Search

Diagnosis of the aging stage of oil-paper insulation is an important technical means for preventing aging failure of oil-paper insulation equipment. This article proposes a Raman spectroscopy diagnostic model to assess the aging status of transformer oilpaper insulation by analyzing the Raman spectra of the insulating oil. Oil-paper samples in different aging stages were obtained from electrical and thermal accelerated aging tests in the lab. The Raman spectra of insulating oil were collected in the range from 400 to 3000 cm−1, which covers the Raman signals of the aging products of oilpaper insulation. Spectral peak contour features were extracted based on Voigt fitting of the Raman spectra and studied using the grid search optimized one-against-one multi-class support vector machine (OAO MCSVM) for diagnosis of the aging stage. The results show that the diagnostic accuracy of the diagnostic model based on the spectral peak contour features was 92% (46/50). Spectral peak contour features demonstrate an ability to reflect the aging process, and the proposed diagnostic method shows potential for monitoring oil-paper insulation.

In this paper, a method of aging diagnosis under different oil-paper ratios is proposed by investigating the accelerated thermal aging data with different oil-paper ratios. Aged oil-paper insulation samples with oil-paper ratios of 10:1, 15:1 and 20:1 are obtained by accelerated thermal aging tests. The Raman spectra of the transformer oil samples are measured by a Raman spectrometer, and screened by random forest algorithm to extract the features. A modified model is established to correcting the features of the aged oil samples under the conditions of 15:1 and 20:1 oil-paper ratio to features under the condition of 10:1 oil-paper ratio. An grid search optimized support vector machine (SVM) classification model is established, and verified by the corrected features. The experimental results show that the proposed method can effectively identify the aging of oil-paper insulation with different oil-paper ratios.

The application of surface-enhanced Raman scattering (SERS) to detect methanol dissolved in transformer oil is one of the most promising online monitoring techniques for evaluating the transformer ...

It is of great significance to monitor internal faults of transformer and ensure its normal and safe operation. Usually, acetylene (C2H2) is considered as the characteristic gas caused by discharge failure in transformer. Using gas sensor technology to analyze dissolved gases in transformer oil is an important strategy. Recently, metal oxide semiconductor (MOS) heterojunctions with tailored microstructures have been developed to fabricate high quality gas sensors for gas detection. In this paper, hierarchical flower-like NiO/ZnO heterostructures assembled with 2D nanosheets have been synthesized by a facile hydrothermal method and calcination process. Noticeably, the introduction of different contents of NiO (3.0 at%, 5.0 at% and 10.0 at%) leads to different assembly manners of the building nanosheets into hierarchical flower-like structures, thus affecting the gas performances. Given this, a variety of microscopic characterization methods were used to observe and compare the differences in the structures and morphologies of the composites. Through the gas sensing test, it was found that the 5.0 at% NiO-modified ZnO based sensor exhibited superior sensing performances to C2H2 compared with that of others. The enhanced properties may be attributed to the formation of p-n heterojunctions as well as high porosity of the nanosheets. This promising approach is versatile for the applications of high-quality gas sensors.

This study explores the method of applying surface enhanced Raman scattering (SERS) in diagnosing the aging stage of oil-paper insulation equipment in the power system, which provides access to the on-line mornitoring of electrical equipment. Oil-paper samples in different aging stages were obtained by accelerated aging experiments. Effective Raman signals of insulating oil were collected by using the self-made hexagonal sliver nano-plates SERS substrates and the self-assembled Raman detection platform. For the SERS data of oil-paper insulation samples, kernel entropy component analysis (KECA) was applied to extract the Raman spectral features and support vector machine (SVM) was used to recognize the the aging stage of oil-paper insulation. The results demonstrate that the KECA-SVM method exhibits a good diagnostic capacity and the recognition accuracy of the propsed method reached 81.43 % (57/70). In summary, a new spectral method is proposed to diagnose the aging stage of oil-paper insulation, which lays a foundation for the actual diagnosis of the aging condition in running oil-paper insulation equipment by Raman spectroscopy.

Rapid identification of aging state of oil-paper insulation is of great significance to the operation safety of power transformers. Raman spectroscopy can rapidly analyze the aging characteristic information dissolved in oil, and it is an effective means for the aging diagnosis of oil-paper insulation. In this paper, Multiple screening KNN Algorithms for Raman spectroscopy analysis of aging oil-paper insulation samples is presented. A large number of aging oil samples were obtained by accelerated thermal aging test. According to the aging days, the samples were divided into 12 categories, and 230 Raman spectra were obtained by Raman spectroscopy. The KNN algorithm is used for classification and regression of Raman Spectra of test samples by Pearson correlation coefficient. Then, based on the traditional KNN algorithm, a multi-screening KNN is proposed according to the actual situation of the aging process of insulating oil Raman spectrum. The prediction of Multiple screening KNN Algorithms accuracy of classification reaches 87.92%, and the RMSE of regression reached 54.28.

In this paper, a method based on Raman spectroscopy to diagnose the aging degree of oil-paper insulation is discussed. According to IEEE guidelines, oil-paper insulation samples were prepared in the laboratory, and the degree of polymerization values of the samples were measured to obtain the exact aging states. Raman detections of the samples were carried out based on the Raman experimental platform. T-S fuzzy neural network was used to analyze the Raman spectrum data of insulating oil. By using the experimental data to train the diagnosis model, constantly modifying the membership function, and mining the internal mathematical relationship between the Raman spectrum characteristics of oil paper insulation and the degree of polymerization of insulating paper, this paper managed to use the Raman spectrum data of insulating oil to predict the aging degree of oil paper insulation. Finally, the deviations between the predicted and the measured degree of polymerization values of the five test samples in this experiment are both less than 150.

