Sensor Design And Image Reconstruction Of Electrical Capacitance Tomography System

Electrical capacitance tomography(ECT)is a kind of computer tomography technology.In this technology,the alternating excitation voltage is applied to the closed space,and the conductive medium in the closed space generates induced charge.The electrodes evenly distributed outside the closed space collect capacitance data,and the capacitance data and sensitivity matrix are used to reconstruct the material distribution of the conductive medium in the closed space.There are some advantages such as non radiation,non invasion,high temperature resistance,low cost and easy access to process parameters.It is an ideal nondestructive testing technology and is widely used in the fields of multiphase flow detection and combustion visualization detection.The performance of ECT system depends on the accuracy and speed of sensor acquisition and image reconstruction,and based on the12 electrode ECT system,the key issues of sensor design and image reconstruction are studied in this paper.The principle and composition of ECT system are discussed in detail.The mathematical model of ECT is deduced,and the structure of sensor is analyzed.The relationship between measured capacitance and sensitivity matrix is established by finite element analysis method.The solution model of image reconstruction is deduced.The theoretical basis for sensor design and image reconstruction of ECT system is established.For the influence of sensor structure parameters on the accuracy of the capacitance data acquisition,the mathematical model of sensor sensitive field is established.When the sensor excitation mode,the plate width,the plate angle,the radial shielding electrode and the insulation filling material change,the sensor performance,electrostatic field distribution and "soft field" effect are analyzed.A new sensor is designed according to the optimal solution of physical parameters.The image reconstruction shows that the optimized sensor has higher performance.For the influence of sensor on the accuracy of ECT system,a sensor based on the photosensitive trigger is designed.At the same time,the optical window structure of transmitting optical signal and shielding electromagnetic wave is adopted to protect electromagnetic shielding and photosensitive components.The data acquisition module of ECT system is analyzed,and for the interference of stray capacitance to the sensor performance,the intra-isoelectric driver cable shielding technology is introduced.The double T-type simulation switch matrix and the digital-analogy demodulation technology are adopted.The experimental results show that the ECT system based on the photosensitive sensor improves the signal-to-noise ratio of the image and the sensitivity of the capacitor.To improve the accuracy of image reconstruction,a fast ECT image reconstruction algorithm based on super-resolution image feature extraction is proposed.Super resolution image is obtained by super-resolution image reconstruction model,and super-resolution image features are extracted based on chamfer distance function.The image edge features are calculated,the center of gravity is used as the pole to polarize,the local limit points of the edge on the amplitude diameter curve are obtained,and the features of the super-resolution image are used to reconstruct the image.Experimental results show that the algorithm can quickly extract superresolution image features and effectively solve the edge effect in the process of image reconstruction.For the low overall accuracy of flow pattern identification,the partial least squares method is used to train the classification model,which based on four typical flow patterns of laminar flow,single drop flow,core flow and circulating flow,and the model is used to identify the flow.Compared with other classification algorithms such as quadratic discriminant analysis and linear discriminant analysis,the partial least squares method combined with linear discriminant analysis has higher overall classification accuracy.