Research On Quantitative Detection Technology Of Coplanar Double-electrode Sensor Scanning

Internal defects in composite materials can seriously harm their performance.The quantitative detection of defects in advance has a significant impact on improving the quality of equipment using this material.The detection technology of coplanar capacitance is suitable for composite material defect detection due to its advantages of non-invasive,single-sided detection,and low cost.The current coplanar array capacitance imaging detection technology is ill-posed due to the uncertainty of the inverse problem,which affects the detection accuracy.It is of great significance to research the solution of the inverse problem,which is used to promote the practicability of the detection technology of coplanar capacitance.However,based on the coplanar double-electrode scanning detection mode,establishing the direct correlation law between the capacitance data and the medium distribution is an effective way to circumvent the inverse problem solution.Because of the nonlinearity of the edge electric field of coplanar electrodes and the uncertainty of the capacitance versus media distribution relationship,it is challenging to directly establish a detection model based on the characteristics of the capacitance data.Therefore,a quantitative detection model based on the scanning of coplanar doubleelectrode sensor and the study of data processing algorithms were developed in this paper.The primary research is as follows:Firstly,based on the analysis of the edge electric field and the mathematical description of capacitance,the electric field of coplanar capacitance in static detection and scanning detection was analyzed.A coplanar double-electrode sensor scanning detection mode was proposed.and a model for quantitative detection of the abnormal dielectric distribution in the materials was constructed.This model can qualitatively identify internal anomalies in materials and quantitatively calculate the size of anomalies by identifying the mutation range of scanning detection capacitance data.Secondly,the influence of electrode parameters and scanning parameters was studied,and the optimal combination conditions for quantitative detection parameters of coplanar dual electrode scanning were obtained.The rectangular coplanar double electrode with intermediate shielding is preferred.The detection conditions with a scanning interval of1 mm and a scanning step of 0.451 mm were determined.Scanning quantitative detection of internal anomalies in different specifications was studied using double-layer composite materials as the research object.The experimental results show that the obtained capacitive data mutation characteristics were consistent with the simulation analysis,and the internal anomaly at this calculation was in the scanning direction dimension(relative error of 1.04%),which verifies the correctness of the quantitative detection model of the established a coplanar double-electrode sensor scanning.Then,accurately identifying the mutation interval of the capacitance change curve is a direct influencing factor on the quantitative detection accuracy.Therefore,a mutation interval automatic recognition algorithm based on probabilistic neural networks was proposed to quantitatively calculate the size of internal anomalies.The probabilistic neural network algorithm for coplanar double-electrode was established through feature extraction,training set selection,and smoothing factor analysis of scanning detection results.The experimental results of double-layer composite material detection showed that the model could automatically identify the length of the mutation interval in the capacitance change curve and quantitatively calculate the internal abnormal size.Finally,the application of the quantitative detection technology of coplanar doubleelectrode scanning was explored,with the adhesive layer defects of multilayer composite material bonding structures as the detection targets.Scanning quantitative detection experiments were conducted on different defect types of experimental samples,considering the adaptability of electrode size and sample size.The experimental results showed that,within the allowable error range,the quantitative detection technology could achieve quantitative detection of the size of adhesive layer defects and determination of the relative position of defects.The effectiveness and applicability of the detection technology were verified.