Hardware Of Sparse Spectral Sampling For Optical Fiber Fabry-Perot Sensor

Among various kinds of optical fiber sensors,Fabry-Perot sensor has the advantages of high reliability,good stability,and strong ability to resistance harsh environments.It is particularly suitable for working in all kinds of harsh working environments,and is an important means for monitoring large-scale equipment structures.However,due to the limitation of the demodulation principle,the measurement speed of Fabry-Perot sensor is not high,and it is difficult to meet the high-speed monitoring requirements for aero engines,nuclear power plants and other special occasions.Therefore,breaking the bottleneck of its measurement speed has potential application value for high-speed monitoring occasions in various harsh environments.To increase its measurement speed under the condition of ensuring measurement accuracy,it is necessary to start from both of the hardware sampling and the software processing.Combining the sparse spectral sampling with the maximum likelihood estimation is a promising way to break through the demodulate speed of optical fiber Fabry-Perot sensor.For this reason,this project focuses on the hardware of sparse spectral sampling based on the maximum likelihood estimation,and the following work are completed:(1)Compare the current demodulation algorithm and maximum likelihood estimation demodulation algorithm in detail.Analyze the requirements of the maximum likelihood estimation demodulation algorithm under sparse spectrum sampling,and put forward the hardware system overall design scheme.(2)According to the overall design scheme of the hardware system,the parameter requirements of each module in the system are analyzed,and various types of devices are selected according to various device performance parameters in the market.The system uses ASE broadband light source with adjustable output power and a flat spectrum.A wavelength division multiplexer is used to filter the wide-band spectrum to generate 40 single-wavelength optical signals to achieve sparse sampling.(3)Design photoelectric conversion and amplification circuits.Analyze the loss of optical path,and put forward the requirements of circuit design,such as amplification,SNR,and bandwidth.Based on the equivalent noise model of the circuit,various noises in the circuit are analyzed in detail and targeted at multiple channels.The analysis of circuit gain consistency and phase consistency has been done.The circuit is debugged and the circuit is tested for performance.The circuit performance meets the design requirements.(4)Design and build an experimental system based on sparse spectral sampling to correct the loss of each path of the hardware system;verify the feasibility of the designed hardware system.The experimental results show that the sparse spectral sampling can restore the original spectrum very well,and the system can achieve a wide range of measurements.When the cavity length is in the range of 250?m-400?m,the demodulation error is less than 1.152%.The feasibility of the sparse spectral sampling hardware system of the optical fiber Fabry-Perot sensor is preliminarily verified.