We demonstrate the utility of cavity-enhanced Raman spectroscopy (CERS) as a unique multigas analysis tool for power transformer diagnosis. For this purpose, improvements have been added to our recently introduced CERS apparatus. Based on optical feedback frequency-locking, laser radiation is coupled into a high-finesse optical cavity, thus resulting in huge intracavity laser power. With 20 s exposure time, ppm-level gas sensing at 1 bar total pressure is achieved, including carbon dioxide (CO2), carbon monoxide (CO), hydrogen (H2), methane (CH4), ethane (C2H6), ethylene (C2H4), acetylene (C2H2), nitrogen (N2), and oxygen (O2). By using the internal standard gas (sulfur hexafluoride, SF6), the quantification of multigas with high accuracy is also realized, which is confirmed by the measurement of calibration gases. For fault diagnosis, transformer oil is sampled from a 110 kV power transformer in service. Dissolved gases are extracted and analyzed by the CERS apparatus. Then the transformer is diagnosed according to the measurement results. CERS has the ability to analyze multigas with high selectivity, sensitivity, and accuracy, it has great potential in gas sensing fields.

As one of the most typical fault characteristic gases in transformer oil, the content of H2 is an important criterion for judging the operation state of power transformers; while online monitoring of such species through gas sensing technology could effectively detect latent faults and avoid accidents. Zinc oxide is a popular sensing material in recent years with limited sensitivity, selectivity and stability for gas detection, while zeolite imidazolate framework-8 (ZIF-8), one kind of metal–organic frameworks (MOFs) materials, possesses adjustable porosity, huge specific surface areas and good thermal stability, so it could perfectly remedy those defects. This paper studied gas-sensing properties of ZnO rods encapsulated by ZIF-8 (ZnO@ZIF-8) for H2. The morphologies, structures, and composition were characterized in detail by means of X-ray diffraction, scanning electron microscope, transmission electron microscope, X-ray photoelectron spectroscopy and energy disperse spectroscopy. H2 gas sensing properties of pure ZnO and ZnO@ZIF-8 composites were implemented based on gas sensing platform. Results suggested that the ZnO@ZIF-8 showed better sensitivity, selectivity and stability than pure ZnO rods, and meanwhile possessed lower optimum operating temperature upon H2 detection. Meanwhile the gas-sensing mechanism was analyzed comprehensively from three aspects of sensitivity, selectivity and stability. Our aim of this work is to propose a novel core–shell material with enhanced performance for H2 sensing and provide a new idea for developing high-performance materials to detect fault characteristic gases.

The CLRS detection characteristics of acetone dissolved in transformer oil were analyzed. Raman spectral peak at 780 cm–1 was used as the characteristic spectral peak for qualitative and quantitative analyses. The effect of the detection depth and the temperature was investigated in order to obtain good Raman signals. The optimal detection depth and temperature were set as 3 mm and room temperature. A quantitative model relation between concentration and the Raman peak intensity ratio I780/I893 was constructed via the least-squares method. The results demonstrated that CLRS can quantitatively detect the concentration of acetone in transformer oil and CLRS has potential as a useful alternative for accelerating the in-situ analysis of the concentration of acetone in transformer oil.

Furfural concentration is an important indicator for the aging diagnosis of internal paper insulation in power transformers. Surface-enhanced Raman scattering (SERS) could directly perform rapid in-situ detection without complex preprocessing steps, which makes SERS suitable for online monitoring. The application of SERS in the detection of furfural concentration in transformer oil was investigated. First, sliver nano-bulks (AgNBs) were synthesized successfully on the copper foil surface by the galvanic displacement. The electric field distributions of AgNBs were analyzed theoretically using the COMSOL software, which indicated that the AgNBs were suitable for SERS application. Then, AgNBs were used as the SERS substrate to detect standard transformer oil samples with different furfural concentrations. As a result, the Raman spectral peak at 1571 cm−1 was selected as the characteristic SERS spectral peak of furfural dissolved in transformer oil for qualitative and quantitative analyses. Subsequently, the Raman spectral peak ratio I1571/I1597 was used as the furfural function in oil to construct the quantitative analysis model using the least squares method. Results showed that SERS method could be used to analyze the furfural dissolved in transformer oil quantitatively and effectively. With further improvements of the detection limit, the SERS method will provide a useful approach to achieving rapid in-situ analysis for furfural dissolved in transformer oil.

As a typical failure characteristic gas dissolved in transformer oil, C2H2 has a great significance to evaluate the running state of transformer by monitoring its content. Previous reports mainly focus on organizing WO3 nanoscale building blocks into well-defined 3D nanostructures. However, little attention has been paid to different 1D building blocks on the impact of gas sensing properties. In this work, nanorod and nanowire-assembled WO3 porous networks were controllably synthesized, which were applied to C2H2 detection. In comparison, sensor based on the latter exhibited superior performance than the former one, which seems indicate the significance of electronic transport in gas sensing process. It is proposed that the former possesses numerous nanojunctions between nanorods, while the latter has features with larger capping area and less junction density.

The use of gas sensors to monitor gases dissolved in transformer oil is of great value to prevent safety issues of power grid. As a part of developing high performance gas sensor, the deeply sensing mechanism in material aspect study has great guiding significance. In this paper, a new surface adsorption model is proposed which provides an approach to achieve the improvement of the gas-sensing performance of Co-doped ZnO sensor theoretically. Utilizing first-principle density functional theory, the adsorption properties of six main characteristic fault gases (C2H2, CO2, CH4, C2H4, CO, and H2) on Co-doped ZnO (0 0 0 1) surface were systemically investigated which indicated that H2 and CH4 molecules were physically adsorbed, whereas CO, CO2, C2H2, and C2H4 were chemisorbed. In addition, using the as-prepared Co-doped ZnO material, the gas-sensing response for these fault gases were experimentally measured, which matches well with the theoretical analysis. This work reveals the gas-sensing performance of Co-doped ZnO composites in microscopic. It lays the foundation for the preparation of gas sensors to serve for monitoring the condition of power transformers.

