Dissolved gas analysis is an effective method for diagnosis of power transformer incipient faults. In this work, the health status of the transformer is evaluated using fuzzy logic approach. Fuzzy inference systems namely Mamdani type-1 and Sugeno type-1 available in MATLAB environment are taken to perform the task. The gases in ppm and fault levels are given as inputs to fuzzy inference system and health status of transformer is predicted. The results infer that Sugeno systems offer better prediction capability and good accuracy even for varying conditions of gas concentrations and fault levels.

The detection, measurement, and classification of Partial Discharges (PD) is an important method to determine the quality of insulation systems used in High-Voltage power apparatus and cables. The PD detector patterns contain the characteristics of an insulating system malfunction. Through PD, the problem is found by taking relevant information out of it. In this proposed work, Convolutional Neural Networks (CNN) extract valuable features from a 3-D Phase-resolved Partial Discharge pattern, where their edges are detected using a canny edge detected approach. CNN performance is improved by optimizing the algorithm hyperparameters. Finally, the proposed technique is compared to cutting-edge CNN algorithms from the literature. According to the findings, the proposed method can be utilized to identify possible challenges, attaining an accuracy of 97.3%.

This paper discusses dissolved gas analysis (DGA) methods, which involve the Duval Triangle Method (DTM), Duval Pentagon Method (DPM), Heptagon Graph Method (HGM), and ratio methods, such as the IEC, Rogers, and Doernenburg method, for identifying issues in power transformers through process. A particular method to recognize the type of issue in a power transformer DGA is the Duval triangle. The fault zone is indicated, and its position in a triangle with equal sides is depicted using the proportionate quantities of three fault gases. The most accurate method of evaluating the condition of a power transformer has been shown to be DGA. Six distinct kinds of DGA power transformer problems can be identified using the DPM, a DGA graphical interpretation technique. They provide a review of the advantages and disadvantages of each plan, which may help decide the DGA method that will finish the test analysis most effectively. It averts further harm to that particular transformer by eliminating early transformer breakdowns. Lastly, the results of the study and future studies will be discussed.

In a high-voltage power transformer, one of the most important metrics to measure the effectiveness of an insulating system is partial discharges (PD). However, in real-world settings, several PD sources (single or multiple) with significant interference may hinder the signal’s acquisition, which could result in an erroneous final diagnosis of the equipment’s condition. Convolutional neural networks have outstanding classification performance and robust automatic feature extraction capabilities, making them adept at recognizing patterns. Nevertheless, the collected samples could be more balanced and a significant amount of unlabelled data, which complicates the ability of the existing approaches to provide accurate diagnoses. This study suggests three distinct sections for the transformer defect recognizing technique using unreliable pseudo-labels (TDRUPL), including teacher and student models, to address the abovementioned issues. A unique variational autoencoder first creates synthetic data to equalize the imbalanced samples. Then, external interference immersed in the measured PD signal is removed by employing a bi-directional long short-term memory network and its hyperparameters are tuned by genetic algorithm. Finally, to deal with unlabelled datasets, the instructor model in TDRUPL includes two separate classifiers that help to extract and classify PD characteristics. They also label some unlabelled data as noisy or noise-free samples for convenience. The student model dynamically modifies the conditional and marginal distribution to recognize transformer defects accurately. The proposed algorithm achieves outstanding performance with an accuracy of 98.9

Sm and Zr co-doped BaTiO3 ceramics were investigated for their microstructure and dielectric characteristics. (Ba1 − xSmx)(Ti0.75Zr0.25)O3−δ (BSTZO) with x = 0.02, 0.04, and 0.06 mol V_O^·· = 1.98 × 1021 cm−3, at 773 K) than that of the other compositions. The measured maximum dielectric constant was found to be 1808, 2010, and 1736 for BSTZO pellet with x= 0.02, 0.04, and 0.06 mol

A frequent investigative method for judging the state of insulation in high-voltage equipment is the partial discharge (PD) test. External interference frequently adversely the measured PD signals. The type-2 fuzzy inference system (FIS) tuned with particle swarm optimization (PSO), an artificial intelligence (AI) technology, is used to de-noise the PD signals in this study. By using PSO, upper and lower membership function parameters are tuned automatically. The proposed method's statistical metrics- Compared to the Fast Fourier Transform and Wavelet Transform-are determined, including the signal-to-noise ratio, cross-correlation coefficient, pulse amplitude distortion, mean square error, and noise level reduction. Compared to the genetic algorithm and PSO, the type-2 FIS algorithm performs better than the others in terms of durability and computational cost.

