Assessment of the current technical condition of electrical equipment based on an artificial neural network

Abstract Assessment of the current technical condition is an important task, so the state of electrical equipment depends on its further operability. Thanks to modern computing devices, it is possible to implement actively artificial intelligence and computer-assisted learning methods that allow achieving high efficiency in data processing. A study was conducted and an algorithm for diagnosing the technical condition based on an artificial neural network was developed. A model based on a multilayer perceptron is proposed, which allows evaluating the technical condition of a high-voltage power transformer. The result of the technical diagnostics of the model is the assignment of the condition to one of the five classes, proposed by the guidelines presented by the International Council on Large Electrical Systems. The methodology is presented on the example of a 250 MVA transformer with a certain defect history, which allowed us to show the reliability and validity of the obtained results. It is shown that the use of the proposed model makes it possible to achieve accuracy in determining the technical condition of 0.95. The introduction of this model into an automated monitoring and diagnostics system will allow assessing the technical condition of electrical equipment in real time with sufficient accuracy.

Download Full-text

Prediction of Stress in Power Transformer Winding Conductors Using Artificial Neural Networks: Hyperparameter Analysis

Energies ◽

10.3390/en14144242 ◽

2021 ◽

Vol 14 (14) ◽

pp. 4242

Author(s):

Fausto Valencia ◽

Hugo Arcos ◽

Franklin Quilumba

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Finite Element ◽

Power Transformer ◽

Network Models ◽

Activation Function ◽

Neural Network Models ◽

Transformer Winding ◽

Artificial Neural ◽

Femm Software

The purpose of this research is the evaluation of artificial neural network models in the prediction of stresses in a 400 MVA power transformer winding conductor caused by the circulation of fault currents. The models were compared considering the training, validation, and test data errors’ behavior. Different combinations of hyperparameters were analyzed based on the variation of architectures, optimizers, and activation functions. The data for the process was created from finite element simulations performed in the FEMM software. The design of the Artificial Neural Network was performed using the Keras framework. As a result, a model with one hidden layer was the best suited architecture for the problem at hand, with the optimizer Adam and the activation function ReLU. The final Artificial Neural Network model predictions were compared with the Finite Element Method results, showing good agreement but with a much shorter solution time.

Download Full-text

Forecasting of Cutting Forces in Dental Adjustment of Ceramic Prostheses Using an Artificial Neural Network

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.152-153.1687 ◽

2010 ◽

Vol 152-153 ◽

pp. 1687-1690

Author(s):

Jian Hui Peng ◽

Xiao Fei Song ◽

Ling Yin

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Cutting Forces ◽

High Speed ◽

Cutting Process ◽

Computer Assisted ◽

Ann Model ◽

Operational Parameters ◽

Artificial Neural

Intraoral adjustment of ceramic prostheses involving cutting process is a central procedure in restorative dentistry because the quality of ceramic prostheses depends on the cutting process. In this paper, an artificial neural network (ANN) model was developed for the first time to forecast the dynamic forces in dental cutting process as functions of clinical operational parameters. The predicted force values were compared with the measured values in in vitro dental cutting of porcelain prostheses obtained using a novel two-degrees-of-freedom computer-assisted testing apparatus with a high-speed dental handpiece and diamond burs. The results indicate that there existed nonlinear relationships between the cutting forces and clinical operational parameters. It is found that the ANN-forecasted forces were in good agreement with the experiment-measured values. This indicates that the established ANN model can provide insights into the force-related process assessment and forecast for clinical dental cutting of ceramic prostheses.

Download Full-text

Applying Artificial Neural Network Model on Investigating the Fiber Diameter of Polybutylene Terephthalate (PBT) Spunbonding Nonwovens: Comparison with Mathematical Statistical Method

Key Engineering Materials ◽

10.4028/www.scientific.net/kem.426-427.692 ◽

2010 ◽

Vol 426-427 ◽

pp. 692-696

Author(s):

Bo Zhao

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Statistical Method ◽

Network Model ◽

Neural Network Model ◽

Artificial Neural Network Model ◽

Fiber Diameter ◽

Computer Assisted ◽

Polybutylene Terephthalate ◽

Artificial Neural

In this paper, two models are founded and introduced to predict the fiber diameter of polybutylene terephthalate spunbonding nonwovens from the spunbonding process parameters. The results indicate the artificial neural network model has good approximation capability and fast convergence rate, and it can provide quantitative predictions of fiber diameter and yield more accurate and stable predictions than the mathematical statistical method. This area of research has great potential in the field of computer assisted design in spunbonding technology.

Download Full-text

Simulating the Filtration Properties of Nonwoven Fabrics: Comparison of Artificial Neural Network, Statistical and Grey Models

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.20-23.1021 ◽

2010 ◽

Vol 20-23 ◽

pp. 1021-1027 ◽

Cited By ~ 1

Author(s):

Ting Chen ◽

Shang Zhen Zhao ◽

Li Li Wu

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Processing Parameters ◽

Computer Assisted ◽

Ann Model ◽

Nonwoven Fabrics ◽

Grey Models ◽

Artificial Neural ◽

Few Data ◽

Filtration Properties

The artificial neural network, statistical and grey models are established for predicting the filtration properties of melt blown nonwoven fabrics from the processing parameters. The results show that the ANN model yields very accurate predictions and a reasonably good ANN model can be achieved with relatively few data points. The statistical model gives satisfactory prediction results for most cases, and the grey model needs to be improved for precise predictions. The results show great perspective of this research in the field of computer assisted design of melt blowing nonwoven technology.

Download Full-text