Allocation of Potentially Environmentally Hazardous Sections on Pipelines

Iakovleva, Emiliia; Belova, Margarita; Soares, Amílcar

doi:10.3390/geosciences11010003

Cited by 12 publications

(18 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Collecting data and forming training samples. At this stage, the collection, systematization, and marking of data on surface defects of steel workpiece are carried out in a unified form [16].…”

Section: Methodsmentioning

confidence: 99%

Synthetic Data Generation for Steel Defect Detection and Classification Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

The paper presents a methodology for training neural networks for vision tasks on synthesized data on the example of steel defect recognition in automated production control systems. The article describes the process of dataset procedural generation of steel slab defects with a symmetrical distribution. The results of training two neural networks Unet and Xception on a generated data grid and testing them on real data are presented. The performance of these neural networks was assessed using real data from the Severstal: Steel Defect Detection set. In both cases, the neural networks showed good results in the classification and segmentation of surface defects of steel workpieces in the image. Dice score on synthetic data reaches 0.62, and accuracy—0.81.

show abstract

Section: Methodsmentioning

confidence: 99%

Synthetic Data Generation for Steel Defect Detection and Classification Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

show abstract

“…There are various models of nonlinear electrical loads, which can generally be classified into simulation models based on models of semiconductor elements being a part of them [14][15][16]; in the form of current sources at harmonic frequencies [17][18][19]; in the form of volt-ampere characteristics obtained by approximation according to various equations [20][21][22]; in the form of nonlinear impedances [23][24][25]; based on time-domain, frequency-domain, or time-frequency modeling approaches [26][27][28][29].…”

Section: Introductionmentioning

confidence: 99%

Computation of Nonlinear Load Harmonic Currents in the Presence of External Distortions

et al. 2022

View full text Add to dashboard Cite

This paper deals with the issues of computation of the nonlinear load harmonic currents in the presence of external distortions based on the real measurements with help of passive harmonic. Such values are necessary when modeling an equivalent nonlinear load as current sources in the presence of external distortions. The passive filter allows to separate external and internal distortions, which is necessary when determining the harmonic currents of a single consumer. The influence of various parameters on the computation of harmonic currents of a nonlinear load, such as the parameters of a passive harmonic filter, the impedance of the power supply transformer, the load parameters of the consumer, taking into account the harmonic generation from the grid side, and from the consumer side, is considered. It is shown that an external source of distortion has practically no effect on the error in estimating the harmonic current magnitudes of a nonlinear load. The obtained simulation results were confirmed experimentally on a laboratory test bench. Recommendations for the selection of passive harmonic filter parameters have been developed to minimize the error in determining the harmonic current magnitudes of a nonlinear load.

show abstract

“…The next fundamental stage in the development of methods for identifying distortion sources is the works based on the analysis of the Norton or Thevenin equivalent circuits [22]- [28]. When implementing this method, the equivalent conductivity of each consumer is determined, which is defined with significant assumptions and can have a significant error on the accuracy of the methods.…”

Section: Introductionmentioning

confidence: 99%

An assessment of the share contributions of distortion sources for various load parameters

Skamyin

Shklyarskiy

Dobush

et al. 2022

IJPEDS

View full text Add to dashboard Cite

The method for assessing the contributions of distortion sources based on measuring consumer currents and calculating their projections onto the supply current vector is considered in the paper. Determination of contributions is carried out on the basis of the developed model of an industrial enterprise in the food industry in MATLAB Simulink software. This study presents various cases of simulation, including variable parameters of linear and non-linear consumer load, changes in parameters of external distortion sources and passive harmonic filters. It is shown that the considered method gives correct results in the absence of external distortions in the electrical grid. The considered criteria for the share contributions make it possible to estimate the most efficient place for installing a passive filter in the absence of external distortions. An indicator for evaluating external distortions has also been developed based on calculating the projection of the harmonic system current onto the harmonic current of the shunt filter at the considered frequency.

show abstract

Allocation of Potentially Environmentally Hazardous Sections on Pipelines

Cited by 12 publications

References 35 publications

Synthetic Data Generation for Steel Defect Detection and Classification Using Deep Learning

Synthetic Data Generation for Steel Defect Detection and Classification Using Deep Learning

Computation of Nonlinear Load Harmonic Currents in the Presence of External Distortions

An assessment of the share contributions of distortion sources for various load parameters

Contact Info

Product

Resources

About