定义：采用光纤装置的光学传感器。

光纤传感器是利用光纤装置敏感一些物理量的器件，通常是温度或机械应力，有时也敏感位移，振动，压力，加速度，旋转（例如基于Sagnac效应的光学陀螺仪测量），或者化学物质的浓度。该装置的基本原理是激光器（通常为单频光纤激光器）或超发光光源发出的光进入光纤中，光纤中或者光纤布拉格光栅中的参数发生轻微改变，然后到达探测器中测量出来该变化。 
与其它类型传感器相比，光纤传感器有下列优势： 

• 包含电学绝缘材料（不需要使用电缆），因此在高压环境中也可以使用。 
• 可以工作在易爆环境中，因为不会产生电火花，即使存在缺陷的情况下。 
• 抗电磁干扰，甚至附近的闪电袭击，并且它们也不会干扰其它装置。 
• 其材料具有化学钝性，即不会污染附近的物体也不会受到腐蚀。 
• 工作的温度范围很宽（比很多电子学装置宽很多）。 
• 具有多路复用能力：即在单根光纤中采用单个光源可以将多个传感器盘在一起。 

目录

• 布拉格光栅传感器
• 分布传感
• 准分布式传感
• 其它方案
• 应用

布拉格光栅传感器 
光纤传感器通常采用光纤布拉格光栅。很多光纤传感器的基本原理是布拉格光栅的布拉格波长（即最大反射率对应的波长）不仅与布拉格光栅周期有关，还与温度和机械应力有关。石英光纤中，布拉格波长对应力的响应比应力本身小20%，因为应力的直接效应在折射率减小的情况下也会相应减小。温度效应接近于热膨胀自身的结果。应力和温度效应可以用不同的技术分别开来（例如，利用一个不受应力的参考光栅，或者结合不同种类的光纤光栅），这样可以同时得出不同的值。对于单纯的压力传感，分辨率在几个με的范围内（相对长度变化是10−6的多少倍），准确率不会降低很多。对于动力学测量过程（例如，声学现象），在1 Hz带宽情况下灵敏度可以大于1nε . 

分布传感 
有些光纤传感器不采用光纤布拉格光栅，而是利用光纤本身。传感原理可以利用瑞利散射，拉曼散射或者布里渊散射。例如，光时域反射就是一种方法，其中采用一个脉冲探测信号光将弱的反射光局域化。还可以利用布里渊频移随温度或者应力的变化。 
有时测量的量是一长段光纤中的平均值。有些温度传感器就是这种情况，例如，Sagnac干涉仪用作陀螺仪。还有的测量随位置变化的量。这称为分布传感。 

准分布式传感 

• 根光纤可能包含很多光栅传感器组成一列来控制整个光纤中的温度和压力分布。这称为准分布式传感。有不同的方法将光栅放在光纤中的特定位置： 
• 在波分复用（WDM）技术或者光频率反射仪（OFDR）中，光栅具有不同的布拉格波长。可以采用波长可调激光器将波长调谐到某一光栅的布拉格波长，而最大反射率的波长得到应力或者温度的影响作用。或者，可以结合一个宽带光源（例如，超发光光源）和波长扫描探测器（例如，采用光纤法布里珀罗干涉仪）或CCD光谱仪。任何一种情况下，光栅数目最多在10-50之间，受限于光源的调谐范围或带宽和单位光纤光栅的波长间隔。 
• 还有一种技术为时分复用（TDM），也是利用弱反射光栅，与短的光脉冲结合。不同光栅的反射光由于到达时间不同可以区分开。时分复用通常与波分复用结合在一起为了提高信道的数目，可以得到几百甚至几千个信道。 
• 采用光开关可以选取不同光纤线，更进一步提高传感器的数目。 

其它方案 
除了以上所说的技术方案，还有一些其它的技术。例如： 

• 光纤布拉格光栅可以应用于干涉型光纤传感器中，其中它们仅作为反射器，并且测量的相移来自于其中的光纤跨度。 
• 还有布拉格光栅激光传感器，其中传感器光栅形成光纤激光器谐振腔的端反射镜，包含一些掺铒光纤，在光纤线中受到980 nm的泵浦光的泵浦。布拉格波长与温度或者应力有关，决定了激光器波长。这种方案可以进一步改善，得到很高的分辨率（由于光纤激光器的线宽很小）和很高的灵敏度。 
• 有的方案将一对布拉格光栅组成光纤法布里珀罗干涉仪，对外界影响非常敏感。法布里珀罗干涉仪也可以用其它方法制作，例如，光纤中可变的气体空隙。 
• 长周期光纤光栅可以用于多参数传感（例如，温度和压力）中，或者作为与温度无关的应力传感器。 

