Development of a tunable diode laser absorption sensor for online monitoring of industrial gas total emissions based on optical scintillation cross-correlation technique

Zhirong Zhang; Tao Pang; Yang Yang; Hua Xia; Xiaojuan Cui; Pengshuai Sun; Bian Wu; Yu Wang; Markus W. Sigrist; Fengzhong Dong

doi:10.1364/OE.24.00A943

1. Introduction

Over the past few years, the particulate matter and gas pollution produced by industry have seriously affected the air quality and consequently human health. For most countries in the world, this situation will continue for a long time. Therefore, the monitoring and controlling of air pollution have become an arduous task for the government and local administration departments. In order to control gas emissions, continuous online monitoring is required. However, many techniques are still under development and have not been fully commercialized yet. For instance, instruments with high performance for both online monitoring of gas flow velocity and gas total emission produced by an individual factory are still immature. If one could simultaneously measure the gas flow velocity and the gas concentrations in a pipeline, industrial emissions could be determined in real-time. This would be beneficial for the implementation of global environmental protection policies for the reduction of gas pollution and for a general environmental management [1].

There are a number of commonly used non-optical methods to detect these emission parameters, consisting of well known “traditional methods”, such as: pressure gauges, thermometers, thermocouple, Pitot tube and electrochemical gas sensor, and so on. These traditional methods have been developed in a long time. And they are easy to install and accessible, although they are limited in terms of meeting the industrial needs described above because of slow response, low sensitivity, complicated pretreatment, etc.. In contrast, optical spectroscopic techniques can work well without preconditioning of the gas. Therefore, they have the potential of in situ measurements with fast response and are ideally suited for industrial applications with sufficient sensitivity and selectivity. The first commercial TDLAS gas monitor was introduced on the market in 1995 using the trademark laser gas by Norsk Elektro Optik company. High-resolution tunable diode laser absorption spectroscopy (TDLAS) represents this kind of powerful and generally accepted technology for selective and sensitive gas sensing, and is widely used in various areas [2–4]. As a result, there are numerous applications of TDLAS, such as: gas pressure monitoring [5], vehicle emissions [6], gas exhaust temperature monitoring [7], carbon isotope measurements [8], and so on [9–14].

Optical scintillation is generated when a light beam passes through a turbulent medium. This phenomenon has found applications ranging from particulate emission monitoring or rain-parameter determination to wind-velocity measurements [15–17]. Using optical scintillation cross-correlation (OSCC) to measure the gas flow velocity has great advantages compared with other traditional techniques, such as Pitot tube, hot wire anemometry or laser Doppler velocimeter (LDV). The widely used Pitot tube is a differential pressure velocimeter and can only perform single point measurements [18]. Hot wire anemometry is widely used in turbulence measurements [19], but the instrument is rather brittle and exhibits a poor measurement stability. The LDV is commonly used in multidimensional flow measurements [20], but the accuracy is easily affected by pipe-vibrations. Up to the present, the optical flowmeter OFS-2000 developed by Wang [21] is the only commercial product for gas velocity measurements. However, this OFS device is not applicable when the temperature fluctuation within the measurement area is small or even negligible. Recently, we have developed a new kind of optical flow sensor which is similar in design to Wang’s device and based on the low-frequency signal. Therefore this new optical flow sensor can measure the flow velocity and particle concentration simultaneously. The Intergovernmental Panel for Climate Change (IPCC) lists methods, manual sampling and measuring method, material balance calculation method and decomposition analysis for gas emissions gross measurements are currently more practical methods. But there are still many limitations, such as estimation errors, time consumption, high cost, etc.. The biggest drawback is the lack of real-time representation and thus the missing of the dynamic impact. In summary, it is necessary to develop a more accurate, economical and fast detection scheme for online monitoring of industrial gas gross emissions.

