T型光聲池的光聲光譜技術(shù)用于同時(shí)檢測(cè)基于三重共振模態(tài)的多組分氣體
 

　　T-type cell mediated photoacoustic spectroscopy for simultaneous detection of multi-component gases based on triple resonance modality
 

　　近日，來自西安電子科技大學(xué)、哈爾濱工業(yè)大學(xué)可調(diào)諧(氣體)激光技術(shù)國(guó)家級(jí)重點(diǎn)實(shí)驗(yàn)室的聯(lián)合研究團(tuán)隊(duì)發(fā)表了《T型光聲池的光聲光譜技術(shù)用于基于三重共振模態(tài)的多組分氣體的同時(shí)檢測(cè)》論文。
 

　　Recently, the joint research team from  School of Optoelectronic Engineering, Xidian University,  National Key Laboratory of Science and Technology on Tunable Laser, Harbin Institute of Technology, published an academic papers T-type cell mediated photoacoustic spectroscopy for simultaneous detection of multi-component gases based on triple resonance modality.
 

　　油浸式電力變壓器是現(xiàn)代電力分配和傳輸系統(tǒng)中最重要的絕緣設(shè)備之一。通過同時(shí)測(cè)量絕緣油中的溶解氣體，如一氧化碳(CO)、甲烷(CH4)和乙炔(C2H2)，可以在電力變壓器的過熱、電弧和局部放電故障的早期診斷中提供合適的解決方案。變壓器故障主要可分為過熱故障和放電故障。CO、CH4和C2H2的含量變化是變壓器故障的主要指標(biāo)。過熱故障包括裸金屬過熱、固體絕緣過熱和低溫過熱。裸金屬過熱的特征是烴類氣體(如CH4和C2H2)濃度的上升。上述兩種氣體的總和占總烴類氣體的80%以上，其中CH4占較大比例(>30 ppm)。CO的濃度(>300 ppm)強(qiáng)烈指示固體絕緣過熱和變壓器故障中的低溫過熱。當(dāng)變壓器處于放電故障時(shí)，C2H2會(huì)急劇增加(>5 ppm，占總烴類氣體的20%-70%)。因此，本研究選擇CO、CH4和C2H2作為目標(biāo)分析物。傳統(tǒng)的多組分氣體定量檢測(cè)方法，如氣相色譜儀、半導(dǎo)體氣體傳感器和電化學(xué)傳感器，在實(shí)時(shí)監(jiān)測(cè)、恢復(fù)時(shí)間、選擇性和交叉敏感性方面存在一定限制。基于光聲光譜技術(shù)的光學(xué)傳感器平臺(tái)具有高靈敏度、高選擇性、快速響應(yīng)、長(zhǎng)壽命和成熟的傳感器設(shè)備等優(yōu)點(diǎn)，在多組分氣體傳感領(lǐng)域發(fā)揮著重要作用。已經(jīng)開發(fā)出多種基于光聲光譜技術(shù)的多組分氣體傳感器模式，如傅里葉變換紅外光聲光譜模式、基于寬帶檢測(cè)的熱輻射體或黑體輻射體使用多個(gè)帶通濾波器、多激光器與時(shí)分復(fù)用(TDM)方法的結(jié)合，以及采用多共振器和頻率分割復(fù)用(FDM)方案。然而，由于寬帶光源的相對(duì)弱強(qiáng)度，弱光聲(PA)信號(hào)易受到背景噪聲的干擾，這是高靈敏度檢測(cè)的主要障礙。
 

　　Oil-immersed power transformer is one of the most important insulation equipment in modern power distribution and transmission systems. Simultaneous measurements of the dissolved gases in insulating oil, such as carbon monoxide (CO), methane (CH4) and acetylene (C2H2), can represent a suitable solution in early diagnosis of overheating, arcing and partial discharge failures of power transformers . Transformer fault can mainly be divided into overheating fault and discharge fault. The content changes of CO, CH4, and C2H2 are the main indicators of transformer failure. Overheating fault includes bare metal overheating, solid insulation overheating and low temperature overheating. The bare metal overheating is characterized by the rising concentration of hydrocarbon gas, such as CH4 and C2H2. The sum of the above two gases accounts for more than 80% of the total hydrocarbon gas, and CH4 accounts for a larger proportion (>30 ppm). The concentration of CO (>300 ppm) strongly indicates the solid insulation overheating and the low temperature overheating in the transformer failure. When the transformer is in discharge fault, the C2H2 will increase dramatically (>5 ppm, 20%− 70% of the total hydrocarbon gas). Therefore, CO, CH4, and C2H2 are selected as the target analytes in this work. The traditional quantitative detection of multiple analytes, such as gas chromatographs, semiconductor gas sensors and electrochemical sensors, were limited in terms of real time monitoring, recovery time, poor selectivity and cross sensitivity. Photoacoustic spectroscopy (PAS)-based optical sensor platforms, which feature the advantages of high sensitivity, high selectivity, fast response, long lifetime and well-established sensing devices, have played an important role in the field of multi-component gas sensing. Various PAS-based multi-gas sensor modalities have been developed, such as Fourier transform infrared PAS modality, broadband detection based thermal emitters or blackbody radiators using several band-pass filters, the use of multi-lasers combined time-division multiplexing (TDM) methods , and multi-resonators with frequency-division multiplexing (FDM) schemes. Due to the relatively poor intensity of the broadband source, the weak photoacoustic (PA) signals were sensitively affected by the background noise, which was a major obstacle to highly sensitive detection.
 

