Research On Displacement Measurement Accuracy Based On Optical Fiber Displacement Sensor

Sensors are an important tool for obtaining information.They are at the forefront of industrial automation production and are widely used in different fields.As a sensor for measuring displacement,the reflective optical fiber displacement sensor studied in this paper has many advantages,such as small probe,simple structure,easy to use,long life,low power consumption,strong anti-electromagnetic interference ability,and so on.However,the measurement accuracy of the sensor is easily affected by some interference factors,among which nonlinear error and ambient temperature are the two most prominent interference factors.In response to these problems,this article has studied the solutions deeply,and developed a measurement system.The core goal of this paper is to improve the measurement accuracy of the sensor.Focusing on this goal,the following work has been specifically done:(1)The working principle of the optical fiber displacement sensor is analyzed from the aspect of geometric optics and composition structure,and the input-output characteristics of the sensor is also analyzed.From the temperature-displacement calibration experiment,the linearity,zero temperature coefficient,sensitivity temperature coefficient and temperature additional error of the optical fiber displacement sensor are calculated.From this calculation,it is known that the sensor has nonlinear error and is easily affected by ambient temperature.In addition,the influence of the selection of materials for each component of the sensor on the improvement of measurement accuracy is also analyzed,and the conditions of each component should meet are given.(2)Aiming at the problem of linearity,this paper uses the look-up table method and the curve fitting method to perform non-linear improvement respectively.Under the condition of the look-up table method,the linearity of the sensor increased from7.86%to 2.50%,which is an increase of 3.14 times.Under the condition of the curve fitting method,the linearity increased from 7.86%to 3.50%,which is an increase of2.25 times.In conclusion,these two methods can improve the linearity of the sensor in different degrees.(3)Aiming at the problem of environmental temperature,this paper respectively established a binary regression analysis model and a particle swarm optimizationOptimization least square support vector machine model for temperature compensation.Under the condition of the binary regression analysis model,the zero temperature coefficient of the sensor is increased from9.78'10^-3/?to 2.72'10^-3/?,the sensitivity temperature coefficient is increased from7.47'10^-3/?to 6.47'10^-3/?,and the temperature additional error is increased from 18.30%to 6.53%.Under the condition of PSO-LSSVM model,the zero temperature coefficient of the sensor is increased from 9.78'10^-3/?to 9.46'10^-4/?,the sensitivity temperature coefficient is increased from 7.47'10^-3/?to 6.72'10^-4/?,and the temperature additional error is increased from 18.30%to 1.65%,this three indexes all get an increase of one order of magnitude.In short,both two models all have achieved good compensation effects.(4)Designed a set of measurement system with STM32F407ZGT6 single-chip microcomputer as the core,solidified the optimized PSO-LSSVM algorithm in it,and used each functional module of the smallest system of the single-chip microcomputer to realize the temperature compensation of the system and realize the intelligent displacement measuring.The research results show that:(1)The effect of using the look-up table method for nonlinear compensation is slightly better than the curve fitting method,the look-up table method make the linearity of the optical fiber displacement sensor increase by 3.14 times;(2)The effect of using the PSO-LSSVM model for temperature compensation is significantly better than the model of the binary regression analysis.Under the condition of the PSO-LSSVM model,the zero temperature coefficient,sensitivity temperature coefficient and temperature additional error of the optical fiber displacement sensor are all increased by an order of magnitude.Practice has fully proved that the look-up table method is effective for improving linearity,and the PSO-LSSVM model is feasible and useful for temperature compensation.