In recent years, China has also increased the efforts to reduce emissions of pollutants. In this paper, we describe the development of a sensor for online monitoring of industrial gas total emission with TDLAS for gas concentration and OSCC for gas flow velocity. Distributed feedback (DFB) diode laser are an appropriate light sources due to their much high modulation bandwidth and significant output power. We successfully developed a TDLAS system for O₂ concentration measurements, because oxygen is a significant gas in steel making process, coking stage, combustion control, etc.. At the same time, we introduce a new instrument to measure the flow velocity, which uses the OSCC signals, and adopts a structure of two parallel light paths. This paper will first briefly describe the operational principles based on TDLAS and OSCC technologies. Then, the application of combining TDLAS with OSCC to detection gas emission gross is also briefly introduced. Finally, experimental results from field test are presented and discussed. These recent results demonstrate that this combined system for industrial pipeline applications has many advantages, such as non-intrusive character, easy maintenance, long service life as well as application also in inflammable and other hazardous environments. Remarkably, the optical scintillation method can also be used to measure the particle density at the same time.

2. Basic principles

2.1 Gas concentration measurement - TDLAS

When light is passing through flue gases, various factors can reduce the light intensity. The Beer-Lambert law describes the part which is related to the gas absorption. The law relates the incident intensity I₀ with the transmitted intensity I by

I = I_{0} \exp (- k L)

here k is the absorption coefficient and L denotes the path length. In the near infrared area, the gas absorption coefficient is usually very small, i.e., kL≤0.05 [22]. Equation (1) can thus be simplified as,

I = I_{0} (1 - k L) = I_{0} [1 - σ (ν) C L]

where

σ (ν)

denotes the absorption cross section at frequency ν, and C stands for the gas concentration. For detecting gas concentrations, wavelength modulation spectroscopy (WMS) is often adopted. The intensity of the second harmonic (2f) signal can be expressed as [23],

I_{2 f} \propto I_{0} σ_{0} C L

If the reference signal and nonlinear least square multiplication method are introduced to fit the target gas 2f signals [24], Eq. (3) may be rewritten as,

C_{M e a} = a \frac{I_{M e a} C_{R e f} L_{R e f}}{I_{R e f} L_{M e a}}

where a is fitting coefficient;

C_{M e a}

and

C_{R e f}

are the concentrations of the target gas to be measured and reference gas in the calibration cell, respectively;

I_{R e f}

,

I_{M e a}

are the intensities of the two split laser beams;

L_{R e f}

and

L_{M e a}

denote the length of the calibration cell and the measurement optical path, respectively. In general, we can use the ratio of the 2f and 1f signals to cancel any laser intensity differences. In this case, we could easily obtain the concentration from the following equation with the ratio values,

C_{M e a} = \frac{{(\frac{I_{2 f}}{I_{1 f}})}_{M e a} C_{R e f} L_{R e f}}{{(\frac{I_{2 f}}{I_{1 f}})}_{R e f} L_{M e a}}

where

{(\frac{I_{2 f}}{I_{1 f}})}_{R e f}

and

{(\frac{I_{2 f}}{I_{1 f}})}_{M e a}

represent the 2f/1f ratio value of the reference and target gas signals, respectively. In most cases, the reference gas cell and the gas in the measurement pipeline were at different temperatures and pressures. We should measure the internal temperature and pressure of the pipeline in real time and modify these effects according to the line-strength temperature relationship and pressure broadening relationship based on the HITRAN database.

2.2 Gas velocity measurement - OSCC

In urban areas, industrial stack gases are one of the major sources of particulate matter and pollution in the atmosphere. Using optical scintillation of stack gas flow to measure velocity offers greater advantage than some traditional techniques as mentioned above. However, the corresponding theory is not consummate yet. The proposed approach with two parallel optical beams originating from two LEDs for stack gas flow velocity measurements is shown in Fig. 1. Each of the two identical light beams is a divergent spherical wave, both propagate along the X-axis and the origins are at x = 0. The diameters of the two transmitting lenses are both D_t, and the diameters of the two receiving lenses are both D_r. The distance between transmitter and receiver is L, and the distance between the two receivers is l. The direction of gas flow is along the Y-axis, and the mean velocity is v. The system with two point optical emitters (LEDs) and two point photon receivers is assumed to be in one plane and the two transmitting beams are absolutely parallel (or minimal errors).

Fig. 1 Layout of optical scintillation measurement.