Traditional RRT-based path planning for an unmanned ground vehicle (UGV) is highly influenced by random sample point generation, number of inflexions towards the target and search efficiency to find the shortest path through the obstacle environment. To deal with these challenges, we propose a path planning method as an enhanced bi-directional-rapidly exploring random tree (RRT) * to handle the shortcomings of the traditional algorithms. It uses the artificial potential field (APF) to identify obstacle space and employ non-uniform sample points accordingly towards the goal point. It improves path generation ability and obstacle avoidance, resulting in faster convergence and fewer inflexion points. For managing system dynamics, the generated pathways are additionally refined by the use of Catmull-Rom Spline Interpolation, producing smoother planned trajectories. Simulation shows that our enhanced bi-directional RRT* algorithm outperforms traditional methods to generate the shortest path while avoiding obstacles.

In the present High voltage application system, natural ester oils are being considered as a viable solution to the replacement of Petroleum-based mineral oil due to their promising environmental conditions, like being renewable, biodegradable, not hazards to water, having a low fire point, and having a high flash point. Blended techniques play a crucial role in this research. This paper investigates that vegetable oils such as Groundnut Oil (GO) and Rice Bran Oil (RBO) have low viscosity, low fatty content, and high breakdown strength in nature, so it suggests better alternatives for mineral oils. Blended GO and RBO oils with various concentrations give the desired performance compared to blended mineral oil. Although this research describes the comparison between various concentrations of vegetable oils and experimental measurements of dielectric properties, From the experimental results, it can be concluded that the mixed ratio samples of 60:40 and 70:30 of groundnut and rice bran vegetable oils enhanced critical properties, which proves the better alternative solution to the transformer mineral oil. Further, the optimized results for the system's insulation can be analyzed using the Grey relational analysis (GRA) method.

Titanium dioxide (TiO2) thin films were deposited on glass substrates using Spray Pyrolysis technique, with variations in multiple deposition parameters. The molarity of the precursor was altered within a small range from 0.09 M-0.15 M. The deposition temperature was systematically adjusted from 200 degrees C to 400 degrees C while the ratio between precursor and chelating agent varied between 1:1,1:2 and 1:3. The thickness of TiO2 films were found to be in the range of 216 nm to 14.9 mu m. Structural analysis conducted by XRD confirmed the formation of anatase TiO2 thin films. Optical studies using UV-Visible spectrophotometer determined the absorption and indirect bandgap ranging from 299 nm to 326 nm and 3.08 eV to 3.44 eV respectively. Electrical studies carried out evaluated the leakage currents and DC resistivity for all individual films with the input voltage applied from +/- 0.5 V to +/- 5V. Impedance studies were conducted by varying input voltages from 0.2V-3V for each film, so as to examine the resultant impedance, dielectric constant, dielectric loss, conductivity, admittance and modulus spectra. The obtained results were analysed to optimise the deposition parameters for designing future memristors with specific individual or combined characteristics, such as high ON/OFF, switching speed, endurance and retention.

The proposed work approaches machine learning based Lithium Iron-phosphate (LIP) battery state of charge (SoC) forecasting method for electric vehicles (EVs). The SoC must be accurately estimated to operate safely and steadily for an LIP battery. It is challenging for the user to determine the battery SoC during the charging segment because of the random nature of their charging process. One of the fundamental components of artificial intelligence, machine learning (ML) is rapidly transforming a wide range of fields with its capacity to learn from given data and resolve complex problems. A Grey Wolf optimization (GWO) algorithm is used to modify the most pertinent hyperparameters of the SVR model to enhance the accuracy of the predictions. The database used in the study was gathered from a laboratory setup and includes information on temperature, voltage, current, average voltage, and, average current. Initially, a preprocessing method for data normalization is utilized to improve the data quality and forecast accuracy. It is demonstrated that the accuracy of the suggested model for estimating the SoC of the LIP battery shows less than 3% MSE between the measured and expected states of charge.

