Research On Structural Optimization And Imaging Algorithms Of Capacitance Tomography Sensors

In the process of industrial pipeline transportation,the medium inside the pipeline is mostly a mixture,and the mixture usually contains two or more kinds of substances.Due to the complex dynamic characteristics,large randomness and difficult to describe in the flow process,the detection the multiphase flow parameters are more difficult.As a part of process tomography,Electrical Capacitance Tomography(ECT)has been favored by many scientific research institutions and scholars,and has gradually become a research hotspot in the development of visual detection technology.ECT technology not only has broad development prospects,but also has rapid imaging,simple structure,and non-invasive characteristics.It is widely used in many fields such as pneumatic transportation,flow pattern identification and visual monitoring.Many fields.As a real-time intelligent detection system,capacitance tomography technology can visually reconstruct the interior of the container by detecting the dielectric constant inside the pipeline and the dielectric constant distribution that causes the change of the capacitance value.When reconstructing the pipeline image,an image reconstruction algorithm needs to be used.Transform data into visualizations.Due to the ill-conditioned characteristics of the ECT problem itself,the visualize measurements will inevitably affect the measurement results.In order to improve the measurement accuracy,this paper starts with the ECT sensor structure and imaging algorithm,and introduces new evaluation indicators to optimize the sensor structure,to improve the image reconstruction algorithm.The main research contents of the paper are as follows:Due to the uneven distribution of the sensitive field inside the ECT sensor,the sensitivity is higher near the electrode plate,and the sensitivity is lower when it is far away from the electrode plate and the center of the area.The structure of ECT sensor was optimized by orthogonal experiments to obtain a relatively uniform distribution of sensitive fields.The optimized parameters of ECT sensor mainly include electrode plate opening angle,shield height,pipe wall thickness and relative permittivity.The relationship between ECT sensor structure and performance indicators is complex,and there are contradictions in evaluating performance.When the number of structural parameters is large,this analytical relationship is even more uncertain.Therefore,a fuzzy satisfaction function is defined to evaluate the relationship between parameters and performance metrics.Define a fuzzy satisfaction function to evaluate the relationship between parameters and performance metrics.Based on the relationship between the two and the fuzzy comprehensive satisfaction index FCSI,the structural parameters of the ECT sensor are optimized.Perform a range analysis of the optimization experimental results.The simulation experiment shows that the fuzzy optimization design based on the multi-index orthogonal experiment makes the sensitive field distribution of the capacitive sensor more uniform.The image error of the structure parameters reduced by 7.6% and 5.3% respectively,and the uniformity objective function was improved compared with the first two evaluation indicators.Aiming at the morbidity of the ECT problem,this paper proposes an improved truncated singular value decomposition(TSVD)regularization algorithm based on implicit iteration,which not only overcomes the influence of a few singular values on the final result,but also increases the fidelity of the final solution and has a faster convergence rate,the solution of the ill-conditioned equation has greater stability.In this paper,five two-phase flow simulation flow patterns are selected to carry out image reconstruction and comparison experiments.The results show that the algorithm proposed in this paper is superior to other imaging algorithms,and can more objectively reflect the distribution of the medium in the pipeline.By comparing the relative image error,correlation coefficient and other parameters,the image error of the improved TSVD regularization algorithm based on implicit iteration is reduced and the image correlation coefficient is significantly improved.