Theoretical Modeling And Implementation Of Micro Optical Fiber Sensors

Due to the advantages of immunity to electromagnetic interference, resistance to harsh environment, remote measurement, and small size, an optical fiber sensor outstands from others and becomes the main direction of advanced sensing technologies. As important one of optical fiber sensors, the intensity reflecting optical fiber sensor has more unique advantages of compact structure, low cost, good reliability and flexibility.In respect of non-contact high-precision micro-displacement measurement and hydrogen gas detection, several key problems of micro optical fiber sensors are studied systematically in this dissertation, which include theoretical modeling, intensity compensation, weak signal processing, as well as developing and testing of prototypes.Firstly, according to different sensing mechanisms in displacement and hydrogen measurement, the theoretical models of reflective optical fiber displacement sensor and micro-mirror optical fiber hydrogen sensor are built, respectively. Their theoretical characteristics are simulated, the influence rules of related parameters on which are analyzed and the validity of the built models are verified by experiment.After analyzing various factors influencing on performance of optical fiber sensors and typical intensity compensation methods, in order to improve their precision, stability, linear range and probe size, a combined optical fiber bundle displacement sensor probe and an optical fiber coupler hydrogen sensor probe are proposed. Their optimizing design methods based on theoretical models are presented. To illustrate compensation performance of the proposed probes, experiments are carried out.An intensity compensation algorithm is studied to increase the precision and linear range of optical fiber sensors further. Applying neural network to intensity compensation of optical fiber sensors, a method based on radial basis function network is proposed. Experiments using the proposed method show that intensity compensation and nonlinearity correction of optical fiber sensors are fulfilled simultaneously.Weak photoelectric signal processing is an important part of optical fiber sensor. Theories about correlation detection and lock-in amplifier are studied. An analog signal processing system and an embedded digital signal processing system for optical fiber sensors are developed, which modulate and drive the light source stably and realtime demodulate the receiving signals. The digital signal processing system has a replaceable optical fiber probe interface and a network interface, and some intelligent functions as self-diagnostics and self-correction.At last, based on the research above, the prototypes of an optical fiber displacement sensor and an optical fiber hydrogen sensor are developed. Corresponding fabricating technics are presented. Testing results show that they satisfied design requirements.