In the development of on-line monitoring technique for transformer aging status, detecting concentration of methanol dissolved in transformer oil by the method of surface enhanced Raman spectroscopy (SERS) is a novel and promising technique. By use of Raman spectroscopy technique, we can improve various shortcomings of traditional detection technology. In this paper, the concentration of methanol dissolved in 2 mL/L methanol-oil solution was effectively detected by surface enhanced Raman spectroscopy. In the first, Raman spectra of single methanol molecule and its SERS were simulated to analyze the Raman peaks of methanol. A SERS substrate was prepared with selective enhancement. Based on prepared SERS substrate, concentration of methanol dissolved in transformer oil with content of 2 mL/L have been effectively detected. This helped us to further study the lower concentration of methanol in oil and make quantitative analysis.

How to accurately detect the concentration of furfural in transformer oil plays an important role in the analysis of the aging state of the transformer. As a fast and non-destructive detection method, Raman spectroscopy is the key to improving the sensitivity of Raman detection. In this paper, based on the extraction technique, the deionized water and methanol are used as extractants to carry out the comparative experiment of furfural Raman spectroscopy in oil. Firstly, the deionized water and methanol extract of furfural in oil are prepared and obtained. The Raman characteristic peaks of furfural are 1372 cm-1 and 1411 cm-1 and 1675 cm-1 by Raman detection, and qualitative analysis is discussed. The experimental results show that deionized water as an extractant has certain feasibility for Raman detection, and it has the lower limit of detection for Raman detection compared with methanol as extractant, which makes a certain contribution in order to improve the sensitivity of furfural Raman detection in transformer oil.

The water content in oil is closely related to the deterioration performance of an insulation system, and accurate prediction of water content in oil is important for the stability and security level of power systems. A novel method of measuring water content in transformer oil using multi frequency ultrasonic with a back propagation neural network that was optimized by principal component analysis and genetic algorithm (PCA-GA-BPNN), is reported in this paper. 160 oil samples of different water content were investigated using the multi frequency ultrasonic detection technology. Then the multi frequency ultrasonic data were preprocessed using principal component analysis (PCA), which was implemented to obtain main principal components containing 95% of original information. After that, a genetic algorithm (GA) was incorporated to optimize the parameters for a back propagation neural network (BPNN), including the weight and threshold. Finally, the BPNN model with the optimized parameters was trained with a random 150 sets of pretreatment data, and the generalization ability of the model was tested with the remaining 10 sets. The mean squared error of the test sets was 8.65 &times, 10&minus, 5, with a correlation coefficient of 0.98. Results show that the developed PCA-GA-BPNN model is robust and enables accurate prediction of a water content in transformer oil using multi frequency ultrasonic technology.

Accurate detection of dissolved aging features in transformer oil is the key to judging the aging degree of oil-paper insulation. In this work, in order to realize in situ detection of furfural dissolved in transformer oil, silver nanoparticles were self-assembled on the surface of gold film with P-aminophenylthiophenol (PATP) as a coupling agent. Rhodamine-6G (R6G) was used as the probe molecule to test the enhancement effect. By optimizing the molecular concentration, molecular deposition time, and silver sol deposition time of PATP, the nanoparticles were made more uniform and compact, and an enhanced substrate with rich hot spots was obtained. The optimum substrate was developed, and surface-enhanced Raman spectroscopy (SERS) detection of trace furfural dissolved in transformer oil was realized. The results showed that the substrate prepared under the conditions of 0.1 mol/L PATP, 5 h deposition in PATP and 12 h immersion in silver sol, had the best reinforcement effect (that is, uniform and compact particle arrangement and no particle clusters). By use of this substrate, the minimum detectable concentration of furfural in transformer oil was about 1.06 mg/L, which provides a new method for fast and nondestructive detection of transformer aging diagnosis.

Furfural (Furan-2-carbaldehyde) dissolved in transformer oil is regarded as an important mark of the thermal and mechanical degradation in oil-paper insulation. Confocal laser Raman spectroscopy (CLRS) is an effective means of detecting trace liquid and has been widely applied in many fields. Using CLRS to detect furfural concentration can overcome many disadvantages of traditional detection methods and accelerate testing. Thus, the application of CLRS in the detection of furfural concentration was investigated in this research. A structure model of a furfural molecule was constructed with Gaussian 09W software. Then, the Raman spectrum of this molecule was analyzed theoretically. Meanwhile, a set of liquid test platforms was set up based on basic CLRS principles. Subsequently, standard transformer oil samples were tested by utilizing extraction technology together with the CLRS liquid test platform. As a result, the peak at 1677 cm−1 was selected as the Raman characteristic peak. The detection limit reached 0.10 mg/L. In addition, a quantitative analysis model was constructed between the concentrations and the Raman characteristic peak areas by applying external standard method and least square method. Finally, six types of oil samples with different furfural concentrations were detected; the test results were compared with those obtained through high-performance liquid chromatography. Comparison results showed that CLRS with extraction can quantitatively and effectively detect the furfural concentration in transformer oil (with a maximum error less than 11.7%). Therefore, a new detection tool is presented for assessing the degradation of insulating papers in power transformers.

SnO2 semiconductor gas sensor as the most commonly sensors are used to check the main fault characteristic of dissolved gases in transformer oil. In current study, SnO2 rods and spheres were synthesized by the hydrothermal routes with tin salts under different conditions. We investigated the intermediate reaction on the formation of SnO2 with different morphologies and proposed their possible formation mechanisms. The structures and morphologies of this hierarchical architecture were characterized in detail by means of powder X-ray diffraction, field-emission scanning electron microscopy. Moreover, the obtained SnO2 rods are found to show the better sensing performances than the other structure with higher gas response during the response-recovery time, lower working temperature, and more sensitive response in the same concentration to carbon monoxide. Our results also illustrate that gas sensing properties of SnO2 can be significantly improved by varying the morphologies of the structures. The results may hold substantial promise in power system gas-sensing applications.

