Breakdown Voltage Prediction Research Articles

Reverse characterization prediction of diamond Schottky barrier power devices using machine learning: Predicting breakdown voltage and Baliga figure of merit

Diamond power devices showed great potential for high-performance and low-loss applications at high frequencies. Schottky power diodes on the diamond have markedly higher threshold reverse operation limits than other power devices, yet they are still far from the ideal limit. Therefore, utilizing the device capabilities across optimizing device parameters is severely time- and resource-intensive considering traditional trial-and-error experimental methodology. On the other hand, data-driven techniques can accelerate device optimization and reduce resource consumption. This work demonstrates machine learning models that predict breakdown voltage in addition to Baliga's power figure of merit (BFOM) for the first time. An experimentally driven dataset is constructed because of its reliability, precision, and potential for generalization. This dataset contains current-voltage characteristics raw data for more than 400 devices fabricated on heteroepitaxial diamond. Four machine learning algorithms are explored and evaluated based on 5-fold cross-validation to prevent overfitting. The extra-tree model shows the highest performance that accurately predicts the breakdown voltage, and the BFOM with R2 equal to 0.924 and 0.911, respectively. Additionally, it showed robust prediction accuracy on unseen data with an 8 % average relative error of 16 devices. Achieving this high accuracy indicates the potential of these models for generalization on various datasets.

Research on the prediction of breakdown voltage of transformer oil based on multi-frequency ultrasound and GWO-RF algorithm

Transformer oil serves as a critical insulating medium within electrical transformers, with breakdown voltage acting as a crucial indicator to assess both the quality of the oil and its insulation characteristics. In this study, 175 groups of transformer oil samples were selected for experimentation. Breakdown voltage was measured, and signals were subsequently collected following the propagation of multi-frequency ultrasonic signals through the oil samples. The study examines the relationship between the amplitude-frequency response, phase-frequency response of multi-frequency ultrasonic acoustic parameters, and breakdown voltage in transformer oil samples. A prediction model for oil sample breakdown voltage was established using the grey wolf optimization algorithm to optimize the random forest algorithm (GWO-RF), with ultrasonic parameters as inputs and transformer oil breakdown voltage as the output. The original 175 datasets were divided into 140 training sets, 20 validation sets, and 15 test sets. The predictive model achieved an average percentage error of 4.04% and a prediction accuracy of 95.96% on the test set, surpassing the performance of the established random forest algorithm (RF) and the sparrow search algorithm optimized random forest algorithm (SSA-RF) for predicting breakdown voltage.

Design Optimization of an Enhanced-Mode GaN HEMT with Hybrid Back Barrier and Breakdown Voltage Prediction Based on Neural Networks

Design Optimization of an Enhanced-Mode GaN HEMT with Hybrid Back Barrier and Breakdown Voltage Prediction Based on Neural Networks

Prediction of DC Breakdown Voltage of Rod–Plate Gaps under Full-Flame Bridging Conditions

Prediction of DC Breakdown Voltage of Rod–Plate Gaps under Full-Flame Bridging Conditions

Prediction of JTE breakdown performance in SiC PiN diode radiation detectors using TCAD augmented machine learning

The design of Junction Termination Extension (JTE) is an important step for meeting the reliability requirement of SiC PiN diode radiation detectors. For early evaluation, Technology Computer Aided Design (TCAD) software is often used to simulate the electronic property of detectors with different JTE parameters. But it is time consuming, which need 1 h or even longer for one case. Here, a TCAD augmented Machine Learning (ML) method based on the fully connected Neural Network (NN) algorithm is proposed to predict the breakdown performance quickly with different parameters of Spatial Modulation (SM) JTE. Utilized ∼5000 datum generated by TCAD simulation, the ML model could be established and achieve good prediction of breakdown voltage and location within a few seconds. As a semi-supervised learning model, its prediction accuracy of breakdown location is higher than 89.4 % and the determination coefficient R2 of breakdown voltage could be up to 0.97 compared with TCAD simulations. Moreover, this model could give the relationship curve of breakdown voltages and doping concentration, which is useful to choose an ideal structure with a wide implantation dose window. Based on this ML prediction model, the design cycle of SiC PiN diode radiation detectors could be reduced significantly.