　　由于吸收和共振圓柱體共同決定了其共振頻率，設(shè)計(jì)并驗(yàn)證了一種T型光聲池作為適當(dāng)?shù)膫鞲衅鳌Ｍㄟ^引入激勵(lì)光束位置優(yōu)化，從模擬和實(shí)驗(yàn)中研究了三種指定的共振模式，呈現(xiàn)了可比較的振幅響應(yīng)。使用QCL、ICL和DFB激光器作為激發(fā)光源，同時(shí)測(cè)量CO、CH4和C2H2，展示了多氣體檢測(cè)的能力。
 

　　A T-type photoacoustic cell was designed and verified to be an appropriate sensor, due to the resonant frequencies of which are determined jointly by absorption and resonant cylinders. The three designated resonance modes were investigated from both simulation and experiments to present the comparable amplitude responses by introducing excitation beam position optimization. The capability of multi-gas detection was demonstrated by measuring CO, CH4 and C2H2 simultaneously using QCL, ICL and DFB lasers as excitation sources respectively.
 

　　圖片顯示了配備了T型光聲池的基于PAS的多組分氣體傳感器配置的示意圖。使用三個(gè)激發(fā)激光器作為激光源，包括DFB ICL(HealthyPhoton，型號(hào)HPQCL-Q)、DFB QCL(HealthyPhoton，型號(hào)QC-Qube)和NIR激光二極管(NEL)，分別在2968 cm−1、2176.3 cm−1和6578.6 cm−1處發(fā)射，以實(shí)現(xiàn)對(duì)CH4、CO和C2H2的同時(shí)檢測(cè)。ICL、QCL和NIR激光二極管在目標(biāo)吸收波長(zhǎng)處的光功率分別為8 mW、44 mW和32 mW，通過熱功率計(jì)(Ophir Optronics 3 A)進(jìn)行測(cè)量。所有激光源都通過調(diào)節(jié)電流和溫度控制來驅(qū)動(dòng)。
 

　　A schematic diagram of PAS-based multi-component gas sensor configuration equipped with the developed T-type PAC is shown in Fig. Three excitation laser sources, including a DFB ICL (HealthyPhoton, model HPQCL-Q), a DFB QCL (HealthyPhoton, model QCQube) and an NIR laser diode (NEL) emitting at 2968 cm−1, 2176.3 cm−1 and 6578.6 cm−1, were employed to realize the simultaneous detection of CH4, CO and C2H2. The optical powers of the ICL, QCL and NIR laser diode measured by a thermal power meter (Ophir Optronics 3 A) at the target absorption lines were 8 mW, 44 mW and 32 mW, respectively. All the laser sources were driven by tuning the current and temperature control.
 

　　Fig.The schematic diagram of multi-resonance PAS-based gas sensor configuration equipped with the developed T-type PAC for multi-component gas simultaneous detection. Operating pressure: 760 Torr.
 

