Isolation of the fault dc lines during initial short period after is a major challenge in multi-terminal (MTDC) system based on voltage-source converter (VSC). Existing protection schemes are mainly designed numerical simulations, which lacks theoretical analysis. In this paper, high-frequency equivalent model VSC-based MTDC first proposed for current calculation. Based analysis, it validated that components flow line and then dramatically decay healthy line. Accordingly, novel detection...

When the distance between resonators is low enough for coupling condition to be greater than critical coupling, single resonant peak at load splits form double peaks. This frequency‐splitting phenomenon results in a reduction power transferred. In this study, an adaptive frequency‐tracking control (AFTC) approach based on closed‐loop scheme implemented overcome problem. An improved ant colony algorithm (IACA) was proposed AFTC track maximum point real time. Then, simulations were performed...

Race detection is increasingly popular, both in the academic research and industrial practice. However, there no specialized comprehensive dataset of data race, making it difficult to achieve purpose effectively evaluating race detectors or developing efficient algorithms.

Traditional multilevel inverters in high voltage power applications have some problems associated with their low frequency switching which produces harmonic distortion. To solve the problem, this paper presents a topology of modified cascade source inverter (VSI) different DC sources. Also, VSI using H-bridge is modeled and analyzed circuit DQ transform detail. implement novel Inverter, hybrid modulation strategy presented through theoretical analysis. The simulation experimental results are...

To solve the problem of extreme learning machine (ELM) on-line training with sequential samples, a new algorithm called selective forgetting (SF-ELM) is proposed and applied to chaotic time series prediction. The SF-ELM adopts latest sample weights old samples iteratively insure that influence weakened. output weight determined recursively during procedure according its generalization performance. Numerical experiments on prediction indicate an effective version ELM. In comparison machine,...

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

In order to solve the hidden-layer neuron determination problem of regularized extreme learning machine (RELM) applied chaotic time series prediction, a new algorithm based on Cholesky factorization is proposed. First, an RELM-based prediction model with one constructed and then added in each training step until generalization performance reaches its peak value. Thus, optimal network structure determined. procedure, used calculate output weights RELM. Experiments indicate that can be...

Object-based image analysis (OBIA) provides a better solution for information extraction from high spatial resolution remote sensing image. Currently, selection of scale parameters is often dependent on subjective trial-and-error methods or post-evaluation multi-segmentation, which directly reduces efficiency land cover classification. This paper proposes OBIA classification method combining statistics based adaptive parameter pre-estimation and Mean-shift segmentation. Series object were...

Acoustic Emission method of nondestructive check is based on exertion wave radiation and their registration during fast local material structure reorganization.It used as a means analysis materials, constructions, productions control diagnosis operating time.In the article, it applied to structural health monitoring Wind Turbines.Acoustic emission testing condition system uses data already collected at wind turbine controller.It an effective way monitor turbines for early warning failures...

The paper analyzes the problems and defects of traditional data storage systems SATA put forward a new system for applications embedded high-speed by two steps: First, carry out improved channels on background. Second, established control distribution module through real-time switching mechanism against rate decreased drastically that affect stability issues when write to channels. ķIndirectly detect channel steep drop inspecting cache. 1) Decide whether switch group or not fine block has...

Acoustic Emission (AE) technique can successfully applied for condition monitoring of low speed rotating components such as rolling bearing and gearbox Wind Turbines.This is able to detect very small energy released rates from incipient defect in a early stage.Wide range signal processing methods be apply diagnosing faults fatigues AE spectrums the changes wave forms are significant recognize failures.Condition Fault identification (CMFI) wind turbine health using automated failure detection...

The current mainstream attack algorithms IFGSM and MI-FGSM will carry out global perturbation to the original example when they adversarial attack. F-MIFGSM algorithm is proposed solve problem of poor concealment. This extracts output information convolutional layer neural network, marks feature region through deconvolution network mapping algorithm, uses specific In experiment, 60,000 training sets in MNIST data set are used train self-constructed 10,000 test attacked, generated examples...

Regular researches of system Sn-Pb alloys with use AE are carried out in various crystallization conditions. In the given technique has been used, which allow to divide signals radiated plastic deformation and crack's formation development from each process. metals majority radiation begins middle crystallization's area comes an end at moment termination However a number materials have registered below excess point on cooling curve firm condition. Essential influence small impurity feature...