Prediction of switching impulse breakdown voltage of complex gap based on SVM

Prediction of switching impulse breakdown voltage of complex gap based on SVM

Electrostatic Field for Positive Lightning Impulse Breakdown Voltage in Sphere-to-Plane Air Gaps Using Machine Learning

Breakdown (BD) voltage is significant in high-voltage power electric machines. Currently, BD voltages are mainly predicted by the semi-empirical formula in strongly inhomogeneous electric fields. However, the equation could not be applied for electrodes with weakly inhomogeneous electric fields. In this paper, positive lightning impulse BD voltages are predicted in various sphere-to-plane air gaps using forms of machine learning such as support vector regression (SVR), Bayesian regression (BR) and multilayer perceptron (MLP). Unlike previous studies, a method is also proposed by introducing streamer propagation characteristics as new features and by removing electric field gradients as unnecessary features to find out how to reduce the feature dimension. The streamer propagation characteristics are suggested to reflect the possibility of a discharge process between electrodes. Predicted voltages from machine learning algorithms are compared with the experimental results and calculated voltages from the semi-empirical formula. Firstly, the predictions from each model agreed well with the datasets. New features were observed to be applied for machine learning algorithms and to be as important as known electrostatic features before discharge. Secondly, predicted BD voltages were more accurate than calculated voltages from the semi-empirical equation in strongly inhomogeneous electric fields. Predictions from each model also agreed well with the experimental results in weakly inhomogeneous electric fields. The prediction accuracy of SVR was better than those of BR and MLP. Machine learning algorithms were also shown to be applied for electrodes with a wide range of inhomogeneities, unlike a semi-empirical method. We expect that the suggested features and machine learning algorithms can be used for accurately calculating BD voltages.

Breakdown Voltage Prediction by Utilizing the Behavior of Natural Ester for Transformer Applications

This research investigates the dielectric performance of Natural Ester (NE) using the Partial Differential Equation (PDE) tool and analyzes dielectric performance using fuzzy logic. NE nowadays is found to replace Mineral Oil (MO) due to its extensive dielectric properties. Here, the heat-tolerant Natural Esters Olive oil (NE1), Sunflower oil (NE2), and Ricebran oil (NE3) are subjected to High Voltage AC (HVAC) under different electrodes configurations. The breakdown voltage and leakage current of NE1, NE2, and NE3 under Point-Point (P-P), Sphere-Sphere (S-S), Plane-Plane (PL-PL), and Rod-Rod (R-R) are measured, and survival probability is presented. The electric field distribution is analyzed using the Partial Differential Equation (PDE) tool. NE shows better HVAC with stand capacity under all the electrodes configuration, especially in the S-S shape geometry. The exponential function is developed for the oils under different electrode geometry; NE shows a higher survival probability. Likewise, the most influential dielectric properties such as breakdown voltage, kinematic viscosity, and water content are used to develop a Mamdani-based control system model that combines two variables to produce the surface model. The surface model indicates that the NE subjected for investigation is less susceptible to moisture effect and higher kinematic viscosity. Based on the surface models of PDE and fuzzy, it is concluded that NE possesses a high survival rate since its breakdown voltage does not react to changes in water content. Hence the application of NE in the transformer application is highly safe and possesses extended life.

Sphere Gap Breakdown Voltage Prediction Based on ISSA Optimized BP Neural Network and Effective Electric Field Feature Set

Electric field distribution is a determinant factor of air gap breakdown voltage. This paper defined an effective electric field zone between the sphere gaps for feature extraction, and 29 parameters were extracted from this zone to reflect the electric field distribution. A data mining model was established by back propagation neural network (BPNN) for sphere gap breakdown voltage prediction, taking the electric field features as inputs. The maximum information coefficient (MIC) method was used to select features strongly correlated to the breakdown voltage, and a multi‐strategy improved salp swarm algorithm (ISSA) was applied to optimize the BPNN parameters to enhance the prediction performance. The standard sphere gaps provided in IEC 60052 were taken as training set and test set, which were divided according to the electric field nonuniform coefficients. The breakdown voltage prediction results of test sample sphere gaps demonstrate good agreements with the experimental values. With input features selected by the MIC method, the ISSA optimized BPNN model has a mean absolute percentage error of only 0.83% and a maximum relative error of 5.86%. By constructing the nonlinear mapping relationship between the effective electric field features and the insulation strength, this study provides a new approach to predict air gap breakdown voltage. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

Prediction of Impulse Breakdown Voltage in Dry Air at High Pressure Using Volume-Time Theory

Dry air is promising as an alternative insulation medium for SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> gas in high-voltage equipment. Since its dielectric strength is about one-third of that of SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> gas, one of the most effective ways to apply dry air to high-voltage gas-insulated switchgears (GISs) is to further increase gas pressure. In this article, impulse breakdown strength of dry air at high pressure under quasi-uniform electric field is measured. Furthermore, the probability distribution of breakdown voltage is predicted using the volume-time theory in consideration of initial electron generation probability in gas gap and field electron emission from an electrode. The predicted distribution of breakdown voltage by the volume-time theory is in line with the experimental results.