HealthyPhoton, model HPQCL-Q
 

HealthyPhoton, model QCQube
 

　　結(jié)論
 

　　建立了基于T型光聲池的多共振光聲光譜氣體傳感器，并驗(yàn)證其能夠進(jìn)行多組分同時(shí)檢測(cè)，達(dá)到ppb級(jí)別的靈敏度。通過有限元分析(FEA)模擬優(yōu)化和實(shí)驗(yàn)光束激發(fā)位置設(shè)計(jì)，三個(gè)指定的諧振頻率的光聲響應(yīng)相互比較，確保了同時(shí)檢測(cè)多種微量氣體的高性能。選擇了CO、CH4和C2H2這三種可燃?xì)怏w作為目標(biāo)氣體，使用QCL(4.59 µm，44 mW)、ICL(3.37 µm，8 mW)和NIR激光二極管(1.52 µm，32 mW)作為入射光束進(jìn)行同時(shí)檢測(cè)驗(yàn)證。F1模式下，光束照射到緩沖腔體壁上，信噪比(SNR)相比通過吸收?qǐng)A柱體的情況提高了4.5倍。實(shí)驗(yàn)得到了CO、CH4和C2H2的最小檢測(cè)限(1σ)分別為89ppb、80ppb和664ppb，對(duì)應(yīng)的歸一化噪聲等效吸收系數(shù)(NNEA)分別為5.75 × 10−7 cm−1 W Hz−1/2、1.97 × 10−8 cm−1 W Hz−1/2和4.23 × 10−8 cm−1 W Hz−1/2。對(duì)濕度交叉敏感性進(jìn)行改進(jìn)的研究提供了對(duì)光聲光譜傳感器在濕度松弛相關(guān)效應(yīng)方面的更好理解。利用單個(gè)光聲腔體和單個(gè)探測(cè)器進(jìn)行多組分氣體傳感的這種開發(fā)的光聲光譜模式，具有在電力變壓器故障的早期診斷方面的獨(dú)特潛力。
 

　　Conclusions
 

　　A T-type cell based multi-resonance PAS gas sensor was established and verified to be capable of multi-component simultaneous ppb-level detection. By the FEA simulation optimization and experimental beam excitation position design, the PA responses of the three designated resonant frequencies are comparable which guarantees the high performance of multiple trace gas detection simultaneously. The three combustible species of CO, CH4 and C2H2 were selected as target gases for the simultaneous detection verification using a QCL (4.59 µm, 44 mW), an ICL (3.37 µm, 8 mW) and a NIR laser diode (1.52 µm, 32 mW) as incident beams. The SNR for F1 mode with the beam irradiating on the buffer wall was increased by 4.5 times than that of passing through absorption cylinder. The experimental MDLs (1σ) were achieved as of 89ppb (CO), 80ppb (CH4) and 664ppb (C2H2) have been acquired, respectively, corresponding to the NNEA coefficients of5.75 × 10−7 cm−1 W Hz−1/2, 1.97 × 10−8 cm−1 W Hz−1/2 and 4.23 × 10−8 cm−1 W Hz−1/2. An improved humidification investigation regarding cross-sensitivity analysis provides a better understanding of PAS sensors in humidity relaxation related effects. This developed PAS modality of utilizing a single PAC and a single detector for multicomponent gas sensing exhibits unique potential for early diagnosis of power transformer failures.
 

　　Simulated spectral distribution characteristics of CO, CH4 and C2H2 based on HITRAN Database. Temperature and pressure: 296 K and 1 atm respectively.
 

　　Schematic structure of the developed T-type PAC.
 

　　Simulated sound pressure distribution of T-type PAC model for the three selected resonance modes by FEA method. Color bar: Simulated sound pressure (Pa).
 

　　Simulation results of the T-type PAC acoustic characteristics with the incident beam position optimization. (a) and (b): Two different incident ways of the excitation beam; (c), (d) and (e): The simulated pressure amplitude response vs. frequency for F1, F2 and F3 detection, respectively.
 

　　The experimental results of PA signals for different resonance modes by scanning the incident excitation beam. (a) Schematic diagram of the light source scanning process in the T-type PAC. Dashed line: Central axis. (b) The PA amplitude of 100 ppm CO vs. the beam position of ICL source. (c) The PA amplitude of 50 ppm CH4 vs. the beam position of ICL source. (d) The PA amplitude of 50 ppm C2H2 vs. the beam position of DFB laser diode. Insert: The irradiated surface of PAC.
 

　　The experimental results for CH4 detection with the incident beam position optimization. (a) Two different ways (I1, I2) of incident excitation beam using ICL for CH4 measurement; (b) The PA amplitude vs. frequency of F1 for the two incident ways; (c) The PA spectra of 100 ppm CH4 in the ICL tunning range using both incidence ways; (d) The PA signal amplitude of CH4 vs. gas concentration for two incidence ways.
 

　　Schematic of the improved humidification system for humidity control.
 

　　Reference
 

　　Le Zhang, Lixian Liu, Xueshi Zhang, Xukun Yin , Huiting Huan, Huanyu Liu, Xiaoming Zhao, Yufei Ma, Xiaopeng Shao,T-type cell mediated photoacoustic spectroscopy for simultaneous detection of multi-component gases based on triple resonance modality,Photoacoustics 31 (2023) 100492.
 

　　https://doi.org/10.1016/j.pacs.2023.100492