Determining accelerated aging power cable spatial temperature profiles using Artificial Neural Networks

Ge, Xufei and Given, Martin and Stewart, Brian G.; (2022) Determining accelerated aging power cable spatial temperature profiles using Artificial Neural Networks. In: 2022 IEEE International Conference on High Voltage Engineering and Applications (ICHVE). 2022 IEEE International Conference on High Voltage Engineering and Applications (ICHVE) . IEEE, CHN, pp. 1-4. ISBN 9781665407519 (https://doi.org/10.1109/ICHVE53725.2022.9961792)

[thumbnail of Ge-etal-ICHVE-2022-Determining-accelerated-aging-power-cable-spatial-temperature-profiles]
Preview
 Text. Filename: Ge_etal_ICHVE_2022_Determining_accelerated_aging_power_cable_spatial_temperature_profiles.pdf
Accepted Author Manuscript
License: Strathprints license 1.0
Download (1MB)| Preview

Abstract

The spatial temperature profile of Medium-Voltage (MV) extruded power cable, undergoing accelerated aging in a tank filled with water according to IEEE standard 1407 guidance, is estimated by combining Finite-Element Modelling (FEM) and an Artificial Neural Network (ANN). ANSYS Fluent is first used to establish a 3-D finite-element model and to simulate the temperature distribution within the power cable. In order to estimate temperature at any position within the power cable thus informing a more accurate aging model based on the variation temperature with position, a 3 layers ANN, trained by Bayesian regularization back-propagation is then developed. For the ANN, the FEA simulation temperature profiles at specified nodes are used as the input information. The resulting model is useful to understand how each position within a cable undergoing artificial aging is affected by different temperatures.

ORCID iDs

Ge, Xufei, Given, Martin ORCID logoORCID: https://orcid.org/0000-0002-6354-2486 and Stewart, Brian G.;
CORE (COnnecting REpositories)