Multigas Analysis by Cavity-Enhanced Raman Spectroscopy for Power Transformer Diagnosis

13CH4/12CH4 sensing using Raman spectroscopy

The paper presents a technique for measuring the concentration of 13CH4 in natural methane using Raman spectroscopy. The peak positions and the relative scattering cross-sections of the Q-branches for the most intense vibrational bands of 13CH4 are determined. Features of the 13CH4/12CH4 ratio measurement methods using Q-branches of the ν1 and ν3 bands were considered. It was shown that the 13CH4/12CH4 ratio can be determined by simulation of the ν3 bands of these molecules without the use of experimental spectra. In our experiments the measurement error of δ13C value was 10 ‰ using the 100-s exposure spectrum at a gas pressure close to 1 atm recorded on the developed Raman spectrometer. In addition, the Raman spectra of alkanes (up to n-hexane) in the range of 2850-3050 cm-1 at a resolution of 0.4 cm-1 are presented, and their integrated intensities in the ranges of the characteristic bands of 13CH4 and 12CH4 are provided. The data obtained make it possible to expand the capabilities of Raman gas analyzers in the mud gas logging industry.

Quantitative detection of multicomponent SF6 decomposition products based on Fourier transform infrared spectroscopy combined with SCARS-DNN

Accurate and efficient quantitative analysis of the decomposition products of the insulating medium SF6 in gas-insulated switchgear (GIS) is important for an effective assessment of its internal insulation status. In this work, a quantitative calibration model of Fourier Transform Infrared Spectroscopy (FTIR) combined with SCARS-DNN (Stability Competitive Adaptive Reweighted Sampling-Deep Neural Network) is proposed for the rapid non-destructive detection of SF6 decomposition products. First, the interference of the background gas SF6 on the absorption spectra of the decomposition products is eliminated according to the Lambert-Beer law, while baseline correction and Savitzky-Golay (S-G) smoothing are used to remove baseline drift and noise. Subsequently, a Monte Carlo cross-validation method is used to detect and eliminate the anomalous samples. Then feature selection is performed using uninformative variable elimination (UVE) and stability competitive adaptive reweighted sampling (SCARS), and finally quantitative calibration models of FULL-DNN (full spectral band), UVE-DNN, and SCARS-DNN are developed. For the quantitative detection of SF6 decomposition products, the SCARS-DNN model had the best prediction performance with a maximum reduction of 96.18% in the root mean square error (RMSE) and 96.11% in the mean absolute percentage error (MAPE). Results reveal that the relative errors are basically kept below 1.36% when predicting the three decomposition products, even in the presence of a high level of SF6 interference. Therefore, the SCARS-DNN model is suitable for high-precision quantitative detection of SF6 decomposition gas.

High-Precision Trace Hydrogen Sensing by Multipass Raman Scattering

Despite its growing importance in the energy generation and storage industry, the detection of hydrogen in trace concentrations remains challenging, as established optical absorption methods are ineffective in probing homonuclear diatomics. Besides indirect detection approaches using, e.g., chemically sensitized microdevices, Raman scattering has shown promise as an alternative direct method of unambiguous hydrogen chemical fingerprinting. We investigated the suitability of feedback-assisted multipass spontaneous Raman scattering for this task and examined the precision with which hydrogen can be sensed at concentrations below 2 parts per million. A limit of detection of 60, 30, and 20 parts per billion was obtained at a pressure of 0.2 MPa in a 10-min-long, 120-min-long, and 720-min-long measurement, respectively, with the lowest concentration probed being 75 parts per billion. Various methods of signal extraction were compared, including asymmetric multi-peak fitting, which allowed the resolution of concentration steps of 50 parts per billion, determining the ambient air hydrogen concentration with an uncertainty level of 20 parts per billion.

Simple technique of coupling a diode laser into a linear power buildup cavity for Raman gas sensing

We report a novel, to the best of our knowledge, and simple technique to lock a 642 nm multi-quantum well diode laser to an external linear power buildup cavity by directly feeding the cavity reflected light back to the diode laser for enhancement of gas Raman signals. The dominance of the resonant light field in the locking process is achieved by reducing the reflectivity of the cavity input mirror and thus making the intensity of the directly reflected light weaker than that of the resonant light. Compared with traditional techniques, stable power buildup in the fundamental transverse mode TEM₀₀ is guaranteed without any additional optical elements or complex optical arrangements. An intracavity exciting light of 160 W is generated with a 40 mW diode laser. Using a backward Raman light collection geometry, detection limits at the ppm level are achieved for ambient gases (N₂, O₂) with an exposure time of 60 s.

