Physics-informed neural networks for modelling power transformer’s dynamic thermal behaviour

Morozovska, Kateryna; Hilber, Patrik; Laneryd, Tor; Luvisotto, Michele

doi:10.1016/j.epsr.2022.108447

Cited by 15 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [55], the authors apply PINN to determine the rotor angle and frequency of test systems. In [56], the authors use PINN to study the thermal dynamic behavior of power transformers. In [57], the authors apply PINN to identify power systems.…”

Section: Introductionmentioning

confidence: 99%

Load Margin Assessment of Power Systems Using Physics-Informed Neural Network with Optimized Parameters

Bento

2024

Energies

View full text Add to dashboard Cite

Challenges in the operation of power systems arise from several factors such as the interconnection of large power systems, integration of new energy sources and the increase in electrical energy demand. These challenges have required the development of fast and reliable tools for evaluating the operation of power systems. The load margin (LM) is an important index in evaluating the stability of power systems, but traditional methods for determining the LM consist of solving a set of differential-algebraic equations whose information may not always be available. Data-Driven techniques such as Artificial Neural Networks were developed to calculate and monitor LM, but may present unsatisfactory performance due to difficulty in generalization. Therefore, this article proposes a design method for Physics-Informed Neural Networks whose parameters will be tuned by bio-inspired algorithms in an optimization model. Physical knowledge regarding the operation of power systems is incorporated into the PINN training process. Case studies were carried out and discussed in the IEEE 68-bus system considering the N-1 criterion for disconnection of transmission lines. The PINN load margin results obtained by the proposed method showed lower error values for the Root Mean Square Error (RMSE), Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) indices than the traditional training Levenberg-Marquard method.

show abstract

Section: Introductionmentioning

confidence: 99%

Load Margin Assessment of Power Systems Using Physics-Informed Neural Network with Optimized Parameters

Bento

2024

Energies

View full text Add to dashboard Cite

show abstract

“…To address this issue, some researchers have adopted combined 1D-3D modeling [20] techniques or have simplified the modeling analysis by focusing only on certain structures [21]. Additionally, neural networks [22], IoT sensor data [23], and thermal lattice network modeling [24] are alternatives for analyzing hotspot temperatures. Nonetheless, these methodologies cannot comprehensively represent the thermal characteristics of transformers.…”

Section: Introductionmentioning

confidence: 99%

Full-Scale Modeling and FBGs Experimental Measurements for Thermal Analysis of Converter Transformer

Yang,

Gao,

Wang

et al. 2024

Sensors

View full text Add to dashboard Cite

As the imbalance between power demand and load capacity in electrical systems becomes increasingly severe, investigating the temperature variations in transformers under different load stresses is crucial for ensuring their safe operation. The thermal analysis of converter transformers poses challenges due to the complexity of model construction. This paper develops a full-scale model of a converter transformer using a multi-core high-performance computer and explores its thermal state at 80%, 100%, and 120% loading ratios using the COUPLED iteration method. Additionally, to validate the simulation model, 24 FBGs are installed in the experimental transformer to record the temperature data. The results indicate a general upward trend in winding the temperature from bottom to top. However, an internal temperature rise followed by a decrease is observed within certain sections. Moreover, as the loading ratio increases, both the peak temperature and temperature differential of the transformer windings rise, reaching a peak temperature of 107.9 °C at a 120% loading ratio. The maximum discrepancy between the simulation and experimental results does not exceed 3.5%, providing effective guidance for the transformer design and operational maintenance.

show abstract

“…On the other hand, international researchers in the field of power transformer defect warning have focused a lot on data-driven models, such as artificial neural networks, support vector machines, random forests, and principal component analysis, to solve these problems better [10][11][12][13]. In [14], a training technique for deriving rules from a functionally approximated ANN utilizing the concentration of dissolved gases in transformer oil as the input is suggested in order to implement fault warning and defect diagnostics in transformers using artificial neural networks.…”

Section: Introductionmentioning

confidence: 99%

Transformer Fault Warning Based on Spectral Clustering and Decision Tree

Liu¹,

Chen

Li³

et al. 2023

Electronics

View full text Add to dashboard Cite

The insufficient amount of sample data and the uneven distribution of the collected data across faults are key factors limiting the application of machine learning in power transformer fault warning, as demonstrated by the poor adaptability of the established data-driven models under actual operating conditions. In this paper, an unsupervised and supervised learning method is designed for power transformer fault early warning based on electrical quantities and vibration signals. The method is based on the Fourier levels of transformer vibration signals under different electrical conditions measured in the field, and the vibration features are clustered according to their intrinsic properties by means of a spectral clustering algorithm. A decision tree model of the vibration characteristics under each cluster is then constructed to calculate early warning values for the transformer vibration spectrum under different electrical conditions, enabling the assessment of transformer production variability. The above process, which is based on field measurement data and data mining analysis methods, is cheaper than the existing transformer fault warning techniques at home and abroad and makes better use of information and training models.

show abstract

Physics-informed neural networks for modelling power transformer’s dynamic thermal behaviour

Cited by 15 publications

References 19 publications

Load Margin Assessment of Power Systems Using Physics-Informed Neural Network with Optimized Parameters

Load Margin Assessment of Power Systems Using Physics-Informed Neural Network with Optimized Parameters

Full-Scale Modeling and FBGs Experimental Measurements for Thermal Analysis of Converter Transformer

Transformer Fault Warning Based on Spectral Clustering and Decision Tree

Contact Info

Product

Resources

About