Research On Structural Design And Preparation Process Optimization Of Thin Film Strain Sensor

The cutting force produced in the cutting process directly affects the cutting heat and the machining quality of the workpiece surface.With the vigorous development of MEMS technology,it is a very important scientific problem to embed the thin film strain sensor into the specific tool to realize the real-time measurement of cutting force.In the process of developing thin film strain sensor,its structural design,preparation process and parameter optimization are the three most important links,which affect the force measurement performance of thin film strain sensor.In this paper,the structure of thin film strain sensor is designed based on the strain and deflection theory of material mechanics.The preparation process of transition layer,insulating layer and sensitive layer is studied.The process parameters of sensitive layer are optimized by least squares support vector machine(LSSVM)and genetic algorithm(GA).Firstly,the basic principle of structural design and simulation analysis of thin film strain sensor is described.It is proposed that the size design of elastic substrate and resistance grid are the two key points of sensor structural design.For the size design of elastic base,eight kinds of elastic base structures are designed,the theoretical model is established according to the strain theory and deflection theory in material mechanics,and the strain and deflection distribution of eight kinds of elastic base are analyzed by ABAQUS software.In order to select the better elastic base structure,the evaluation model of elastic base is established.It is concluded that Hshaped base and symmetrical circular arc elastic base are the better structures.Then the strain transfer efficiency of the resistance grid is analyzed,and the effects of the resistance grid structure under different thickness and width on the strain distribution and sensor sensitivity are studied.With a resistance of 800 ?,the resistance grid structure is designed through the resistance formula.Then,the preparation process principle of thin-film strain sensor is studied,the preparation process of thin-film strain sensor is systematically explained,the preparation process parameters of the composite insulating layer composed of transition layer and insulating layer are determined,the characterization performance of the composite insulating layer is analyzed,and the electrical performance of the sensitive layer is analyzed,which lays a foundation for exploring the influence of the preparation process of sensitive layer on its characterization performance.Finally,the preparation process optimization of the sensitive layer film of thin film strain sensor is studied.LSSVM is proposed to establish the nonlinear regression function between the preparation process parameters of Ni Cr alloy thin sensitive layer and its characterization performance,and GA is used to find the optimal value of the nonlinear regression function of each characterization performance.The experiments between sputtering power,sputtering pressure,argon flow rate,substrate bias and film deposition rate and roughness were designed by orthogonal test method.The sample data were obtained,and the effects of process parameters on deposition rate and roughness were analyzed.Using the established optimization model,the value of film deposition rate under the optimal process parameters was 76.92nm/ min;The value of surface roughness under the best process parameters is 10 nm.The test of film resistivity is designed by orthogonal test method,and the sample data are obtained.The influence laws of sputtering power,sputtering pressure,substrate bias and substrate temperature are analyzed.The prediction model of resistivity is established by LSSVM,and the resistivity under the best process parameters is 1.34 ?·?m by GA.