Power transformer is one of the critical and expensive apparatus in high voltage power system. Hence, using highly efficient gas sensors to real-time monitor the fault characteristic gases dissolved in transformer oil is in pressing need to ensure the smooth functionalization of the power system. Till date, as a semiconductor metal oxide, zinc oxide (ZnO) is considered as the promising resistive-type gas sensing material. However, the elevated operating temperature, slow response, poor selectivity and stability limit its extensive applications in the field of dissolved gases monitoring. In this respect, rigorous efforts have been made to offset the above-mentioned shortcomings by multiple strategies. In this review, we first introduce the various ZnO hierarchical structures which possess high surface areas and less aggregation, as well as their corresponding gas sensing performances. Then, the primary parameters (sensitivity, selectivity and stability) which affect the performances of ZnO hierarchical structures based gas sensors are discussed in detail. Much more attention is particularly paid to the improvement strategies of enhancing these parameters, mainly including surface modification, additive doping and ultraviolet (UV) light activation. We finally review gas sensing mechanism of ZnO hierarchical structure based gas sensor. Such a detailed study may open up an avenue to fabricate sensor which achieve high sensitivity, good selectivity and long-term stability, making it a promising candidate for transformer oil monitor.

As an important arc discharge characteristic gas dissolved in power transformer, acetylene $(\mathrm {C}{}_{2}\mathrm {H}_{2})$ plays a quite significant role in transformer fault diagnosis. Current study indicates that shape control of 3D hierarchical flower-like ZnO structures used for $\mathrm {C}{}_{2}\mathrm {H}_{2}$ detection remains a challenge. In this work, taper-flower and bundle-flower ZnO nanostructures were synthesized with assistance of active agents in a hydrothermal method, and the possible growth mechanisms were proposed. As for $\mathrm {C}{}_{2}\mathrm {H}_{2}$ detection, sensor based on bundle-flower not only has a lower optimum operating temperature, but also exhibits higher sensitivity, better selectivity and stability than these of the taper-flower one. The results suggest that the bundle-flower ZnO provides an ideal platform for gas sensor applications.

The accurate detection of trace fault characteristic gases dissolved in transformer oil is the key to achieve transformers’ operation status diagnosis. In this paper, we designed and established a detection platform based on fiber enhanced Raman spectroscopy (FERS), which uses a 1 m long hollow-core photonic crystal fiber (HC-PCF) as the sample container, thus simultaneously increasing the effective interaction path length of light-analyte and improving the photon collection efficiency. The simultaneous Raman spectroscopy detection of 7 kinds of mixed fault characteristic gases, such as H 2 , CO, CO 2 , CH 4 , C 2 H 2 , C 2 H 4 and C 2 H 6 , was achieved, and the characteristic Raman shift lines were respectively identified as 4156, 2143, 1388, 2914, 1977, 1344, 2954 cm-1. Taken H 2 as an example, the measured Raman signals shows excellent relationship with the sample concentration. The experiment results prove that FERS provides a very promising method for highly selective and sensitive gas analysis in transformers’ operation status diagnosis.

Oil is an important liquid insulating medium in oil-immersed transformer. The water content in oil is closely related to the deterioration of the performance of the insulation system. 160 oil samples that were collected from various transformers, 150 of which were set as training sets and 10 forecast sets. Input variables to the model is 242-dimension multi-frequency ultrasonic testing data, and the output of the model is the water content of the oil sample. Through the experimental method to determine the hidden layer consists of 11 neurons, the nonlinear mapping relationship. Genetic algorithm (GA) was applied to optimize the BP neural network connection weights and threshold of every layer. The results show that the correlation coefficient of water content of oil prediction model is 0.97207. The average absolute percentage error MAPE of the proposed GA-BPNN model is 9.4%. All results provide a new online detection method for water content in transformer oil.

Dissolved gas analysis (DGA) is an efficient method to diagnose faults of oil-immersed power transformer in the early stages. Tin oxide $(\mathrm {S}\mathrm {n}\mathrm {O}_{2})$ has been extensively studied as a gas sensing material since it is well known to be effective in sensitively detecting various gases. However, the sensitivity of gas sensor will be deteriorated seriously when the humidity of working detecting environment changes, which makes response curve fluctuate and drifts distinctively. As humidity has a huge impact on the detection accuracy, it becomes one of the most important considerations in the practical use of sensors. In this paper, three gas sensors based on $\mathrm {S}\mathrm {n}\mathrm {O}_{2}, \mathrm {P}\mathrm {d}/\mathrm {S}\mathrm {n}\mathrm {O}_{2}$ and $\mathrm {Z}\mathrm {n}/\mathrm {S}\mathrm {n}\mathrm {O}_{2}$ sensing materials were synthesized by hydrothermal method and characterized by X-ray diffractometry (XRD), scanning electron microscopy (SEM) and Energy-dispersive X-ray spectroscopy (EDS). Meanwhile, the performance of the gas-sensing under different environmental humidity was tested. The results show that metal doped gas sensors are better sensitivity and stability than that of the pure $\mathrm {S}\mathrm {n}\mathrm {O}_{2}$ gas sensor. Moreover, with the increase of humidity the detection sensitivity of $\mathrm {P}\mathrm {d}/\mathrm {S}\mathrm {n}\mathrm {O}_{2}$ and $\mathrm {Z}\mathrm {n}/\mathrm {S}\mathrm {n}\mathrm {O}_{2}$ sensors emerge a phenomenon of increasing first and decreasing then. Ultimately, based on the sensing properties of pure, Pd and Zn-doped gas sensors, the possible sensing mechanism for them in humid atmosphere was investigated, which will make contribution to the application of $\mathrm {S}\mathrm {n}\mathrm {O}_{2}$-based gas sensors in the online monitoring of dissolved gases in transformer oil.

As a new mark of internal insulation degradation, the analysis of methyl acetate in transformer oil has attracted considerable attention. Confocal Laser Raman spectroscopy (CLRS) could accomplish the non-destructive and non-contact detection for the liquid samples, which is more suitable for online monitoring and in-situ detection. In this study, the feasibility of CLRS method applied in the analysis of methyl acetate dissolved in transformer oil has been verified. The molecule structure model of methyl acetate has been constructed by Gauss View software, and its Raman spectrum was analyzed theoretically. Then, different oil samples containing methyl acetate was detected by the CLRS test platform. In order to perform qualitative and quantitative analysis, the Raman spectral peak at 628 cm-1 was selected as characteristic Raman peak. Moreover, the quantitative analysis method was calculated by the least square method and internal standard method. CLRS method can be applied in the qualitative and quantitative analysis of methyl acetate in transformer oil, which provide a useful approach for online monitoring and in-situ analysis of methyl acetate in transformer oil.