Download Full Size | PDF

The flue gas composition in industrial pipeline is complex. The number of particles appears fluctuations in a statistical sense, and leads to the fluctuation of the optical signal with time, when the particles random move in and out the field of view. The correlation time of the scintillation is approximately equal to the time that the particle moves across the lens. As a result, the characteristic frequency (CF) of the optical scintillation [25] is,

f = \frac{v}{D_{r}}

Suppose that the extinction coefficient of stack gas flow is $α^{'} (r, t)$ and the received logarithmic light intensity is lnI(t) [26], then

\ln I (t) = \ln < I > - \int_{0}^{L} α^{'} (r, t) d x

where

< * >

is the assemble average, r is the position vector and the term

α^{'} (r, t)

stands for the perturbation part. For homogeneous isotropic time stationary turbulence, the correlation function is only relative to the distance between the two receivers and the time delay. Then the cross-correlation function becomes,

C_{\ln I} (l, τ) = \int_{0}^{L} \int_{0}^{L} R_{α} (| r_{1} - r_{2} |, τ) d x_{1} d x_{2}

where

τ

is time delay between these both receiving lenses, and

R_{α} (r_{1} - r_{2}, τ)

is the correlation function of the extinction coefficient. Because of the movement of the stack gas along the Y-axis, the Taylor frozen turbulence hypothesis, and the geometric layout shown as Fig. 1, we get

R_{α} (r_{1} - r_{2}, τ) = R_{α} (x_{1} - x_{2}, l - v τ, 0)

After a series of complex mathematical derivations and transformations (specific processes can be found in [27]), the cross-correlation function can be expressed as,

C_{\ln I} (l, τ) = C_{α}^{2} \int_{0}^{L} (L - x) (L_{0}^{2 / 3} - r^{2 / 3}) d x, (r = \sqrt{x^{2} + {(l - v τ)}^{2}})

where L₀ is the outer-scale of the turbulence. Numerical computer simulations of Eq. (10) are shown in Fig. 2. The time delay at the peak of the cross-correlation function is 0.1 s, which is just equal to

l / v

with l = 0.5 m and v = 5 m/s. The result illustrates that it is feasible to calculate the velocity by the OSCC function.

Fig. 2 The numerical computer simulations.

Download Full Size | PDF

The reasons of optical scintillation caused by a light beam passing through gas flow are very complex. Particles move in or out the view of sight in random and will induce optical intensity fluctuations, but it is difficult to receive the scintillation signals in an industrial environment. Turbulence causes the optical scintillation, the frequency of this kind of scintillation commonly reaches hundreds or thousands of Hz. Particle concentration fluctuations at random will also induce optical scintillation, the frequency is commonly less than hundreds of Hz. The low frequency part of optical scintillation can be used to measure gas flow velocity and particle concentration simultaneously. Many methods have been suggested or demonstrated for wind measurements by the temporal cross correlation of the optical scintillation [17], such as the peak technique, slop technique, frequency technique, Briggs technique, and covariance technique. Among these methods, the peak technique is the most resistant to interference, exhibits a better time resolution and is easier to implement than the other methods [28]. The average velocity along the optical path can be obtained by the following equation,

\bar{v} = \frac{l}{τ}

the size of the particulate matter inside an industrial pipeline varies, so the optical scintillation signal contains the influence of different scales. If, the non-uniform scale is defined as $ξ$ , the correlation time of the scintillation [25] can be rewritten according to Eq. (6),

f \approx \frac{v}{ξ + D_{r}}

above equation indicates that the smaller the scintillation frequency, the larger the non-uniform scale and vice versa. Therefore, the effect of large scale and small scale nonuniformity in the optical scintillation signal can be removed according to the CF, Eqs. (6) and (12).

2.3 Total emissions calculation – TDLAS & OSCC

Total gas amount G is the time integral of flow rate Q within the time T,

G = \int_{0}^{T} Q (t) d t

hence, the total gas emission quantity M can be expressed as,

M = \int_{0}^{T} m (t) \cdot Q (t) d t

where m is the gas mass concentration in the pipeline. According to Eqs. (13) and (14), the total gas emission quantity depends on the measurement of the real-time flow rate. As discussed in section 2.2, the LED emits a spherical light wave which is then constrained to a parallel beam by a large aperture lens. The parallel light beams pass through the gas flow and arrive at the detector. The light signal received at the detector is the result of the integral of the extinction along the entire optical path. But for the actual viscous flow, the flow velocity is different at different positions across the pipe section. Therefore, the velocity is a function of the space position. The correlation function of the optical scintillation signal can be written as,