应用
尽管发展了很多年，光纤传感器在商业上仍未大量应用，因为很难取代目前的技术，虽然它们也有其自身的局限。但是在有些领域中，光纤传感器正逐渐的被大量接受。尤其是需要工作于很恶劣的环境中时，例如传感高压和高功率器械，或者在微波炉中。布拉格光栅传感器还可以用来控制条件，例如飞机的两翼中，涡轮机中，桥，大坝、油井和管道。集成光纤传感器也被称为“智能结构”；可以实现对内部条件的控制，获取结构中不同部分得到的压力、老化现象、振动等的有用信息。智能结构是光纤传感器今后发展的一个主要动力和方向。 

Definition: optical sensors based on fiber devices

Alternative term: fiber sensors

Fiber-optic sensors (also called optical fiber sensors) are fiber-based optical sensors for some quantity, typically temperature or mechanical strain, but sometimes also displacements, vibrations, pressure, acceleration, rotations (measured with optical gyroscopes based on the Sagnac effect), or concentrations of chemical species. The general principle of such devices is that light from a laser (often a single-frequency fiber laser) or from a superluminescent source is sent through an optical fiber, experiences subtle changes of its parameters either in the fiber or in one or several fiber Bragg gratings, and then reaches a detector arrangement which measures these changes.

One distinguishes intrinsic and extrinsic sensors. Intrinsic sensors are those where a fiber itself (possibly in a modified form, e.g. containing a Bragg grating) acts as the sensor. Extrinsic sensors use fibers only for transporting light to and from the actual sensor.

Many fiber-optic sensors are based on single fibers, but others are made with fiber bundles. For example, there are extrinsic sensors where some illumination light is sent to a sample through some of the fibers of a bundle, and reflected light or into used fluorescence light is sent back through other fibers.

Compared with other types of sensors, fiber-optic sensors exhibit a number of advantages:

• They consist of electrically insulating materials (no electric cables are required), which makes possible their use e.g. in high-voltage environments.
• They can be safely used in explosive environments, because there is no risk of electrical sparks, even in the case of defects.
• They are immune to electromagnetic interference (EMI), even to nearby lightning strikes, and do not themselves electrically disturb other devices.
• Their materials (e.g. fused silica) can be chemically passive, i.e., do not contaminate their surroundings and are not subject to corrosion.
• They have a very wide operating temperature range (much wider than is possible for many electronic devices).
• They have multiplexing capabilities: multiple sensors in a single fiber line can be interrogated with a single optical source (see below).

Bragg Grating Sensors

Many fiber-optic sensors are based on fiber Bragg gratings. The basic operation principle is often that the Bragg wavelength (i.e., the wavelength of maximum reflectance) of a fiber Bragg grating depends not only on the Bragg grating period but also on temperature and mechanical strain.

For optical strain sensors based on silica fibers, the fractional response of the Bragg wavelength to strain is roughly 20% smaller than the strain itself, since the direct effect of strain is to some extent reduced by a decrease in refractive index. The temperature effect is close to that expected from thermal expansion alone. The effects of strain and temperature can be distinguished with various techniques (e.g. by using reference gratings which are subject to the same temperature but not to the strain, or by combining different types of fiber gratings), so that strain and temperature are obtained at the same time.

For pure strain sensing, the resolution can be the range of a few με (i.e., relative length changes of a few times 10⁻⁶), and the accuracy may be of the order. For dynamic measurements (e.g. of acoustic phenomena), sensitivities better than 1 nε in a 1-Hz bandwidth are achievable.

There are also Bragg grating laser sensors, where small fiber lasers are realized, consisting of two gratings and a rare-earth doped fiber in between. Alternatively, there can be one FBG and a broadband reflector on the other side. When supplied with pump light, such a device produces an output with a wavelength close to the Bragg wavelength. That emission wavelength can then be measured, and notably it can hardly be influenced even during propagation in a rather long fiber – in contrast to amplitude-coded signals, which may be affected by attenuation.