Diagnosis of power transformer faults is essential for the stable operation of an electrical power system. Dissolved gas analysis is a prevalent method used for this purpose, which assesses the transformer oil quality and analyses the gases liberated during continuous transformer operation. The proposed work uses a support vector machine (SVM) optimized firefly algorithm for transformer fault diagnosis. This algorithm optimizes the SVM hyperparameters and reduces k-fold loss, improving the model’s accuracy and fault prediction capability. The SVM’s radial basis function (RBF) kernel helps handle the non-linearly separable data. The results showed that the accuracy obtained after incorporating the optimization algorithm is comparatively high, evidencing increased performance of the model in predicting transformer faults. The approach provides a reliable way for quick and early detection of transformers’ inherent faults, facilitating improved maintenance strategies and reduced equipment downtime.

Reprocessing waste things into a functional product gives two-edged benefits, an eco-friendly environment and waste management. Nowadays, Ceramic materials based on calcium phosphates, which include Nanohydroxyapatite (nHAp) and Tri-calcium Phosphates (TCP) were extensively used in biomedical applications for their outstanding biocompatibility and bioactivity properties. In this work, nHAp (pure) and its polymeric composites were prepared through the precipitation method using the coral skeleton as a calcium precursor. Besides, to change the surface functional characteristics, the polymers were utilized as a capping agent. The synthesized samples were examined by spectrographic tools such as XRD, FT-IR, SEM/EDAX, TG-DTA, XPS and HR-TEM to find the phase, functional groups, size, morphology and thermal stability together with phase transition as well as bonding types which present in the prepared samples respectively. In addition, Simulated Body Fluid (SBF) analysis, antimicrobial assay and hemolytic test were intended to test the apatite-forming ability, antimicrobial effectiveness and hemocompatibility of the prepared samples. Therefore, the present findings suggest the possibility of developing novel HAp polymeric nanocomposites that exhibit improved biocompatible, osteoconductive which impersonates the structure of natural human bone. So, that can use for making bone and dental implants using corals skeleton as a precursor at a lesser cost.

(Ba0.9Sm0.1) (Sn0.05Ti0.95) O3 (BSST) ceramic powders were synthesized by solid state reaction route and were pelletized and sintered at 1673 K for 4 h. The BSST thin film was also prepared by electron beam evaporation technique by using the sintered pellets. X-ray diffraction of the BSSTO confirmed that the material has formed in crystalline nature and also in single phase. Raman spectroscopy was used to study the morphology, structure and phase transition behaviour of BSST bulk as well as thin film. The Raman analysis indicated that the film formed at 973 K by electron beam evaporation is less crystalline than that of the bulk BSST due to inadequate substrate temperature.

Uninterrupted power supply to power consumers has increasingly become a global requirement for monitoring and assessing power apparatus's health online and offline for lifetime estimation. Traditionally, extraction of low-level features from aerial images was used for condition monitoring of high-voltage insulators. It is hard to attain insulator defect identification with a complex background. Convolution fusion Network (CFN) with Enhanced Cat Swarm Optimization (ECSO) algorithm for feature selection addresses these issues in this proposed work. CFN can integrate different level features and combine them into wealthy visual features. Electrical insulator photographs captured inside a studio make realistic Overhead Power Distribution Lines (OPDLs). Deep CFN is the right choice to extract in-depth features from the insulators. ECSO can exclusively differentiate the intact and defective insulators and their material types to select optimal features from the dataset. Ensemble multi-class support vector machine (MCSVM) with sigmoidal kernel function used for insulator recognition.

The Electric Field (E-Field) and potential distribution plays a major role in deciding the lifetime of an insulator. Different climate and pollution conditions lead to deterioration and degradation of insulator properties by increasing the electric stress and exceeding the threshold value during long working conditions, leading to failure of the insulator. This article mainly focuses on the reduction of electric stress by using different side positions of the corona ring on a 110 kV AC transmission line composite insulator. The Corona ring plays a significant role in the reduction of electric stress. This study analyses the performance of three different positions of the corona ring: HV and LV side, LV side, and HV side achieving uniform distribution. The obtained results confirm the effectiveness of three different corona ring positions, wherein the HV side highly reduces the electric stress.

