Non-intrusive sensing of air velocity, humidity, and temperature using tunable diode laser absorption spectroscopy
This work will report the non-intrusive sensing of air velocity, humidity, and temperature using tunable diode laser absorption spectroscopy (TDLAS), and discuss the potential applications of such sensors for in situ monitoring and active control for wind energy. The sensing technique utilizes the absorption features of water vapor in ambient air to monitor multiple flow parameters including velocity, humidity, and temperature simultaneously and non-intrusively [1-3]. The TDLAS technique does not require pre-calibration or seeding and extensively employs fiber-optics technologies to facilitate its implementation. As a result, the sensor enjoys advantages such as low maintenance cost and scalability, which are especially attractive characteristics for practical deployment in power generation systems such as wind turbines. In this work, we will discuss the fundamentals of TDLAS technologies and the results obtained in a series of laboratory demonstrations. Figure 1 shows a schematic to illustrate the concept of TDLAS. The output of a laser diode (both in terms of it power and wavelength) was changed by modulating its driving current (for example, a 2 kHz ramp signal in this work). As a result such modulation, the wavelength of the output scans across a certain spectral range (near the 1392 nm for example in this work) to detect an absorption line of water vapor. Typically, the output of the laser were split into three parts using fiber couplers. The first part, a small portion (10% in this work) of the output, was fed into a Mach-Zehnder interferometer (MZI) to monitor the wavelength scan during the modulation as shown. The rest of 90% of the output was then split into two equal parts to be used as the probe beams. The probe beams are pitched at an angle and directed into the target flow as shown. The transmitted laser beams are detected by two photodiodes at the opposite side of the flow, whose signals are then collected by a data acquisition systems for further analysis. In the signal analysis, the magnitude of the absorption peak was used to determine the concentration of the water vapor in the measurement region, the relative strengths of two absorption peaks to determine the temperature, and the Doppler shift between the two beams to determine flow velocity . A set of example measurements is shown in Figure 2 below to illustrate the data analysis mentioned above. The results show an absorption peak near 1392 nm, measured by two probe laser beams, one crossed the flow at -45 degrees (slide blue line) and the other at 45 degrees (dashed red line). As mentioned above, the absolute magnitude of such absorption signals (defined as absorbance as shown in Figure 2) is used to determine the concentration of water vapor. As shown in Figure 2, the absorption signals peak at different wavelengths because the Doppler shift caused by the flow. This Doppler shift (Δν) was calculated to be 4 x 10^(-4) cm-1 for this measurement as shown in the inset of Figure 2, based on which the flow velocity was determined to be 12.4 m/s. In this measurement, the target flow (which was generated by a small open jet tunnel) was also characterized by hotwire for comparison purposes. Figure 3 shows the flow velocity measured at multiple locations by the hotwire. We can see the velocity followed a distribution in the measurement region. The velocity distribution peaked at 20 m/s at the center of the flow, and the arithmetic mean of distribution was 11.7 m/s. In comparison, the TDLAS technique determines a line-of-sight integrated measurement of 12.4 m/s. In summary, the above example demonstrated the advantages of the TDLAS technique for monitoring multiple flow properties simultaneously and non-intrusively. Due to the advancement and maturity in diode lasers and fiber technologies, such technique is especially attractive for in situ application in practical systems.