Acetylene (C 2 H 2 ) is one of the most important and typical characteristic fault gas dissolved in power transformer oil, and on-line monitoring its concentration could reflect the operational state of transformer timely, which could prevent safety issues of power grid effectively. Considering the traditional zinc oxide (ZnO) gas sensor exhibiting some shortcomings, such as higher operating temperature, lower response, poor selectivity, and so on. The paper proposes the ZnO-based (pure ZnO, cobalt-doped ZnO) gas sensors to contrastively study. The characterization results of nanomaterials (XRD, FESEM, HRTEM) show that the as-grown products possess the good crystal structure and high purity. The gas-sensing experiment results on optimum working temperature (Co-ZnO: 160 °C, pure ZnO: 180 °C), response-recovery time (Co-ZnO: 30–33 s, pure ZnO: 35–51 s), sensitivity (Co-ZnO: 9.886, pure ZnO: 1.971; 200 ppm C 2 H 2 ), the limit of detection (3 ppm, Co-ZnO), stability (20, 50, 100, 200, 400 ppm C 2 H 2 ), and selectivity (H 2 , CH 4 , C 2 H 2 , C 2 H 4 ; 50, 100, 200 ppm) are obtained. They correspond to the theoretically analysis which based on the DFT-D calculations. The result of gas-sensing properties indicates that the dopant of Co is beneficial to reduce the optimum working temperature and response-recovery time significantly. In addition, it enhances the sensitivity of gas sensor evidently and process a good stability as well as selectivity. The orbital hybridization occurred during C 2 H 2 adsorption, which promotes the electron transfer and conductivity changes. The results lay a solid foundation for the preparation of new ZnO-based gas sensor that serve for the transformer condition monitoring and fault diagnosis.

Carbon monoxide (CO) is one of the most important fault characteristic gases dissolved in power transformer oil. With the advantages of high sensitivity and accuracy, long-term stability, and short detection time, photoacoustic spectroscopy (PAS) has been proven to be one promising sensing technology for trace gas recognition. In this investigation, a tunable PAS experimental system based on a distributed-feedback (DFB) diode laser was proposed for recognizing dissolved CO in transformer oil. The molecular spectral line of CO gas detection was selected at 1.567 μm in the whole experiment. Relationships between the photoacoustic (PA) signal and gas pressure, temperature, laser power, and CO gas concentration were measured and discussed in detail, respectively. Finally, based on the least square regression theory, a novel quantitative identification method for CO gas detection with the PAS experimental system was proposed. And a comparative research about the gas detection performances performed by the PAS system and gas chromatography (GC) measurement was presented. All results lay a solid foundation for exploring a portable and tunable CO gas PAS detection device for practical application in future.

The detection of furfural in transformer oil through surface enhanced Raman spectroscopy (SERS) is one of the most promising online monitoring techniques in the process of transformer aging. In this work, the Raman of individual furfural molecules and SERS of furfural-Mx (M = Ag, Au, Cu) complexes are investigated through density functional theory (DFT). In the Raman spectrum of individual furfural molecules, the vibration mode of each Raman peak is figured out, and the deviation from experimental data is analyzed by surface charge distribution. In the SERS of furfural-Mx complexes, the influence of atom number and species on SERS chemical enhancement factors (EFs) are studied, and are further analyzed by charge transfer effect. Our studies strengthen the understanding of charge transfer effect in the SERS of furfural molecules, which is important in the online monitoring of the transformer aging process through SERS.

To evaluate the thermal degradation of oil-paper insulation, analyzing the furfural dissolved in transformer oil is determined as an important method. Surface enhanced Raman spectroscopy (SERS) is used to detect the concentration of furfural, which not only overcomes the shortcomings of the traditional detection methods, but also realizes nondestructive testing. In this paper, SERS for detecting the furfural in transformer oil was investigated. we used Electrochemical displacement to synthesize the two-dimensional Ag Nano-branch on copper foil, and then applied for measuring oil samples with different furfural concentrations by SERS. The SERS signal of furfural was used for qualitative and quantitative analysis at ∼931 cm−1, and the mathematical model was constructed, depending on the Raman peak intensity ratios Ι931/Ι770 and furfural concentrations, basing on the least squares method. As a result, without complex preprocessing steps, SERS can detect the transformer oil's furfural concentration quantitatively. Therefore, it proposes a new method to detect furfural concentration in transformer oil with fast in-situ capability.

A sensitive optical system which can be used to measure accurately the concentrations of fault gases dissolved in samples extracted from transformer oil is described, and some test results are presented.

Hierarchical flower-like ZnO nanorods, net-like ZnO nanofibers and ZnO nanobulks have been successfully synthesized via a surfactant assisted hydrothemal method. The synthesized products were characterized by X-ray powder diffraction and field emission scanning electron microscopy, respectively. A possible growth mechanism of the various hierarchical ZnO nanostructures is discussed in detail. Gas sensors based on the as-prepared ZnO nanostructures were fabricated by screen-printing on a flat ceramic substrate. Furthermore, their gas sensing characteristics towards methane were systematically investigated. Methane is an important characteristic hydrocarbon contaminant found dissolved in power transformer oil as a result of faults. We find that the hierarchical flower-like ZnO nanorods and net-like ZnO nanofibers samples show higher gas response and lower operating temperature with rapid response-recovery time compared to those of sensors based on ZnO nanobulks. These results present a feasible way of exploring high performance sensing materials for on-site detection of characteristic fault gases dissolved in transformer oil.