C_{l n I} (l, τ) = 2 \int_{0}^{L} (L - x) R_{α} (x, l - v (x) τ) d x

Accordingly, the correlation function of the system structure contains the information of the flue gas velocity at different positions. The optical scintillation measurement of the velocity yields the average velocity along the light path. When the upstream straight pipeline is long enough, the gas fluid flow is fully mixed and developed and the cross section velocity exhibits the Centro-symmetric distribution. The flow rate can thus be expressed as,

Q_{T} = v \cdot s

where

v

is the average velocity, and

s

is the pipeline section area. However, the actual industrial pipeline generally differs from a long straight pipeline, and the velocity distribution is also affected by various other factors than the pipeline geometry. Hence, the velocity distribution is no longer symmetric. The correction coefficient for the velocity distribution

ω

is introduced in order to improve the accuracy of the gas flow rate measurement. The flow rate calculation formula becomes,

Q = ω \cdot v \cdot s

The correction factor $ω$ can be obtained by additional flow measurement methods such as multi-points measurements or multi-line measurements. Finally, the gas total emission E can be written as,

E (K g / s) = {\frac{[Q (m^{3} / s) \cdot C_{g a s} (%) \cdot 1000 (L / m^{3})]}{22.4 (L / m o l) \cdot 1000 (g / k g) \cdot t (s)}} \cdot M (g / m o l)

where C_gas denotes the gas concentration (identical to C_mea of Eq. (5)) derived from the TDLAS measurements.

3. Measurement systems

The total emission measurement system essentially contains two parts: the gas concentration monitoring system and the velocity measurement system. The schematic diagram of the online total emission measurement experimental setup is shown in Fig. 3. The TDLAS technique with modulation for the gas concentration measurement has been described in details elsewhere (see e.g [1, 2, 7].). The DFB diode laser is tuned to the rotational-vibrational line of the target oxygen gas at 760 nm by a homemade current and temperature controller. The laser wavelength is scanned across the selected absorption line by a saw-tooth signal at a low frequency of 10 Hz and simultaneously modulated by a sinusoidal signal at a high frequency of 10 kHz from the homemade Signal Generator. The modulated laser beam is divided into two parts with a 1 × 2 fiber splitter. One arm (10%) is used to pass through a 5 cm calibration cell as a reference signal for locking the wavelength and compensating for intensity fluctuations whereas the other arm (90%) is used to measure the pipeline gas concentrations. Both laser beams are collimated and collected by two coincident photodetectors after passing through absorption gases. These two current signals are then transmitted into the homemade digital lock-in-amplifier to gain the first and second harmonic signals with 1024 points each period [29]. And this sampling width contains 3-5 times absorption linewidth to avoid the wavelength drift. The lock-in time constant is set at 0.3 ms to compromise the amplitude attenuation and obtain the best signal-to-noise ratio. In this combination system, the CaF₂ lens and nitrogen purging are used to combat continual high temperatures or dust influence.

Fig. 3 The schematic diagram of the online total emission measurement experimental setup.

Download Full Size | PDF

The gas flow velocity in the pipeline is recorded by a sensor based on the low frequency part of the double minimal errors parallel path OSCC as described in section 2.2. Two LED light sources are modulated at 10 kHz and emit ideal Gauss spherical waves at a wavelength of 630 nm and an output power of >1 W. The receivers are silicon photoelectric diodes (Hamatsu-S5106). The received signals are amplified, filtered by band pass and low pass filters to obtain the two signals correlation times. The digital voltmeter is used to observe the signal amplitudes. A pressure and temperature sensor is introduced for modifying the temperature and pressure influence for gas concentration monitoring. At last, the above three signals (two signals for OSCC & one signal for TDLAS) are collected by a data acquisition card and processed by a computer. These two systems are combined into one complete sensor for in situ monitoring industrial gas total emissions. Figure 4 shows two photographs of the two independent measuring set ups. We anticipate to combine them into one non-invasive system in the future to achieve the concentration and velocity measurement, simultaneously.

