Key Technology Of Shape Sensing Of Thin Diameter Rod Using Fiber Bragg Grating

Fiber Bragg grating shape sensoring rod have great potential in shape reconstruction and positioning applications in the field of noninvasive and minimally invasive surgery,such as the positioning of spinal puncture needle,reconstruction of intestinal endoscope,shape display of cardiac vascular catheter,and positioning of tip position of laryngeal surgery robot,etc.the shape reconstruction,the accuracy of the terminal positioning and the technology of diameter refinement have been the key technical problems limiting the clinical application.In this paper,the fiber Bragg grating(FBG)flexible sensor structure is taken as the research object.On the basis of deep research on the theory and method of shape sensing and the realization of technology exploration,the experimental platform and structure model are built to carry out the relevant shape reconstruction experiment verification.The key technologies of the research include: the design and preparation of the thin-diameter long sensor with substrate,the shape reconstruction algorithm based on orthogonal curvature fitting,the FBG sensor network configuration and non-uniform grid point optimization layout method,the bending torsion separation method and shape reconstruction technology,the packaging and shape reconstruction technology of the thin-diameter sensor without substrate,and the strain transfer rate analysis technology without substrate.The main work of this paper is as follows:(1)Aiming at the problem of packaging angle deviation of sensor with substrate,this work proposes to improve the positioning angle of circumferential optical fiber by improving the packaging fixture of sensor,increasing the accurate positioning technology of multiple slender optical fibers(diameter 200 ?m,length 810mm)on the substrate of thin nickel titanium alloy wire(diameter 700 ?m),and increasing the nesting technology of FBG sensor point positioning.A sensor with a length of 810 mm and diameter of 2 mm is encapsulated and reconstructed in plane and space.The reconstruction error is 2.83% in plane and 3.49% in space.(2)In the process of shape reconstruction,based on the shortcomings of the error accumulation of the original reconstruction algorithm,a new shape reconstruction algorithm is proposed to improve the reconstruction accuracy.The original recursive algorithm is transformed into a multi-point line fitting reconstruction algorithm,and a new calculation formula of the fitting algorithm is derived and verified by experiments.It is found that the maximum shape errors of the end positions are 2.65%,2.7% and 3.4% in the single bending,double bending and spatial bending experiments,respectively.The maximum shape error of the end position is2.53%,2.60% and 3.32% respectively by using the unfixed point fitting method.(3)To improve the reconstruction accuracy of the algorithm and the validity of the data collected by FBG sensor network,the optimal layout of sensor grid points is studied.Firstly,based on the characteristics of the error accumulation of the reconstruction algorithm and the influence of the starting position on the end position,the optimization algorithm model is constructed;secondly,based on the operation characteristics of the orthogonal curve network,the optimization objective is analyzed and the objective function is constructed;then,the particle swarm optimization algorithm is used to optimize the layout scheme of the reconstruction algorithm,and compared with the uniform layout scheme.The result of shape reconstruction experiment shows that the error of the sensor's end position is 2.94%,but the end position error of non-uniform grid point is 2.71%.(4)To reconstruct the spatial curve with torsion angle,a theoretical method is proposed to separate the bending and torsion of the shape sensor during the deformation of the sensing rod by FBG sensor array.According to the distribution of triangular FBG sensor array,the relationship between bending curvature and bending strain is derived,and the function of packaging angle of FBG sensor and torsion induced strain under different torsion angles is established.Combined with the influence of bending and torsion,a nonlinear matrix equation is established to solve the three unknown parameters of wavelength displacement caused by maximum strain,bending direction,torsion and temperature.The analysis results show that these three parameters are enough to separate bending and torsion,and obtain two scalar functions of curvature and torsion,which can describe the three-dimensional shape of the curve according to Frenet Serret formula.(5)To reduce the diameter of the sensor,the non substrate packaging method is adopted by improving the positioning sheet,glue injection method and packaging process.The sensor has a diameter of 1.5mm and a sensing length of 200 mm.Four fiber Bragg gratings form 90° each other around the center.There are three gratings on each fiber.Six measuring points are obtained by using the combination technology of space division multiplexing and wavelength division multiplexing.The shape error of the sensor in the plane is 2.71%,and the maximum normalization error of the shape sensor in the space is 3.42%.(6)The strain transfer law of FBG shape sensor without substrate is modeled and calculated by finite element method.A simplified strain transfer model was established,and the mechanical analysis was carried out.Finally,the formula for calculating the average strain transfer rate was deduced.The effect trend of the main parameters of the adhesive layer length,package diameter and adhesive elastic modulus on the average strain transfer rate of the sensor was obtained.It is found that the average strain transmissivity of FBG shape sensor increases with the increase of package length and adhesive elastic modulus,but decreases with the increase of package diameter.