Genetic algorithm based artificial neural network and partial least squares regression methods to predict of breakdown voltage for transformer oils samples in power industry using ATR-FTIR spectroscopy

The current study proposes a novel analytical method for calculating the breakdown voltage (BV) of transformer oil samples considered as a significant method to assess the safe operation of power industry.Transformer oil samples can be analyzed using the Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy combined with multivariate calibration methods. The partial least squares regression (PLSR) back propagation–artificial neural network (BP-ANN) methods and a genetic algorithm (GA) for variable selection are used to predict and assess breakdown voltage in transformer oil samples from various Iranian transformer oils.As a result, the root mean square error (RMSE) and correlation coefficient for the training and test sets of oil samples are also calculated. In the GA-PLS-R method, the squared correlation coefficient (R2pred) and root mean square prediction error (RMSEP) are 0.9437 and 2.6835, respectively. GA-BP-ANN, on the other hand, had a lower RMSEP value (0.2874) and a higher R2pred function (0.9891).Considering the complexity of transformer oil samples, the performance of GA-BP-ANN has resulted in an efficient approach for predicting breakdown voltage; consequently, it can be effectively used as a new method for quantitative breakdown voltage analysis of samples to evaluate the health of transformer oil..

Intelligent computing and analysis of breakdown voltage of rod-rod long air gap

The discharge characteristics of air gap is one of the important factors to determine the insulation level which is crucial for electrical engineering construction. In order to calculate the 50% breakdown voltage of rod-rod air gap under different meteorological conditions and different gap distances, we take the positive standard switching impulse test data of rod-rod air gap under different meteorological conditions as samples. Based on training set, a breakdown voltage prediction model of rod-rod air gap under different meteorological conditions was established by cuckoo search optimized weighted support vector regression (CS-ω-SVR). The results show that the average absolute percentage error of the test set is 3.6%, which verifies the validity of this computing method. The method can provide reference for the computation of air gap breakdown voltage under different altitudes and extreme meteorological conditions.

Prediction of Transformer Oil Breakdown Voltage with Barriers Using Optimization Techniques

A new procedure to optimally identifying the prediction equation of oil breakdown voltage with the barrier parameters’ effect is presented. The specified equation is built based on the results of experimental works to link the response with the barrier parameters as the inputs for hemisphere-hemisphere electrode gap configuration under AC voltage. The AC HV is applied using HV Transformer Type (PGK HB-100 kV AC) to the high voltage electrode in the presence of a barrier immersed in Diala B insulating oil. The problem is formulated as a nonlinear optimization problem to minimize the error between experimental and estimated breakdown voltages (VBD). Comprehensive comparative analyses are addressed using three recent innovative marine predators, grey wolf, and equilibrium optimization algorithms to reduce the error between the experimental and estimated breakdown voltages. In addition, the experimental results are expressed in two different models via grey and real coding. Finally, simulation results are conducted with statistical indices that show the effectiveness of the proposed models for experimental verification. The total percentage errors of the tested samples between the observed and estimated VBD are 1.207, 1.222, and 1.207 for EO, GWO, MPA, respectively. The marine predator algorithm has the best performance compared with the other two competitive algorithms for grey and actual codes.

Breakdown Voltage Prediction for Sphere and Semispheroid Geometries With Gaussian Process Regression-Based Model Under the Application of Lightning Impulses of Both Polarities

The design of high-voltage (HV) systems is principally dependent on the discharge voltage of their insulation. Sphere geometry and semispheroid geometry are extremely important in HV systems, such as ground rods and gas-insulated substations (GISs). Hence, in this work, a machine learning algorithm is proposed to develop a model to predict the discharge characteristics of air for sphere and semispheroid geometries. Finite element method (FEM) simulations have been performed to extract different electric fields and energy features of air gaps in the range of 5-40 mm under lightning impulses of both polarities. While developing the model, these features along with gap lengths are considered. The features have been used for training a machine learning algorithm based on the Gaussian process regression (GPR) to develop the model. The outcomes received from the model are ratified with measured experimental data. A good comparison between the two establishes the fidelity of the novel model. The proposed methodology is also compared with the other state-of-the-art techniques and found good. Remarkable performance has been acquired for other gap geometries as well.

Deep neural network-based approach for breakdown voltage and specific on-resistance prediction of SOI LDMOS with field plate

Breakdown voltage (BV) and specific on-resistance (R on,sp), are critical indicators to measure the quality of the power device. To improve the device performance, field plate technology has been developed by modifying the electric field distribution. However, the current technology computer-aided design (TCAD) simulation cannot predict the effect of field plate on BV and R on,sp simultaneously, and the operation process is complicated. This paper proposes a deep neural network (DNN)-based model instead of TCAD to predict the effect of field plates on BV and R on,sp of silicon on insulator lateral double diffused metal oxide semiconductor. The experimental results show that, compared with TCAD simulation, the average deviation for BV and R on,sp is 5.06% and 2.55%, respectively. Also, the time for BV prediction is accelerated significantly up by 1.23 × 105. Our DNN model provides a potential direction to predict device performance with higher efficiency.