Fig. 4 Photographs of the (a) WMS gas concentration measurement system and (b) OSCC velocity measurement system. The two systems are combined into one device thus reducing the volume and improving the easiness of operation.

Download Full Size | PDF

4. Results and discussions

In order to demonstrate the feasibility of the sensing device, measurements were carried out at a special steel plant. The pipeline section is rectangular 1.5 m*0.6 m. The velocity of the pipeline gas flow can be calibrated with a commercial Pitot tube (CP300), the temperature is about 110 °C, and the particle diameters and number density can also be measured by multi-channel optical particle counter (AE3321). The distance between transmitter and receiver corresponds to the length of the stack of 1.5 m, and the distance between the three receivers (two for OSCC & one for TDLAS) is 0.3 m. The transmitter apertures and receiver apertures diameters are 30 mm (as shown in photograph B of Fig. 4). In Fig. 5 the received data from both receivers (channels A and B) are plotted versus time. The temporal variations in both channels are similar to those trends as seen in Fig. 5(a). Based on the correlation of the two signals the maximum relative time delay is found to be approximately $4 \times 10^{- 4} s$ . However, this is obviously not true for the calculated velocity 10³ m/s according to Eq. (11) as seen in Fig. 5(b). The main reason causing this error is because both the low frequency (the large scale turbulence vortex) and high frequency (the small scale turbulence vortex) increase the value of the zero delay and offset the maximum cross correlation time delay. Therefore, the high frequency CF caused by refractive index fluctuation is estimated about 267 Hz with Eq. (6). And also, if we take pipe section length 1.5 m as the minimum large scale concentration nonuniform element ( $ξ$ ), the CF is about 5 Hz with Eq. (12). Meanwhile, the distance between the two light paths is 0.3 m, so the CF to be extracted is about 24 Hz. Also, if we take the lens diameter of 0.03 m as the maximum small scale concentration nonuniform element ( $ξ$ ), the CF is far greater than 133 Hz with Eq. (12). Therefore, it is crucial to remove the interference between low frequency and high frequency components. Above all, the applied empirical value of 30 Hz low-pass filter is introduced and remove the low frequency components below 5 Hz for pre-processing the data. In other words, the filter frequency can be determined by the diameter of lens, the pipe dimensions and gas velocity inside of the pipeline regardless of shape of the pipe, pressure and temperature. Meanwhile, the temperature and pressure will not change acutely under industrial conditions. Then, the OSCC is carried out by using the polar method [30] as shown in Fig. 5(c), the maximum relative time delay is about $4.1 \times 10^{- 2} s$ , the calculated velocity mean value is 7.3 m/s in Fig. 5(d). The research result confirms to the actual status and has practical significance.

Fig. 5 The channel A and channel B original and processed optical scintillation signals within 5.5 s form data acquisition card. (a) the collected two sets signals; (b) cross correlation data, maximum is 4 × 10⁻⁴ s, v = 1000 m/s;(c) the band-pass filtered signals; (d) cross correlation data, maximum is 4.1 × 10⁻² s, v = 7.3 m/s.

Download Full Size | PDF

Figure 6 shows the long term results of continuous measurements of the gas flow velocity with the Pitot tube compared to the OSCC data in the industrial pipeline. The OSCC data represent the average value: in space, the results correspond to the average velocity along the light path; in time, the velocity is the average value within 30 s. As a result, the velocity record appears relatively smooth, and the averaged value is about 8 m/s. However, the Pitot tube measures single point data. The results shown in the Fig. 6 are only for the central point velocity within 1 s. Hence, its long time behavior changes more abruptly than the OSCC curve. Furthermore, the measurement data may be affected by the insertion of the Pitot tube into the pipeline. Due to the differences in the measurement principle and application mode, the measurement results are not exactly the same for a given time. The mean velocity of the OSCC and Pitot tube measurements are 7.59 m/s and 8.20 m/s, respectively. These values agree with the reality that the velocity is higher at the center of pipeline where the Pitot tube recording takes place. The statistical standard deviations of the gas velocity are 1.37 m/s and 1.47 m/s, respectively, i.e. above the deviation between the two mean values. And also, these results demonstrate that the velocity derived from the optical scintillation method is more stable than the Pitot tube measurements. Furthermore, the OSCC device may need to be calibrated with a more accurate measurement instrument than offered by the Pitot tube.

