Comparison of Machine Learning Algorithms for the Prediction of Mechanical Stress in Three-Phase Power Transformer Winding Conductors

Fausto Valencia,Hugo Arcos,Franklin Quilumba,Renato Procopio

doi:10.1155/2021/4657696

Fausto Valencia, Hugo Arcos + Show 2 more

Open Access

https://doi.org/10.1155/2021/4657696

Copy DOI

Abstract

This research compares four machine learning techniques: linear regression, support vector regression, random forests, and artificial neural networks, with regard to the determination of mechanical stress in power transformer winding conductors due to three-phase electrical faults. The accuracy compared with finite element results was evaluated for each model. The input data were the transient electrical fault currents of power system equivalents with impedances from low to high values. The output data were the mechanical stress in the conductors located in the middle of the winding. To simplify the design, only one hyperparameter was varied on each machine learning technique. The random forests technique had the most accurate results. The highest errors were found for low-stress values, mainly due to the high difference between maximum and minimum stresses, which made the training of the machine learning models difficult. In the end, an accurate model that could be used in the continuous monitoring of mechanical stress was obtained.

Highlights

Journal of Electrical and Computer Engineering magnetic material extends towards infinite
FEM is still necessary to get the training data, it is no longer used for the rest of the power transformer lifetime after the model is obtained. e drawback is the difficulty of training the artificial neural networks (ANNs). ey have many hyperparameters that affect the model accuracy [18]
Is research explores the use of four machine learning techniques for the determination of mechanical stress: linear regression (LR), support vector regression (SVR), random forests (RF), and ANN. e objective is to compare each technique’s accuracy when varying only one hyperparameter, simplifying the model design and implementation

Summary

Units MVA kV kV Hz

Core diameter Core window height Limb-limb separation Low voltage winding inside diameter Low voltage winding outside diameter High voltage winding inside diameter High voltage winding outside diameter Low voltage disk height High voltage disk height Spacer block (located between disks). Equation (3) shows the formulation for the low voltage winding transient current, ILV, where ω is the angular frequency of the system, t is the time, φ is the angle representing the fault starting point, θ is the angle between phases (120° in a three-phase balanced system), and λ ωr/xl, where r and xl are the equivalent resistance and inductive reactance seen by the fault. E radial force in the middle conductors of the windings represents the highest value [17]. Erefore, a simplified model of the winding conductor can be used to determine the stress, where the conductor is modelled as a ring with radius Radring and cross-sectional area Sc. us, the force P normal to the section of the conductor subjected to a radial force per length Fr and the stress σ are calculated by the following equations, respectively:.

Results

Phase C

FEM SVR

Conclusions