Electrostatic Field Feature Selection Technique for Breakdown Voltage Prediction of Sphere Gaps Using Support Vector Regression

This article proposes a methodology for air gap breakdown voltage (BV) prediction based on support vector regression (SVR), taking various features extracted from the electrostatic field calculation results as input variables. The genetic algorithm (GA) is applied for feature selection to improve the performance of the SVR model. A case study on sphere gap BV prediction is reported to demonstrate the validity of the proposed technique. This study provides a reference for dielectric strength prediction by artificial intelligence algorithms, thus to guide the optimal design of air insulation structures.

Relative breakdown voltage and energy deposition in the liquid and gas phase of multiphase hydrocarbon plasmas

Pulsed electrical discharges in a gas–liquid mixture deposit energy into both phases. Here, we propose a model to simulate breakdown in multiphase based on experimental data. Furthermore, we estimated breakdown voltage in each phase and then estimated energy deposition in each phase. Discharge in pure liquid showed a highly stochastic nature, having a wide breakdown voltage distribution, while the mean value closely follows a one term power law as a function of gap spacing. When there is external gas injection to the gap, breakdown voltage increased significantly due to charge dissipation on bubble surface. This effect was simulated to predict breakdown voltage in liquid with gas injection at different rates. A multiphase system model was developed to simulate breakdown in the gas–liquid phase. The model is a superposition of power law and Meek criteria physical models for the liquid and gas phases, respectively, with empirically derived coefficients. Energy deposition into each phase was estimated by this model. The gap spacing is the primary factor determining breakdown voltage and energy distribution. In studied conditions, we were able to predict the breakdown voltage and estimate energy deposition into different phases. When the gap and flow rate vary between 2 and 10 mm and flow rate 0–1 LPM, 50%–93% of electrical energy is deposited into the liquid. This model allows for predicting breakdown voltage in a multiphase. Furthermore, it allows for control of the energy distribution among the phases in a multiphase pulsed discharge system.

Research on effectiveness of lightning impulses with different parameters for detecting protrusion defects in GIS

The focus of this paper is on the effectiveness of lightning impulse (LI) with different parameters for detecting protrusion defects in gas insulated switchgear (GIS). The study examines the 50% (U 50 ) breakdown probability for typical protrusion defects on the conductor of a 363 kV class GIS. A quantitative method is developed to predict breakdown voltage, calculate minimum detectable length of defect, and express the effectiveness of the LI test. The results show that there is little difference of U 50 between aperiodic and oscillating LI. Corresponding detecting effectiveness decreases for LI with long wavefront, which is expressed quantitatively by detectable area. The critical defect length detected increases with increased wavefront. This study provides a general method to express the effectiveness of LI with different parameters for detecting protrusion defects and it support the LI test for on-site testing of GIS.

Discharge of Air Gaps During Ground Potential Live-Line Work on Transmission Lines

Ground potential live-line work (GPLW) is a typical type of live-line work on transmission lines. In order to ensure the safety of live-line workers and reduce the consumption of test resources, the discharge of air gaps during GPLW on transmission lines was studied. Firstly, a breakdown voltage prediction method of GPLW air gaps was proposed, which can predict the breakdown voltage of GPLW air gaps. Subsequently, a discharge model of GPLW air gaps was proposed as well, so that the discharge characteristics of GPLW air gaps under different gap distances can be obtained more convenient. The prediction method and the discharge model were verified by switching impulse discharge tests of GPLW air gaps on UHV transmission lines, and the calculated results were close to the test results. Finally, the influence of live-line worker postures on discharge characteristics was studied. The results showed that workers should avoid causing protrusions of body parts in the course of work, which is more conducive to safety.

Computation of breakdown voltage of long rod-plane air gaps in large temperature and humidity range under positive standard switching impulse voltage

The 50% breakdown voltage of the air gap is an important basis for the design of the external insulation of the high voltage power systems, and the atmospheric conditions have a large impact on the breakdown voltage of the air gap. The temperature and humidity of the operating environment of the high voltage power systems have a wide range. The breakdown voltages of the gaps obtained in the laboratory are mostly near the standard atmospheric conditions, and there is no good way to calculate the discharge voltage outside the standard atmospheric conditions. This paper established a 50% positive switching impulse breakdown voltage prediction model based on the cuckoo search algorithm optimized support vector regression for the rod-plane air gap. This calculation model is used to predict the selected prediction data. The predicted results are basically consistent with the experimental results, and the effect is better than the atmospheric correction results. Finally, the method was used to calculate the 50% positive impulse breakdown voltage of the rod-plane air gap under the extreme temperature and humidity conditions in Beijing. This method provides a new idea for obtaining the air gap breakdown voltage under different atmospheric conditions.