Fig. 6 The continuous measurement results of gas flow velocity with Pitot tube (upper panel) and OSCC method (lower panel).

Download Full Size | PDF

Owing to the harsh ambient conditions in the pipeline, the harmonic signals are inevitably distorted by the scattering from dust or vapor besides the gas absorption itself. Since the O₂ absorption is rather weak at the wavelength of 760 nm, the determination of the gas concentration by the traditional Beer-Lambert law from the 2f signals of the reference and target gas would certainly be inaccurate if the influence by other factors would not be taken into account. Hence, we collect temperature and pressure data in additon to 2f/1f harmonic signals to improve the anti-interference ability and accuracy. The long term monitoring results are shown in Fig. 7(a). The oxygen concentrations changing trend can be seen clearly. So, the technical staff can adjust the amount of oxygen in real time according to measurement. As discussed above in section 2.3, the correction coefficient $ω$ in Eq. (17) can be calculated by the Pitot tube and OSCC mean values resulting in $ω=1 .08$ . Thus, the flow rate calculation Eq. (17) can be expressed as $Q (m^{3} / s) = 1.08 \cdot v_{o s c c} \cdot S_{S e c t i o n}$ . The unit time gas total emissions E are finally obtained according to Eq. (18) based on the flow rate Q and the concentration C, as illustrated in Fig. 7(b).

Fig. 7 Continuous measurement results of oxygen. (a) the oxygen concentration derived from TDLAS; (b) the oxygen flow rate based on Eq. (18) and OSCC measurement data.

Download Full Size | PDF

Furthermore, it is worth noting that the low-frequency OSCC method can also be used to monitor the particle concentrations. However, this paper focuses on the gas total emission rather than on the particle aspect. Therefore, the particle concentration calculation method describe in [27] is not outlined here. Distribution measurements of the particle number density versus particle size in air and four different measurements in industrial flue gas all obtained with a commercial multi-channel optical particle counter are depicted in Fig. 8(a). It can be seen that the particle size in an industrial environment is larger than in normal air, mostly around 10 μm. Continuous measurements of particle concentrations in flue gas are shown in Fig. 8(b).

Fig. 8 (a) Measurements of the distribution of particle number density vs particle size in air and 4 measurements in flue gas; and (b) continuous measurement results of particle concentrations.

Download Full Size | PDF

5. Conclusions

TDLAS can be used to monitoring gas concentrations and the OSCC can be employed to monitor the gas velocity and particle concentration. Above two techniques are not related with each other, seemingly. Actually, in order to achieve the gas total emission measurement, we have done some researches to combine these two techniques into one system to monitor the concentration and velocity simultaneously. However, it should be emphasized that the measurement results for online monitoring of industrial emissions reported here are still in an early stage. The reasons for optical scintillation caused by a light beam passing through a stack gas flow are very complex. Further work is needed to fully explore the potential of the technique. For instance, the influence of pipeline shape and size represents a very detailed study aspect and the derivation of the optimum distance between the three light paths used needs further theoretical and experimental studies. Furthermore, a new complex theory based on path-weighted function and average gas flow velocity to calculate the total emissions with the help of gas concentrations needs to be derived and developed. However since nowadays there are no instruments available to verify the measurement accuracy. We also aim to expand the capabilities of our approach for simultaneous measurements of particle matter density and concentrations of other gases like CO, CO₂, NH₃, H₂S, CH₄, and so on. This would be particularly useful in the field of pollutant gas emissions gross control or industrial production process cost reduction. In summary, the combination of these two methods provides a new tool for accurate measurements of total emissions, and it is expected that the feasibility of more versatile instruments will be demonstrated in the future.

Acknowledgments

This research is financial supported by the National Key Technology Research and Development Program of the Ministry of Science and Technology of China (Grant No. is 2014BAC17B03); the National Natural Science Foundation of China (Grant No. is 11204320, 11204319, 41405034), the Special Fund for Basic Research on Scientific Instruments of the Chinese Academy of Science (Grant No. is YZ201315) and Chinese academy of sciences president’s international fellowship initiative (PIFI, No. is 2015VMA007).