Multiple Gas Detection by Cavity-Enhanced Raman Spectroscopy with Sub-ppm Sensitivity

Accurate and sensitive detection of multicomponent trace gases below the parts-per-million (ppm) level is needed in a variety of medical, industrial, and environmental applications. Raman spectroscopy can identify multiple molecules in the sample simultaneously and has excellent potential for fast diagnosis of various samples, but applications are often limited by its sensitivity. In this contribution, we report the development of a cavity-enhanced Raman spectroscopy instrument using a narrow-line width 532 nm laser locked with a high-finesse cavity through a Pound-Drever-Hall locking servo, which allows continuous measurement in a broad spectral range. An intracavity laser power of up to 1 kW was achieved with an incident laser power of about 240 mW, resulting in a significant enhancement of the Raman signal in the range of 200-5000 cm^-1 and a sub-ppm sensitivity for various molecules. The technique is applied in the detection of different samples, including ambient air, natural gas, and reference gas of sulfur hexafluoride, demonstrating its capability for the quantitative measurement of various trace components.

Trace photoacoustic SO2 gas sensor in SF6 utilizing a 266 nm UV laser and an acousto-optic power stabilizer

A sulfur dioxide (SO₂) gas sensor based on the photoacoustic spectroscopy technology in a sulfur hexafluoride (SF₆) gas matrix was demonstrated for SF₆ decomposition components monitoring in the power system. A passive Q-switching laser diode (LD) pumped all-solid-state 266 nm deep-ultraviolet laser was exploited as the laser excitation source. The photoacoustic signal amplitude is linear related to the incident optical power, whereas, a random laser power jitter is inevitable since the immature laser manufacturing technology in UV spectral region. A compact laser power stabilization system was developed for better sensor performance by adopting a photodetector, a custom-made internal closed-loop feedback controller and a Bragg acousto-optic modulator (AOM). The out-power stability of 0.04% was achieved even though the original power stability was 0.41% for ∼ 2 hours. A differential two-resonator photoacoustic cell (PAC) was designed for weak photoacoustic signal detection. The special physical constants of SF₆ buffer gas induced a high-Q factor of 85. A detection limit of 140 ppbv was obtained after the optimization, which corresponds to a normalized noise equivalent absorption coefficient of 3.2 × 10^-9cm^-1WHz^-1/2.

Diffusion Mechanisms of Dissolved Gases in Transformer Oil Influenced with Moisture Based on Molecular Dynamics Simulation

Dissolved Gas Analysis (DGA) of insulating oil is widely used for diagnosing transformer incipient faults. Moisture is a major contaminant and degradation byproduct of transformer insulating oil. In this paper, molecular dynamics simulation was used to study the influence of moisture on the diffusion movement of dissolved gases in the insulating oil. Cycloalkanes (C₂₀H₄₂), alkanes (C₂₀H₃₈), and aromatic hydrocarbons (C₂₀H₂₆) are selected as the basic structural units in the constructed transformer oil simulation system. 0%, 1%, 3%, and 5% moisture molecules are added to insulating oil, respectively, and the insulating oil generates seven kinds of gas molecules through cracking. With an anhydrous model used as a benchmark, we researched the diffusion trajectory, the diffusion coefficient (D), free volume (V _F), and the moisture-gas interaction energy of each gas molecule as a function of moisture content. Through this study, we found that the increase of moisture content enlarges the V _F value of dissolved gas in insulating oil, which makes the gas in oil easier to diffuse. Besides, the moisture can also alter the mean square displacement (MSD) of dissolved gases; the mutual energy of molecules is mainly affected by the electrostatic interaction energy. This study can contribute to a better understanding of the influence of different moisture contents on the diffusion movement of dissolved gas in transformer oil from the micro level.

Dense-pattern multi-pass cavity based on spherical mirrors in a Z-shaped configuration for Raman gas sensing

We report a dense-pattern multi-pass cavity (MPC) based on four spherical mirrors placed in a Z-shaped cavity configuration for improving the Raman signals from gases. The folding structure of the cavity causes dense patterns of spots, and at least 420 beams are reflected in the cavity. Raman spectra of ambient air, methane, and ethylene are recorded to demonstrate the performance of our apparatus. At atmospheric pressure, ppm-level detection limits of the gases are achieved with 10 s of exposure time. The Raman signal intensities of the gases show excellent linearity with the gases' partial pressures, which means that high-accuracy detection is also feasible.

Genetic algorithm based artificial neural network and partial least squares regression methods to predict of breakdown voltage for transformer oils samples in power industry using ATR-FTIR spectroscopy

The current study proposes a novel analytical method for calculating the breakdown voltage (BV) of transformer oil samples considered as a significant method to assess the safe operation of power industry. Transformer oil samples can be analyzed using the Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy combined with multivariate calibration methods. The partial least squares regression (PLSR) back propagation-artificial neural network (BP-ANN) methods and a genetic algorithm (GA) for variable selection are used to predict and assess breakdown voltage in transformer oil samples from various Iranian transformer oils. As a result, the root mean square error (RMSE) and correlation coefficient for the training and test sets of oil samples are also calculated. In the GA-PLS-R method, the squared correlation coefficient (R²pred) and root mean square prediction error (RMSEP) are 0.9437 and 2.6835, respectively. GA-BP-ANN, on the other hand, had a lower RMSEP value (0.2874) and a higher R²pred function (0.9891). Considering the complexity of transformer oil samples, the performance of GA-BP-ANN has resulted in an efficient approach for predicting breakdown voltage; consequently, it can be effectively used as a new method for quantitative breakdown voltage analysis of samples to evaluate the health of transformer oil. .