Quasi-distributed Sensing

A single fiber may contain many grating sensors (see above) in series to monitor the temperature and strain distribution along the whole fiber. This is called quasi-distributed sensing. There are different techniques to address the single gratings (and thus certain locations along the fiber):

• In one technique, called wavelength division multiplexing (WDM) or optical frequency-domain reflectometry (OFDR), the gratings have slightly different Bragg wavelengths. A wavelength-tunable laser in the interrogator unit can be tuned to the wavelength belonging to a particular grating, and the wavelength of maximum reflectance indicates the influences of strain or temperature, for example. Alternatively, a broadband light source (e.g. a superluminescent source) may be used together with a wavelength-swept photodetector (e.g. based on a fiber Fabry–Pérot) or a CCD-based spectrometer. In any case, the maximum number of gratings is typically between 10 and 50, limited by the tuning range or bandwidth of the light source and the required wavelength interval per fiber grating.
• Another technique, called time division multiplexing (TDM), uses identical weakly reflecting gratings, interrogated with short light pulses. The reflections from different gratings are then distinguished via their arrival times. Time division multiplexing is often combined with wavelength division multiplexing in order to multiply the number of different channels to hundreds or even thousands.
• An optical switch allows one to select between different fiber lines, further multiplying the possible number of sensors.

Distributed Sensing

Other fiber-optic sensors do not use fiber Bragg gratings as sensors, but rather the fiber itself. The principle of sensing can then be based on Rayleigh scattering, Raman scattering or Brillouin scattering. For example, optical time domain reflectometry is a method where weak reflections can be localized using a pulsed probe signal. It is also possible, e.g., to exploit the temperature or strain dependence of the Brillouin frequency shift.

In some cases, the measured quantity is a kind of average over the full fiber length. This is the case for certain temperature sensors but also for Sagnac interferometers used as gyroscopes. In other cases, position-dependent quantities (e.g. temperatures or strains) are measured. This is called distributed sensing. A single fiber may then e.g. replace a large number of electronic temperature sensors.

See the articles on optical temperature sensors and optical strain sensors for more details.

Other Approaches

Apart from the approaches described above, there are many alternative techniques. Some examples are:

• Fiber Bragg gratings may be used in interferometric fiber sensors, where they merely serve as reflectors, and the measured phase shift results from fiber spans between them. Pairs of fiber Bragg gratings are used as fiber Fabry–Pérot interferometers, which can react particularly sensitively to external influences. (A Fabry–Pérot interferometer can also be made with other means, e.g. with a variable air gap in the fiber.) Such techniques can also be implemented with multiple fiber segments; the method (usable for dynamic sensing) is then called fiber segment interferometry [12]. Using a sinusoidal optical frequency modulation, which can be easily achieved with current modulation of a single-frequency laser diode, in combination with suitable signal processing (e.g. using a field programmable gate array (FPGA)), one can separately measure the average strain in any particular fiber segment with high temporal resolution [11].
• There are Bragg grating laser sensors, where a sensor grating forms the end mirror of a fiber laser resonator, containing e.g. some erbium-doped fiber, which receives 980-nm pump light via the fiber line. The Bragg wavelength, which depends on e.g. temperature or strain, determines the lasing wavelength. This approach, which has many further variations, promises very high resolution due to the small linewidth of such a fiber laser, and very high sensitivity.
• Long-period fiber gratings are particularly interesting for multi-parameter sensing (e.g. of temperature and strain), and alternatively for strain sensing with very low sensitivity to temperature changes.

Applications

Even after a substantial number of years of development, fiber-optic sensors have still not become very wide-spread, since it is difficult to replace already well-established technologies, even if they exhibit certain limitations. For some application areas, however, fiber-optic sensors are increasingly recognized as a technology with very interesting possibilities. This holds particularly for harsh environments, such as sensing in high-voltage and high-power machinery, or in microwave ovens. Bragg grating sensors can also be used to monitor the conditions e.g. within the wings of airplanes, in wind turbines, bridges, large dams, oil wells and pipelines. Buildings with integrated fiber-optic sensors are sometimes called “smart structures”; they allow one to monitor the inside conditions and to gain important information on the strain to which different parts of the structure are subject, on aging phenomena, vibrations, etc. Smart structures are a main driver for the further development of fiber-optic sensors.

Bibliography