References and links

1. F. Z. Dong, W. Q. Liu, Y. N. Chu, J. Q. Li, Z. R. Zhang, Y. Wang, T. Pang, B. Wu, G. J. Tu, H. Xia, Y. Yang, C. Y. Shen, Y. J. Wang, Z. B. Ni, and J. G. Liu, Real-Time in situ Measurements of Industrial Hazardous Gas Concentrations and Their Emission Gross (InTech Publisher, 2011), Chap. 5.

2. Z. R. Zhang, H. Xia, F. Z. Dong, T. Pang, B. Wu, P. S. Sun, G. X. Wang, and Y. Wang, “Simultaneous detection of multiple gas concentrations with multi-frequency wavelength modulation spectroscopy,” Europhys. Lett. 104(4), 44002 (2013). [CrossRef]

3. O. Witzel, A. Klein, C. Meffert, S. Wagner, S. Kaiser, C. Schulz, and V. Ebert, “VCSEL-based, high-speed, in situ TDLAS for in-cylinder water vapor measurements in IC engines,” Opt. Express 21(17), 19951–19965 (2013). [CrossRef] [PubMed]

4. C. Liu, L. Xu, J. Chen, Z. Cao, Y. Lin, and W. Cai, “Development of a fan-beam TDLAS-based tomographic sensor for rapid imaging of temperature and gas concentration,” Opt. Express 23(17), 22494–22511 (2015). [CrossRef] [PubMed]

5. D. T. Cassidy and J. Reid, “Atmospheric pressure monitoring of trace gases using tunable diode lasers,” Appl. Opt. 21(7), 1185–1190 (1982). [CrossRef] [PubMed]

6. F. Z. Dong, W. Q. Liu, J. G. Liu, X. H. Tu, Y. J. Zhang, F. Qi, P. H. Xie, Y. H. Lu, S. M. Wang, Y. P. Wang, and Q. N. Wei, “On-line roadside vehicle emissions monitoring,” J. Test. Mea. Tec. 19(2), 119–127 (2005).

7. X. Liu, J. B. Jeffries, R. K. Hanson, K. M. Hinckley, and M. A. Woodmansee, “Development of a tunable diode laser sensor for measurements of gas turbine exhaust temperature,” Appl. Phys. B 82(3), 469–478 (2006). [CrossRef]

8. K. Tanaka and K. Tonokura, “Sensitive measurements of stable carbon isotopes of CO₂ with wavelength modulation spectroscopy near 2μm,” Appl. Phys. B 105(2), 463–469 (2011). [CrossRef]

9. A. D. Sappev, P. Masterson, E. Huelson, J. Howell, M. Estes, H. Hofvander, and A. Jobson, “Results of closed-loop coal-fired boiler operation using a TDLAS sensor and smart process control software,” Combust. Sci. Technol. 183(11), 1282–1295 (2011). [CrossRef]

10. A. Sepulveda-Jauregui, K. Martinez-Cruz, A. Strohm, M. W. A. Katey, and F. Thalasso, “A new method for field measurement of dissolved methane in water using infrared tunable diode laser absorption spectroscopy,” Limnol. Oceanogr-Meth. 10(7), 560–567 (2012). [CrossRef]

11. S. Svanberg, “Gas in scattering media absorption spectroscopy – from basic studies to biomedical applications,” Laser Photonics Rev. 7(5), 779–796 (2013). [CrossRef]

12. Y. Deguchi, T. Kamimoto, Z. Z. Wang, J. J. Yan, J. P. Liu, H. Watanabe, and R. Kurose, “Applications of laser diagnostics to thermal power plants and engines,” Appl. Therm. Eng. 73(2), 1453–1464 (2014). [CrossRef]

13. A. Pogány, A. Klein, and V. Ebert, “Measurement of water vapor line strengths in the 1.4–2.7µm range by tunable diode laser absorption spectroscopy,” J. Quant. Spectrosc. Radiat. Transf. 165, 108–122 (2015). [CrossRef]

14. D. W. Choi, M. G. Jeon, G. R. Cho, T. Kamimoto, Y. Deguchi, and D. H. Doh, “Performance improvements in temperature reconstructions of 2-D tunable diode laser absorption spectroscopy (TDLAS),” J. Therm. Sci. 25(1), 84–89 (2016). [CrossRef]

