Fabrication And Sensing Signal Processing Of Polyimide Optical Fiber F-P Cavity Sensing Probe

With the development of optical fiber sensing technology,optical fiber sensor has been successfully applied in the fields of petroleum chemical industry,metallurgical industry,and aerospace to measure various physical quantities,how to achieve temperature measuring precisely and fast in these fields has been a hot research problem in recent years.For the characteristics of small volume,strong anti-electromagnetic interference ability,strong ability to resist a complex environment and higher interference precision,interferometric optical fiber sensors such as M-Z interferometric optical fiber sensor,Michelson interferometric optical fiber sensor and Fabry-Perot interferometric optical fiber sensor have been hot research objects.Currently,materials for fabricating optical fiber F-P cavity sensor sensing probe are almost quartz glass,single mode quartz optical fiber,pure silica core optical fiber and photonic crystal optical fiber,compared with above materials,polymer material has a higher coefficient of thermal expansion and is more sensitive to temperature,and has better light transmittance property,photosensitive property,lower transmission loss,better corrosion resistance and higher mechanical strength.In this paper,a kind of Polyimide optical fiber F-P cavity temperature sensing probe was designed and fabricated,and its multi-peaks reflected sensing signal interference spectrum wavelength peak seeking algorithm was studied,works were carried out from the following aspects:(1)Through theoretical principle,the influence of F-P cavity surface reflectance,cavity surface tilt angle and length size on reflected signal spectrum were analyzed,and an intrinsic F-P cavity sensing probe structure was designed whose cavity length and cavity depth sizes were 25?m and 55?m respectively,and the reflected signal of F-P cavity has obvious interference spectrum.(2)The influence of cavity length direction feed on cavity length processing size,the influence of cavity depth direction feed on cavity depth processing size and the influence of femtosecond laser power on F-P cavity processing size were explored,by comparing size error between processed size and designed size,the processing parameters were determined: spot diameter size was 4?m,the processing speed was5mm/s,the laser power was 45 m W,cavity length direction feeds was 3?m,and processing times was 10,cavity depth direction feed was 5?m,and processing times was5,total processing times was 50.Polyimide optical fiber and single mode quartz optical fiber F-P cavity sensing probe structure were fabricated by femtosecond laser whose cavity length and cavity depth were 25.66?m and 55.8?m respectively,the signal spectrum and temperature sensitivity of the two sensing probes were compared and analyzed,the experimental results show that Polyimide optical fiber F-P cavity's interference effect is more obvious,and reflected sensing signal interference spectrum has higher quality;In the temperature environment of 20??80?,the temperature sensitivity of single mode quartz optical fiber F-P cavity sensing probe is 2.56pm/?,the temperature sensitivity of Polyimide optic fiber F-P cavity sensing probe is4.42pm/?,so Polyimide optical fiber F-P cavity sensing probe has higher temperature sensitivity and better temperature sensing performance.(3)The characteristic differences between FBG reflection sensing signal spectrum and F-P cavity reflected sensing signal spectrum were compared and analyzed,for the problem that traditional peak seeking algorithms of FBG signal spectrum cannot be applied to multi-peaks signal spectrum peak seeking,according to the characteristic of F-P cavity reflected sensing signal interference spectrum,a multi-peaks wavelength peak seeking algorithm was proposed,the simulation test average peak seeking error of multi-peaks algorithm is 3.2pm;in 30.7 ? constant temperature experimental test condition,multi-peaks algorithm average peak seeking error is 3.7pm,direct peak seeking algorithm average error is 12.1pm,Monte Carlo algorithm average error is5.2pm,Gauss fitting algorithm average error is 4.1pm;in 20? ? 80? variable temperature experimental test conditions,the multi-peaks algorithm average peak seeking error is 4.3pm,direct peak seeking algorithm average error is 13.2pm,Monte Carlo algorithm average error is 6.5pm,Gauss fitting algorithm average error is 4.7pm,compared with traditional peak seeking algorithms,in the same test condition,multi-peaks seeking algorithm all has lower peak seeking error.