge

Energy, 2024, vol. 309, issue C

Abstract: Power system transient stability prediction (TSP) is particularly important as power systems change and evolve, including the rapid growth of renewable energy, the proliferation of electric vehicles, and the construction of smart grids. Traditional time-domain simulation methods are time-consuming and cannot achieve online prediction. Direct methods are poorly adapted and cannot be applied to complex power systems. Existing machine learning algorithms only classify the transient stability without providing the degree of transient stability of the system. Therefore, a fast and accurate power system TSP method is needed to assist operators in implementing timely measures to improve the stability of the power system running. This study proposes a Mobileception-ResNet network, Mobileception-ResNet is formed by Inception-ResNet-v2, MobileNet-v2, and a fully connected layer. In this study, Mobileception-ResNet and nine comparison models are experimented on two node systems, i.e., the IEEE 10–39 and 69–300 systems. In the IEEE 10–39 system, the root mean square error, mean absolute error, and mean absolute percentage error of Mobileception-ResNet are 44.13 %, 36.74 %, and 39.96 % lower, and the coefficient of determination is 0.04 % higher, respectively, when compared to the comparative model with the best evaluation indicator; in the IEEE 69–300 system, the corresponding values are 2.6 %, 12.83 %, 12.55 %, and 0.01 %, respectively.

Keywords: Transient stability; Deep learning; MobileNet-v2; Convolutional neural network; Inception-ResNet-v2 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544224029384
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:309:y:2024:i:c:s0360544224029384

DOI: 10.1016/j.energy.2024.133163

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().