Research On Methods Of High-precision Dynamic Measurement Of Light Field Linear Type Time Grating Displacement Sensor

Precision measurement technology guarantee a national manufacture industry goes to high grade,precision and advanced.with the development of science and technology,automation equipments rely on precision measurement,precision dynamic measurement hold important position in modern science and technology area.Time grating is a new displacement sensor which have its own intellectual,and light field type linear time grating is a branch research direction of time grating.At present,the precision of light field type linear time grating sensor can reach to ±2?m when run in static experiment conditon and can reach to ±8?m precision when run in dynamic condition,however,when the velosity of experiment conditon greater than 5mm/s,the precision decreased obviously,existing time-space out of synchronism phenomenon,which lead to generate dynamic errors and result in precision decreasing.The thesis got the support of the Natural Science Foundation of China,for precision falling problems,I carried out studying on methods of light field type linear time grating which with velosity greater than 5mm/s dynamic measurement condition to achieve high accuracy and high resolution.Research target:Studying two new methods in light field linear type time grating area,to make sensor's dynamic error within ±4?m during dynamic experiment with speed greater than 5mm/s.Firstly,coming up with two methods to improve time grating dynamic performance–BP neural network prediction arithmatic,Serial dynamic phase comparation method,Meanwhile,building mathematic models for two methods and analysing two mathematic models feasibility based on time-garting theory.Secondly,hardware circuit schematic diagrams and pcbs were designed,BP neural network prediction arithmatic hardware circuit with FPGA as core,mainly consist of Excitation signal module,A/D sampling module,Traveling wave synthesis module,and UART module etc.For serial dynamic phase comparation method hardware circuit mainly include A/D sampling module,dule-direction phase comparator circuit,D/A conversion module etc.,with FPGA as MCU to do logic operation.Finally,expriment proof was built to do dynamic calibration and verify two methods,when accelerated velosity close to 0cm ?,the velosity of experiment condition at 9mm/s,BP neural network prediction arithmatic method's error within ±4?m,Serial dynamic phase comparation method's error within ±3?m.when accelerated velosity reach to 4~5mm ?,BP neural network prediction arithmatic method's error within ±9?m,Serial dynamic phase comparation method's error within ±6?m.The experimental result demonstrate: Both of two methods fulfilled reaearch target,but under high acceleration velocity condition,BP neural network prediction arithmatic and Serial dynamic phase comparation method's precision are decreasing.