Compact QEPAS humidity sensor in SF6 buffer gas for high-voltage gas power systems

In SF₆ insulated high-voltage gas power systems, H₂O is the most problematic impurity which not only decreases insulation performance but also creates an acidic atmosphere that promotes corrosion. Corrosion damages electrical equipment and leads to leaks, which pose serious safety hazards to people and the environment. A QEPAS-based sensor system for the sub-ppm level H₂O detection in SF₆ buffer gas was developed by use of a near-infrared commercial DFB diode laser. Since the specific physical constants of SF₆ are strongly different from that of N₂ or air, the resonant frequency and Q-factor of the bare quartz tuning fork (QTF) had changed to 32,763 Hz and 4173, respectively. The optimal vertical detection position was 1.2 mm far from the QTF opening. After the experimental optimization of acoustic micro-resonator (AmR) parameters, gas pressures, and modulation depths, a detection limit of 0.49 ppm was achieved for an averaging time of 1 s, which provided a powerful prevention tool for the safety monitoring in power systems.

Hazardous Gas Detection by Cavity-Enhanced Raman Spectroscopy for Environmental Safety Monitoring

We demonstrate the practicability of cavity-enhanced Raman spectroscopy (CERS) with a folded multipass cavity as a unique tool for the detection of hazardous gases in the atmosphere. A four-mirror Z-sharped multipass cavity results in a greatly extended laser-gas interaction length to improve the Raman signal intensity of gases. For Raman intensity maximization, the optimal number of intracavity beams of a single reflection cycle is calculated and then the cavity parameters are designed. A total of 360 intracavity beams are realized, which are circulated four times in the cavity based on the polarization. ppb-Level Raman gas sensing at atmospheric pressure for several typical explosive gases and toxic gases in ambient air, including hydrogen (H₂), methane (CH₄), carbon monoxide (CO), hydrogen sulfide (H₂S), and chlorine (Cl₂), is achieved at 300 s exposure time. Our CERS apparatus, which can detect multiple gases simultaneously with ultrahigh sensitivity and high selectivity, is powerful for detecting hazardous gases in the atmosphere, and it has excellent potential for environmental safety monitoring.

A Versatile Multiple-Pass Raman System for Industrial Trace Gas Detection

The fast and in-line multigas detection is critical for a variety of industrial applications. In the present work, we demonstrate the utility of multiple-pass-enhanced Raman spectroscopy as a unique tool for sensitive industrial multigas detection. Instead of using spherical mirrors, D-shaped mirrors are chosen as cavity mirrors in our design, and 26 total passes are achieved in a simple and compact multiple-pass optical system. Due to the large number of passes achieved inside the multiple-pass cavity, experiments with ambient air show that the noise equivalent detection limit (3σ) of 7.6 Pa (N2), 8.4 Pa (O2) and 2.8 Pa (H2O), which correspond to relative abundance by volume at 1 bar total pressure of 76 ppm, 84 ppm and 28 ppm, can be achieved in one second with a 1.5 W red laser. Moreover, this multiple-pass Raman system can be easily upgraded to a multiple-channel detection system, and a two-channel detection system is demonstrated and characterized. High utilization ratio of laser energy (defined as the ratio of laser energy at sampling point to the laser output energy) is realized in this design, and high sensitivity is achieved in every sampling position. Compared with single-point sampling system, the back-to-back experiments show that LODs of 8.0 Pa, 8.9 Pa and 3.0 Pa can be achieved for N2, O2 and H2O in one second. Methods to further improve the system performance are also briefly discussed, and the analysis shows that similar or even better sensitivity can be achieved in both sampling positions for practical industrial applications.

Fiber-enhanced Raman spectroscopy for highly sensitive H2 and SO2 sensing with a hollow-core anti-resonant fiber

An innovative fiber-enhanced Raman gas sensing system with a hollow-core anti-resonant fiber is introduced. Two iris diaphragms are implemented for spatial filtering, and a reflecting mirror is attached to one fiber end that provides a highly improved Raman signal enhancement over 2.9 times than the typical bare fiber system. The analytical performance for multigas compositions is thoroughly demonstrated by recording the Raman spectra of carbon dioxide (CO₂), oxygen (O₂), nitrogen (N₂), hydrogen (H₂), and sulfur dioxide (SO₂) with limits of detection down to low-ppm levels as well as a long-term instability < 1.05%. The excellent linear relationship between Raman signal intensity (peak height) and gas concentrations indicates a promising potential for accurate quantification.