Research On Model For Metal Plate’s Surface Temperature Field Based On Distributed Fiber Bragg Grating

In the mechanical system,the temperature rise is likely to bring up the thermal deformation of the original machinery,leading damage of the structure.Therefore,it is of great vital to gain temperature distribution during machinery health monitoring and managing,contributing to exact production,development of structures’ life-time and loss reduces.It has been proved that there are many advantages of Fiber Bragging Grating(FBG)at temperature measurement such as strong electromagnetic corrosion resistance,heat-resistance,small attenuation of signals etc.By multiplexing technique of Fiber Grating Sensors,multi-point multi-parameter measurement is realized,which is pretty conducive to the establishment of temperature field’s distributed measurement network.Through distributed measurement network,multi-point temperature data are acquired at the same time and proper modeling method is designed based on the multi-point data,which is a relationship between temperature and space or time,leading to a more comprehensive understanding of the temperature information on the various positions and short-term temperature forecast.In this paper,the steady state temperature field model and the non-steady state prediction model are studied for metal plate’s surface and a distributed measurement system based on FBGs is established.The main work is as follows:(1)In view of the numerical modeling method of the steady state temperature field,the Bspline model is optimized by the particle swarm optimization(PSO)algorithm to reduce the error,named PSO-Bspline.Respectively,for the single peak and double peak functions,the proposed modeling method is adopted compared to the three spline model and Gaussian radial basis functions.The finite element analysis software ANSYS is used to simulate the steady temperature field of the metal plate,and the dispersion points’ temperature is applied for modeling,proving that PSO-Bspline shows the smaller error and the better correlation.(2)In view of the numerical modeling method of the non-steady state temperature field,gray model(GM)and autoregressive model(AR(p))are combined according to their respective advantages and gray autoregressive model(GM-AR)are proposed.GM-AR model is adopted for supposed functions.Meanwhile,ANSYS is used to simulate the non-steady temperature field of the metal plate and the dispersion points’ temperature is applied for modeling,proving that proposed GM-AR model is more likely to have better forecast results.(3)Studying on the principle of distributed optical fiber grating temperature sensing and multiplexing technology,distributed temperature measurement system is established based on distributed FBG sensors in this paper.Specifications and implement monitoring processes are in deep design.The data measured by infrared camera are gotten to prove the accuracy of measuring system.(4)According to the distributed temperature measurement system,distributed multi-point temperature data are achieved to build the steady state temperature field model and non-steady state temperature field model.The performances of different modeling methods are compared.