Shed configuration is one of the important factors affecting icing characteristics and flashover performance composite insulators. In this paper, non-energized energized glaze tests are conducted on 12 types typical 220 kV insulators with different shed configurations in artificial climate chamber. Combining numerical calculations electrical field distributions, monitoring leakage current, measurement ice-melting water conductivity shooting arc paths, influences ice growth voltages...

Sensor drift is often application-dependent and results in a reduction the overall long-term performance of electronic noses. Even with compensation it remains challenging to transfer these models other application scenarios. In order remedy this deficiency, different/generic chemicals are needed provide general-purpose calibration approach that can be applied wide range article, we investigated method identify based on four criteria (universality, safety, sensibility, differentiation). This...

Gas sensor array (GSA) data usually has high-dimensional features and a small sample size. When classifier is directly used for GSA classification, it prone to overfitting high time cost. The traditional solution perform feature dimensionality reduction before classification. However, selecting suitable method time-consuming laborious, some useful classification may be lost after reduction, especially the weak response low-concentration gases. In this article, we proposed ensemble-based...

Abstract Speckle noise is a kind of ubiquitous in medical image, which will damage the texture structure image and affect analysis by doctors. Therefore, we propose an denoising model based on fuzzy logic, can eliminate speckle well, improve recognition facilitate acquisition information The main work arrangement algorithm to design membership function that traverse preprocess make smooth. Then, mask template 5 × designed definition g‐l calculus, there mainly unknown parameter this template....

The drift of electronic nose (e-nose) sensors is the most challenging problem. An effective calibration method for e-nose proposed in this paper, which based on Projection to Convex Sets (POCS) and Extreme Learning Machine (ELM) method, including an automatic sampling platform sensor model. It realizes a unified universal both recognition regression application solve problem under long-term working conditions. Isopropanol acetone gases that respond were chosen as training Experiments show...

Deep learning models can autonomously learn intricate feature representations directly from raw data, thereby alleviating the burden of engineering and achieving notable performance enhancements across various tasks. However, due to limited sample size gas sensor array data sets, deep are susceptible issues such as overfitting in odor classification This paper proposed a pyramid scattering convolutional network framework (PSCN) based on wavelet transform address above issues. PSCN first...

This study aims to enhance the robustness of electronic nose (e-nose) against noise interference and sensor damage through a biomimetic approach. Inspired by structure biological mechanisms mammalian olfactory system, we have proposed novel model for processing signals from gas-sensitive arrays e-nose. simulates functions epithelium bulb, encoding into spike sequences, incorporates concept wavelet filter banks scattering networks, effectively capturing local variations in signals. Utilizing...

This article presents the development of an artificial olfactory bulb (OB) using electronic nose with thermally modulated metal-oxide sensors. Inspired by animal OBs, our approach employs thermal modulation to replicate spatial encoding patterns glomeruli clusters and subclusters. new enhances classification capabilities traditional noses offers insights for biomimetic olfaction. Molecular receptive range (MRR) analysis confirms that OB effectively mimics glomerular distribution OBs....

With the improvement of medical conditions, doctors can diagnose patients with ultrasound images, but images currently formed always have a certain proportion speckle noise, which will interfere doctor's diagnosis. Therefore, research on removal noise has significance. In this paper, we propose an image denoising model based fuzzy logic. Firstly, assigns weights to by means membership function, obtains required weight table, and then traverses entire image. The role is pre-smoothed, 5x5...

This paper proposes a new way to clear up the fog which softly combines dark channel with color attenuation, namely DCP and CAP. The main task is solve bad performance in sky white bright regions based on problem of incomplete defogging near object foreground A quad-tree search method applied look for atmospheric light A. We can choose pixel average last image block as excessive saturation. guided filter adjust final medium transmission. By experiments, results prove that proposed effective.

Signal processing in electronic nose (e-nose) odor detection involves a series of complex and time-consuming steps, such as noise reduction, filtering, normalization signal preprocessing, well feature generation, selection, dimensionality reduction. When dealing with different gas sensors, these methods might require meticulous adjustments, each sensor need own specific strategy. Such strategies are often tedious overly reliant on intricate engineering, leading to operations poor...

Abstract There are many ways to super-resolution reconstruction of a single image. This method require the association between low-resolution images and high-resolution images, has achieved good results in different applications. article also pursues correlation but we combine with linear networks explore possibilities combining two domains. In order achieve purpose restoring image details, used network, from traditional single-picture super-resolution, using network delay packet loss...