Research On Dynamic Error Compensation Of Fiber Bragg Grating Strain Sensor

With the continuous development of industrial technology and equipmentbreakthroughs, high precision dynamic measurement for the equipment has becomethe mainstream monitoring, and it puts forward higher requirements on the dynamicperformance of the sensors. Generally, sensors itself with nonlinear structure andaffected by various factors such as the environment in the practical engineeringapplication, so there must be measurement errors in the measurement system.Therefore, the research of dynamic compensation is of great importance for achievingeffective monitoring equipment.FBG sensor is a newly developed optical fiber sensors in recent decades, and hasoutstanding advantages in the measurement of temperature, strain, acceleration andother physical parameters. This paper details FBG strain and temperature sensingprinciples and analysis sensor’s dynamic performance through theoretical analysisand experimental research. According to numerical analysis method, this paperrealizes temperature compensation and dynamic compensation to improve theaccuracy of measurement.This paper contains the following contents:1. Based on the working and sensing principle of FBG, the performance of FBGstrain sensor are analyzed through strain characteristic test and temperaturecharacteristic test.2. By using the ANSYS finite element analysis method, the finite element modelof FBG strain sensor is established. Based on the thermal analysis and thetemperature and strain coupling experiment, the influence of temperature on thesensor’s structure and accuracy are discussed.3. According to the least squares theory and principle of temperaturecompensation, this paper realize temperature compensation effectively. According tothe BP neural algorithm theory and principle of temperature compensation, thestructure and parameters of BP neural network are identified and the BP neuralnetwork simulation training and test are accomplished.4. This paper obtained the working frequency bandwidth and the response time of FBG strain sensor with the analysis of transient response and step response.5. Based on the analysis of step response and according to the BPneural algorithm theory and principle of dynamic compensation, the dynamiccompensation for the FBG strain sensor using BP neural network are accomplished.