The operation conditions of oil immersed power transform will affect the safety and stability of power system directly. As one of the main fault characteristic gases dissolved in transformer oil, acetylene (C 2 H 2 ) gas can effectively reflect the discharge faults which occur in oil-paper insulation system. Monitoring its content of C 2 H 2 gas has a great significance in diagnosis and evaluation the running state of transformer. In this paper, tin oxide (SnO 2 ) and reduced graphene oxide (rGO) nanocomposite based C 2 H 2 gas sensor was developed. The gas sensing material was synthesized by a facile hydrothermal method and characterized by X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM) and transmission electron microscope (TEM). Then, the gas sensing properties of as-synthesized nanomaterials were tested by exposing to different concentration of C 2 H 2 gas with a temperature range of 100–300 °C and its gas sensing mechanism of C 2 H 2 gas was discussed. The characterization results indicated that the SnO 2 and SnO 2 /rGO were succeed synthesized. Under the optimum operating temperature (180 °C) of SnO 2 /rGO, it exhibited a low detection limit (1 μL/L), high response (50 ppm, 13.2) and good linearity (1–10 μL/L). The gas sensing mechanism for the better sensing properties to C 2 H 2 gas was explained based on superhigh surface areas and heterojunction between SnO 2 and rGO. This results highlight SnO 2 /rGO can effective enhance the C2H2 gas sensing properties for monitoring the dissolved C 2 H 2 gas in transformer oil with high performances.

Methyl ethyl ketone dissolved in transformer oil can be used as the mark of thermal insulation degradation. As an effective detecting method for trace liquid, Laser Raman Spectroscopy (LRS) is more suitable for in-situ detecting methyl ethyl ketone concentration than traditional detection methods. The application of LRS in detecting methyl ethyl ketone concentration has been studied. Firstly, the structure model of methyl ethyl ketone molecule was built by Gauss View 5.0. Then, the Raman spectrum of methyl ethyl ketone was analyzed by Gaussian 09W software theoretically. And the liquid test platform was built based on confocal Raman technology in the lab. The Raman characteristic peak of methyl ethyl ketone dissolved in oil was selected by testing standard samples of transformer oil directly. Moreover, applying the least square method, the quantitative analysis method was built between the areas of Raman characteristic peak and the concentrations. The experimental results showed that LRS can be used to effectively detect the concentration of methyl ethyl ketone dissolved in transformer oil. Therefore, a new detection tool is presented for transformer health diagnosis.

Thermal fault threatens the safe and stable operation of oil filled equipment. Based on TG-DSC and ReaxFF (reactive force field), in this paper, the reaction of molecular dynamics simulation method was used to establish the simulation model of mineral insulating oil molecules under different temperatures and to study dynamic pyrolysis process and produce regularity of gaseous molecules in transformer oil. First, thermal analysis technique TG-DSC is used to observe the quality and energy changes during the process of transformer oil decomposition. Then, by using reactive reaction force field, the three typical hydrocarbons in transformer oil-alkane, cycloalkane and aromatic is simulated in a liquid system and the microscopic decomposition mechanism of transformer oil is studied. Through the establishment of a liquid system consisting of three different kinds of hydrocarbons, and the minimization of energy balance system, the pyrolysis of them in NVT system is simulated under different temperatures. The initial decomposition temperatures of the three kinds of hydrocarbons is that alkane is lower than cycloalkane and aromatic is the highest. Then the relationship between temperature and the process of pyrolysis is concluded. This study reveals the pyrolysis dynamic mechanism of the transformer oil on the atomic level, corresponding to the real pyrolysis successfully.

Methane, ethane, ethylene and acetylene are the effective failure characteristic gases for power transformer, which can effectively reflect the electric and overheat faults. Monitoring these gases have a great significance in diagnosis and evaluation the running state of transformer. In this paper, Ptdoped SnO 2 nanoflower gas sensor was prepared, then its detection characteristic and response mechanism of hydrocarbon gases had been studied. The results indicated that, as the increasing of specific surface area, compared to the pure SnO 2 nanoparticle gas sensor, the Pt-doped SnO 2 nanoflower gas sensor has higher sensitivity and faster response characteristic to hydrocarbon gases, and responses versus gas concentration kept a good linearity when the gas concentration between 5 μL/L and 50 μL/L. The role of the dopant of Pt into the SnO 2 and the sensing mechanism had also been discussed in this work.

As an aging product of insulating paper, methyl formate dissolved in transformer oil can reflect the progress of oxidization and affect the insulating property of transformer oil. And the concentration of methyl formate dissolved in transformer oil has been given more and more attention. Confocal Laser Raman spectroscopy (CLRS) can directly perform non-contact fast measurement to transformer oil samples, especially being suitable for online monitoring. Using CLRS to detect the concentration of methyl formate dissolved in transformer oil can overcome many disadvantages of traditional methods. In this paper, the application of CLRS in the detection of methyl formate dissolved in transformer oil was studied. Firstly, Gaussian 09W was applied to analyze the Raman spectrum of methyl formate. Based on the CLRS test platform set up in the lab, oil samples with different methyl formate concentrations were detected. And methyl formate was characterized by Raman signal at 903 cm−1, where can be used to quantitative analysis with the least squares method. The results show that LRS can be used in the analysis of methyl formate dissolved in transformer oil, which lays the foundation for the online monitoring of dissolved methyl formate in transformer oil with CLRS.

Acetic acid produced in oil-paper insulation aging process would greatly reduce the quality of insulation performance of a transformer, and seriously affect the healthy operation of the transformer. The monitoring of acetic acid content dissolved in the transformer oil is of great significance for ensuring the safe and reliable operation of the transformer. Laser Raman spectroscopy is a molecular analysis technology based on the Raman effect, which can directly perform non-contact measurement to acetic acid dissolved in the transformer oil by a laser with single frequency. Compared to the conventional methods, laser Raman spectroscopy method is more simple and time-saving. In this paper, the application of laser Raman spectroscopy in the detection of acetic acid dissolved in the transformer oil was studied. Firstly, Gaussian 09W simulation was applied to analyze the Raman spectrum of acetic acid. The spectral peaks and Raman vibration modes of acetic acid were analyzed and assigned. Acetic acid dissolved in the transformer oil was directly detected based on the laser Raman spectroscopy liquid detection test platform. Acetic acid was characterized by Raman signal at 618cm−1, where no spectral interferences caused by oil-derived Raman signals occur. And the qualitative and quantitative analysis methods for the acetic acid based on the characteristic spectra and the least squares method were established. The results show that laser Raman spectroscopy can be used in the analysis of acetic acid dissolved in the transformer oil with good detection stability, which lays a good foundation for quantitative detection and the online monitoring of acetic acid dissolved in the transformer oil with laser Raman spectroscopy.