15. E. Ryznar, “Dependency of Optical Scintillation Frequency on Wind Speed,” Appl. Opt. 4(11), 1416–1418 (1965). [CrossRef]

16. A. Chen, J. Hao, Z. Zhou, and K. He, “Particulate Concentration Measured from Scattered Light Fluctuations,” Opt. Lett. 25(10), 689–691 (2000). [CrossRef] [PubMed]

17. T. I. Wang, G. R. Ochs, and R. S. Lawrence, “Wind measurements by the temporal cross-correlation of the optical scintillations,” Appl. Opt. 20(23), 4073–4081 (1981). [CrossRef] [PubMed]

18. R. Klopfenstein Jr., “Air Velocity and Flow Measurement Using a Pitot Tube,” ISA T. 37(4), 257–263 (1998). [CrossRef]

19. M. Y. Xu, C. Du, and J. C. Mi, “Centreline statistics of the small-scale turbulence of a circular jet and their dependence on high frequency noise,” Wuli Xuebao 60(3), 034701 (2011).

20. K. Maru and T. Hata, “Nonmechanical scanning laser Doppler velocimeter for cross-sectional two-dimensional velocity measurement,” Appl. Opt. 51(34), 8177–8183 (2012). [CrossRef] [PubMed]

21. T. I. Wang, “Optical flow sensor using fast correlation algorithm,” US Patent US20020145727 (2003–8-26).

22. J. Reid and D. Labrie, “Second-harmonic detection with tunable diode lasers–Comparison of experiment and theory,” Appl. Phys. B 26(3), 203–210 (1981). [CrossRef]

23. S. T. Wang, J. Li, R. S. Che, and T. Y. Wang, “A Methane Gas Sensor with Optic Fiber Based on Frequency Harmonic Detection Technique,” J. Appl. Opt. 25(2), 44–47 (2004).

24. R. F. Kan, W. Q. Liu, Y. J. Zhang, J. G. Liu, M. Wang, D. Chen, J. Y. Chen, and Y. B. Cui, “A high sensitivity spectrometer with tunable diode laser for ambient methane monitoring,” Chin. Opt. Lett. 5(1), 54–57 (2007).

25. A. Ishimaru, Wave Propagation and Scattering in Random Media (Wiley-IEEE Press, 1999).

26. W. Q. Liu, H. L. Liu, Z. Y. Zeng, and Y. Jiang, “Analysis of spectrum characteristics of optical scintillation in stack gas flow,” Chin. Phys. 15(8), 1777–1782 (2006). [CrossRef]

27. Y. Yang, F. Z. Dong, Z. B. Ni, T. Pang, Z. Y. Zeng, B. Wu, and Z. R. Zhang, “Theoretical Analysis of Stack Gas Emission Velocity Measurement by Optical Scintillation,” Chin. Phys. B 23(4), 040703 (2014). [CrossRef]

28. H. L. Liu, Z. Y. Zeng, and W. Q. Liu, “The study of stack gas velocity measurement using optical signal cross-correlation method,” Opt. Technol. 32(6), 920–925 (2006).

29. Z. R. Zhang, F. Z. Dong, Y. Wang, B. Wu, T. Pang, H. Xia, and G. Tu, “Online monitoring of industrial flue gases using tunable diode laser with a digital-control module,” Proc. SPIE 7853, 785313 (2010). [CrossRef]

30. J. Zhou, Z. F. Yuan, X. G. Pu, K. Wang, Y. G. Lu, and K. F. Cen, “Study of zero-crossing polarity correlation for velocity measurement of high temperature flue gas,” Proc. CSEE. 19(3), 11–13, 45 (1999).

Development of a tunable diode laser absorption sensor for online monitoring of industrial gas total emissions based on optical scintillation cross-correlation technique

Abstract

1. Introduction

2. Basic principles

2.1 Gas concentration measurement - TDLAS

2.2 Gas velocity measurement - OSCC

2.3 Total emissions calculation – TDLAS & OSCC

3. Measurement systems

4. Results and discussions

5. Conclusions

Acknowledgments

References and links

Cited By

Figures (8)

Equations (18)

Optics Express