This research focuses on classifying the meat quality using a convolution neural network (CNN). The study mainly focused on classifying the meat quality using the microscopic camera without adding chemicals. In this proposed research, the meat is checked in two different ways: using the pH method and image processing using deep learning. Our method has the advantage of double-step verification of the meat by using pH and image processing. The quality of meat prediction by the proposed method is 97% accurate when compared to the conventional way of categorizing meat. The performance of analyzing the quality of meat is further improved by taking more significant number of data set collections with different angle of meat image. It enriches the CNN algorithm performance with tuned hyperparameters.

The proposed work is used to define the type of fault in power transformer using dissolved gas analysis (DGA). It is based on a semi-supervised deep neural network (DNN) architecture and hyperparameter-tuned transfer learning algorithms (DGA). To address the significant issue that the major pseudo-labeling encountered in available data are unlabelled in nature, which causes the impact of semi-supervised learning importance in this proposed work. Second, various demi-supervised learning approaches are used to determine the proposed work's effectiveness. Third, hyperparameters in the DNN are optimized using the Whale Optimization Algorithm (WOA) for different training optimizers. The suggested scenarios are based on 500 lab-gathered samples and literature samples. The word “results” refers to how the initial decomposing material affected the transformer fault's severity and demonstrates the suggested solution's viability. This model has a superior accuracy (97.8 % ) for diagnosing transformer fault types compared to other DGA techniques.

The benzoxazines made from cardanol-aniline (C-a), bisphenol-A-aniline (BA-a), and bisphenol-F-aniline (BF-a) were separately reacting with paraformaldehyde through Mannich condensation reaction. Hybrid polymeric blends of binary and ternary compositions were developed using DGEBA, C-a and BA-a/BF-a matrices. Bio-silica was obtained from rice-husk, which was functionalized using glycidoxypropyltrimethoxysilane, reinforced with combinations of binary and ternary blends of epoxy/benzoxazine resins, and then cured with triethylenetetramine (teta). From thermogravimetric analysis, it was inferred that 100 wt% of bio-silica reinforced hybrid poly(C-a/BA-a/DGEBA) and poly(C-a/BF-a/DGEBA) composite samples possess better thermal stability than that of neat matrices. Among the composites, silica reinforced composites possess a lower dielectric constant than that of other composites. The value of break down voltage of 100 wt% of bio-silica reinforced polymer composites namely, poly(DGEBA-teta), poly(C-a 50wt%/DGEBA 50 wt%), poly(C-a/BA-a/DGEBA) and poly(C-a/BF-a/DGEBA) were observed at 25.93, 31.44, 31.74 and 32.30 kV, respectively. According to the data obtained from different experimental studies, hybrid composites made of epoxy resin and benzoxazine with bio-silica reinforcement possess better performance characteristics and can be considered a better suited material for high voltage insulation applications.

Measurement and analysis of Partial Discharge (PD) patterns have appeared as an emerging field in assessing insulation failure in High Voltage apparatus. This paper uses a PD signal combined with the deep convolution-optimized learning machine classifier (DC-OLMC) to predict the location of water droplets in 11 kV polymer insulators subjected to alternating currents. There are two major confront when applying the proposed algorithm: i) Contamination is a significant issue in PD signal measurement, which causes a reduction in recognition rate (RR), and ii) with minimal computing time, high-level feature extraction and recognition. Traditional condition monitoring methods of insulators concentrated on extracting fewer priority features from the input patterns. In the current work, to address this problem, an Alexnet with Bacterial Foraging Algorithm (BFO) based optimized kernel parameter classifier and Translation Invariant Wavelet Transform (TIWT) is employed to remove interference from PD signals. The analysis demonstrates that the suggested technique, with an identification rate of 99.17%, is considered a valuable tool for locating water droplets in high-voltage insulators.