The analysis of methanol concentration is regard as a new means of evaluating the aging degree of insulating paper in power transformers. Using Laser Raman Spectroscopy (LRS) to detect the concentration of methanol in transformer oil can overcome many shortcomings of traditional detection techniques and realize fast test. In this paper, the application of LRS in detecting methanol concentration has been studied. First of all, Gaussian 09W was used to build the structure model of methanol molecule and analyze the Raman spectrum of methanol molecule. LRS liquid detection test platform was built based on confocal Raman technology. Standard samples of transformer oil were tested utilizing this test platform directly, and the Raman characteristic peak at 1030 cm−1 was selected for quantitative analyses. In order to further improve the detection sensitivity, extraction technology was applied. At last, applying the least squares method, the quantitative analysis method was built between the areas of Raman characteristic peak and methanol concentrations. The experimental results showed that LRS method can effectively detect the concentration of methanol in transformer oil, which proposes a new way to realizing the fast test of methanol in transformer oil.

Three ratio method of dissolved gases is a most recognized method to evaluate the insulation of oil-filled equipment, but it lacks the quantitative relation of dissolved gases and severity of the defects. This paper study the reaction mechanism of transformer oil pyrolysis based on molecular simulation. First, thermal analysis technique TG-DSC is used to observe the quality and energy changes during the process of transformer oil decomposition. Then, by using reactive reaction force field (ReaxFF), the three typical hydrocarbons in transformer oil-alkane, cycloalkane and aromatic is simulated in a liquid system and the microscopic decomposition mechanism of transformer oil is studied. Through the establishment of a liquid system consisting of three different kinds of hydrocarbons, and the minimization of energy balance system, the pyrolysis of them in NVT system is simulated under different temperatures. Then the relationship between temperature and the process of pyrolysis is concluded. This study reveals the pyrolysis dynamic mechanism of the transformer oil on the atomic level, corresponding to the real pyrolysis successfully.

Accurate detection of trace fault characteristic gases (CH4, C 2 H 2 , C 2 H4, C 2 H 6 , H 2 , CO, CO 2 and O 2 ) is the key to ensure safe service of power transformer. Nowadays, the existing defection method for the fault characteristic gases have problems on gas cross interference, aging, and so on. In this paper, the technology of Raman spectroscopy was used to detect fault gases extracted from transformer oil, which has advantages of antiaging, simultaneous multi-element analysis. Based on the built-up confocal microprobe Raman spectroscopy platform, the simultaneous measurement of eight kinds of fault gases as well as N 2 was achieved. Then, the characteristic Raman spectra and the minimum measurable concentrations of each gas had been obtained. Besides, to solve the bottleneck problems of limited gas minimum measurable concentrations of Raman spectroscopy technology, a cavity-length modulation frequency-locking enhanced cavity was proposed to combine with Raman spectroscopy. Based on which, the detection limit of minimum measurable concentrations for fault characteristic gases were improved for about 11.8 times, reaching CH4 (17.4 μL/L), C 2 H 2 (32.0 μL/L), C 2 H4 (52.8 μL/L), C 2 H 6 (29.3 μL/L), H 2 (75.5 μL/L), CO (130.2 μL/L), CO 2 (90.6 μL/L), O 2 (80.7 μL/L) and N 2 (85.1 μL/L). The findings of this study lay the foundation for a new method for simultaneously measuring fault gases extracted from transformer oil.

Recently, copper sulfide which was formed by corrosive sulfur in transformer oil reacting with copper windings caused many faults of high voltage electric equipments. Online detection of dibenzyl disulfide (DBDS) has become a new way of avoiding happening major accident. Comparing with the traditional monitoring technologies, Laser Raman spectroscopy (LRS) has incomparable advantages, such as non-contact detection, quickly integrated detection of multiple characteristic substances, using single-frequency laser, which are necessary to achieve online monitoring. Based on the basic principle of LRS, we build up a platform using LRS for quantitative and qualitative analysis and develop the relevant experimental methods. Thermal aging tests are done at 140 °C with an air atmosphere up to 48h with copper strip. The test results indicate that Laser Raman spectroscopy can be applied to identify corrosivity of mineral oil for a running power transformer.

Recently, copper sulfide which was formed by corrosive sulfur in transformer oil reacting with copper windings caused many faults of high voltage electric equipments. Online detection of dibenzyl disulfide (DBDS) is the key to avoid happening major accident. Comparing with the traditional monitoring technologies, Laser Raman spectroscopy (LRS) has incomparable advantages, such as non-contact detection, quickly integrated detection of multiple characteristic substances, using single-frequency laser, which are necessary to achieve online monitoring. Based on the basic principle of LRS, we build up a platform using LRS for quantitative and qualitative analysis and develop the relevant experimental methods. Thermal aging tests are done at 140 °C with an air atmosphere up to 48h with copper strip. The test results indicate that Laser Raman spectroscopy can be applied to identify corrosivity of mineral oil for a running power transformer.

This paper investigates the applications of photoacoustic spectrometry (PAS) to analyze dissolved gases in transformer oil. An inexpensive experimental device with an electric pulse infrared light and six filters has been developed to detect six main fault gases in transformer oil, namely, CH 4 , C 2 H 6 , C 2 H 4 , C 2 H 2 , CO, and CO 2 . The method of choosing the characteristic wavelengths for the above six fault gases is discussed in detail. A weighted least-square error method is proposed to analyze the measured data and to determine the component concentrations of dissolved gas in oil. Two sample sets of gas mixtures are analyzed by the experimental device for the purpose of verification. The PAS measurements are compared with the true values and the values measured by a conventional gas chromatograph (GC) method. The comparison results show that PAS is an effective way of analyzing dissolved gases in transformer oil, which exhibits some advantages in the gas detection such as consuming no gas, separating no gas, high accuracy, high stability, and rapid measurements.

