A combined model based on improved information entropy and vague support vector machine (IVSVM) is introduced into transformer fault diagnosis using dissolved gas analysis in oil (DGA). The method used to obtain the weights of each weight raw data, processed training data corresponding types are inputted (VSVM) classifiers. Firstly, weighted by discretise original from mixed state for subsequent classifier training. Then, set divides events true, false unknown factors, which can optimise...

Abstract Metal tip defect is a typical insulation in gas insulated switchgear (GIS). Simulation research on the under multi‐physical field coupling was conducted order to explore realistic electric distribution. The results of simulation were confirmed by partial discharge experiment. On one hand, influence location and structural parameters distribution investigated established model. other experimental platform built obtain initial voltage this defect. show that presence defects can...

This study presents a combined model based on the exploratory factor analysis (EFA) and least square support vector machine (LSSVM) to predict contamination degree of insulator surface. Firstly, EFA method is utilised reduce numerous influence variables into few variables, which could decrease complexity model. Then, regarding above as new input LSSVM established degree. In order obtain optimal predictive value, non-dominated sorting genetic algorithm II applied optimization parameters. The...

Composite insulators, which use high temperature vulcanized (HTV) silicone rubber as shed material, are widely applied in transmission lines. Ice accumulation on their surfaces may inflict flashover accident and even massive power outage. In this study, a superhydrophobic (SH) coating HTV was fabricated by the sol–gel process combined with plasma jet treatment method. It found that as-prepared SH exhibited prominent superhydrophobicity an excellent self-cleaning property water contact angle...

The intestinal immune system is crucial for protection from pathogenic infection and maintenance of mucosal homeostasis. We studied the microenvironment in a Salmonella enterica serovar Typhimurium mouse model. Intestinal lamina propria macrophages are main effector cells innate resistance to intracellular microbial pathogens. found that S augmented Tim-3 expression on CD4+ T enhanced galectin-9 F4/80+ CD11b+ macrophages. Moreover, promoted activation bactericidal activity via...

Abstract Basin‐type insulator often has small cracks due to stress concentration. The current method cannot accurately reflect the condition of find concentration areas. To solve these problems, a for detecting two‐dimensional plane ( δ 1 and 2 ) within different depth ranges in basin‐type is proposed based on critically refracted longitudinal (LCR) wave. First, acoustoelastic equation characterising relationship between variation LCR wave propagation time was derived. Next, characteristics...

Ladder network synthesis is very significant for winding fault diagnosis within transformer. To synthesize the reversely based on measurable frequency response analysis (FRA) data only applicable way. date iron core and non-tested winding, which result in frequency-dependent components multiple ladders some ranges, are usually not considered reverse process thus precision largely limited. This paper proposes to ladder networks different regions handle this problem. Firstly, sensitivities...

As the scale of power system continues to expand, equipment fault rate gradually increases, which puts forward higher requirements for transformer diagnosis. Through diagnosis technology, faults can be found in advance during operation, measures taken time reduce possibility accidents. It is that composition and content dissolved gas oil are closely related types defects, play an important role prediction operating state In order obtain accuracy, not only oil, but also correlation between...

At present, the most widely used method of winding fault detection is frequency response analysis (FRA). To surmount defects slow convergence and low classification accuracy caused by inappropriate parameter selection support vector machines (SVM), this paper proposes a transformer based on bald eagle search (BES) algorithm to optimize kernel g penalty coefficient C in SVM model, denoted as BES-SVM. The faults shows that BES- model can diagnose more accurately. Compared with traditional...

In order to accurately predict the oil temperature of transformer top layer, based on traditional neural network algorithm, this paper proposes a prediction method VMD (Variational Modal Decomposition) and GRU (Gated Recurrent Unit) network. Traditional methods have less consideration interaction between different trend items in series, problem is time series problem, so historical at previous also needs be considered. response above problems, introduced decompose original sequence into...

To investigate the impact of temperature on internal electric field distribution and oil streamer discharge development under AC-DC composite voltage, dielectric parameters insulating pressboard at different temperatures are measured experimentally. The nonlinearity with is compared analyzed in terms its effect distribution. Additionally, based bipolar carrier transport model drift-diffusion model, influence process simulated studied. results indicate that considering non-linear dependence...

Transformer familial defects are different types, specifications, series and even types of transformers produced by the same manufacturer that have type defect. In this paper, dissolved gas in transformer oil is analyzed using improved SVM (support vector machine) algorithm based on apriori (frequent item set), a defect detection method proposed. Apriori was used to analyze correlation oil, variation content normalized obtain confidence degree various gases under fault so as judge weight...

Dissolved gas analysis (DGA) in oil is an essential approach for transformer defect prediction. Most of the prediction studies use artificial intelligence methods to build individual classifiers. Artificial techniques are highly sensitive data. Transformer data often unbalanced set, which leads supervised learning models that focus more on a larger variety samples, resulting poorer model performance. To address this situation, paper uses synthetic minority oversampling technique (SMOTE)...

Multi-Robot Multi-Target Tracking (MR-MTT) addresses the problem that a swarm of mobile robots actively detect and move to maintain surveillance team dynamic targets, which is fundamental in modern robot system has enormous potential numerous fields. This paper investigates series MR-MTT algorithms, with detailed demonstrations two distributed approaches solving finite iterations. On one hand, novel local algorithm introduced solve multi-robot multi-target assignments within limited...

Oil paper insulation is the main form of large transformers. Partial discharge(PD) in oil will not only damage performance, but also precursor and manifestation deterioration. Weibull distribution parameter an important characteristic PD. In this a method proposed to extract PD based on three-parameter model, then gray wolf optimization algorithm particle swarm are used calculate parameters model. The result shows that two algorithms can get quickly accurately, model effectively reflect...

Transformer mechanical condition assessment methods based on transformer vibration signals have received a lot of attention due to their non-stop, safe and other characteristics. At present, many studies body are amplitude, which has high mistaken judgment rate. the same time, for different loads types transformers, single frequency varies widely, making it difficult reflect effectively. In this paper, energy share signal is calculated in bands according characteristics reduce influence...