Incipient transformer faults can be effectively diagnosed by on-line monitoring of dissolved gases in transformer oil. Raman spectroscopy can simultaneously detect the type and concentration of seven kinds of characteristic dissolved gases in oil excited by a laser at one wavelength. It is a promising technique without facing the issues of aging and calibration compared to traditional gas chromatography techniques. In this paper, the principle of Raman spectroscopy is briefly introduced. The experimental platform based on the confocal micro Raman system is explained, the detection method of Raman spectroscopy for dissolved gases in transformer oil is determined and the Raman detection of seven fault gases (CO, CO 2 , CH 4 , C 2 H 2 , C 2 H 4 , C 2 H 6 and H 2 ) is fundamentally realized. On this basis, characteristic Raman spectra are chosen and the corresponding molecular vibrational modes are analyzed simultaneously. To lower detection limits, a surface-enhanced Raman scattering (SERS) substrate is introduced since near-field coupling effects between adjacent Au nanoparticles decorated on SiO 2 substrate can greatly increase local electromagnetic field. The simulation results of electric field distribution are presented, all of which lay a good foundation for online monitoring of dissolved gases in transformer oil.

H 2 is one of the main fault characteristic gases dissolved in transformer oil, which can indicate the electric faults, such as high energy discharge, spark discharge and partial discharge, and partial oil overheating phenomenon. Detection the content of H 2 has an important significance for transformer diagnosis and state assessment. Gas sensing detection technology is the core of online monitoring device. In this paper, for the detection of dissolved H 2 , based on the density functional theory and the first-principles, pure and Ag-doped SnO 2 models and gas adsorption models were built, and theoretical calculations were conducted. Meanwhile, pure and Ag-doped SnO 2 gas sensing materials were synthesized with hydrothermal method. Then SnO 2 based gas sensors were fabricated and their gas sensing properties were measured. Finally, its gas sensing mechanism was discussed based on the macro gas sensing properties and micro simulating calculations. The results indicated that, Ag doping can improve the gas sensing properties of SnO 2 nanostructures to H 2 , Ag doping sensor has better effect than pure sensor for H 2 detection, such as a lower optimum operating temperature of 340 °C, lower detection limit of 10 µL/L with higher sensitivity. The role of the dopant of Ag into the SnO 2 and the sensing mechanism had also been discussed in this work. The experimental results verifies the feasibility and accuracy of study the gas sensing performances of SnO 2 based gas sensors using the first-principles calculation based on the density functional theory, further perfecting its gas sensing mechanism of SnO 2 based gas sensor and providing us a fresh idea and feasible way to develop different kinds dopant of metal or metal-oxide gas sensors with high performances.

Analyzing the concentration of low molecular acids dissolved in oil is vital in the oil-paper insulation aging diagnostic procedure of power transformers. The existing methods cannot distinguish between different acid types and their strengths. In this study, an improved solution Raman detection platform is fabricated. The direct measurement of dissolved acetic acid, a kind of low molecular acids, is observed in transformer oil without extraction. The Raman shift line of oil-dissolved acetic acid at 891 cm−1 corresponding to H–C–H symmetrical swing and O–H swing modes is taken as its characteristic value. Taking Raman shift line of pure oil at 932 cm−1 as an internal standard, a linear regression curve for quantitative analysis is obtained with a slope of 0.19. The best platform parameter of accumulation number is 300, which is determined by Allan deviation analysis. The current concentration detection limit and accuracy for oil-dissolved acetic acid are obtained at about 0.68 mg/mL and 91.66%, separately. The results show that Raman spectroscopy could be a useful alternative method for evaluation insulation aging state of an operating power transformer in the future.

In the past decades, the application of nanostructures and nanotechnology has attracted intensive attention in many fields from electronic devices to chemical sensors. The monitoring of toxic or inflammable gases has been increasingly important for humans, environment and various areas. In the power system field, power transformers are important equipment of electricity transmission and distribution, and its operation condition is associated with the safety of power system. If the power transformers break down, there will be a huge lose in national economic. Hydrogen (H 2 ) is one of the main fault characteristic gases dissolved in transformer oil, which can indicate the high energy discharge, spark discharge, partial discharge and partial oil overheating. The gas sensor technology is the key of on-line monitoring trace amount of gases. This paper made a research on the detection characteristics of CuO-SnO 2 gas sensor to the H 2 gas dissolved in transformer oil. In this study, a facile hydrothermal method was adopted to fabricate SnO 2 and CuO-SnO 2 nanoparticles. The CuO content was chosen as 5mol-% (sample 1), 10mol-% (sample 2) and 15mol-% (sample 3). Microstructures and surface morphologies for all samples were characterised by X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM). Then the gas sensing mechanism of gas sensor was analyzed. The results proved that: A lot of p-n heterojunction would form which change the gas sensing properties of composite SnO 2 -based gas sensor to hydrogen; compared with the SnO 2 gas sensor, the CuO-SnO 2 performed better sensitivity and quicker response to the H 2 ,and also kept a good linearity and stability. The results offer a new thought to study the application of composite SnO 2 -based gas sensor on detecting the gases dissolved in transformer oil on line and represent an advance of heterojuction nanostructures in further enhancing the functionality of gas sensors towards H 2 .

Partial discharge in oil-paper insulation is one of the main reasons for aging and disruption of the power transformer. Understanding the emergence of partial discharge and the gas variation law in time can determine the development of internal latent fault. Therefore, based on the air-gap discharge model, which is the primary discharge type, partial discharge characteristics and gas generation law were researched. The experimental results show that: Sk with Ku of Hη(φ) and Sk of Hqmax(φ) ascend as time rises; air-gap partial discharge mainly produces H2, CH4 as well as CO and the absolute gas generation rate of main gases show a downward trend; combined with fuzzy diagnosis, the relationship between partial discharge and dissolved gas in oil was analyzed, revealing that the code of the approved three-ratio method correlated to air-gap partial discharge most is “110”.