Research On Data Fusion Technology Of Electrical Detection

With the development of science and technology in human world,higher level detection technology is required by medical and industrial applications.It is a research hotspot and a main direction of soft detection which reconstructs the unmeasurable variables according to the mechanism-empirical composite modeling through the indirect sensing.The two most important application of soft dection,with the background of medical imaging and industrial process control,are imaging and flow detection,which belong to the basic technologys.And the improvements of their performance are significant.Electrical tomography and electromagnetic flow detection are proposed for these backgrounds repectively with high research value.Because of the restriction from underdetermined problem and soft field effect,it is difficult to obtain effective and stable 2D image reconstruction effect by ordinary image reconstruction algorithm.Then the reconstruction of 3D electric tomography image is not effective.Electromagnetic flowmeter technology has been mature in theory.But its application is always be restricted by supply of power,quality of power and harsh EMI enviroment in complex industrial scene,which make the power supply from battery necessary.Thus,the reduction of power consumption,resistance of measurable noise,stablization of measurement quality,and improvement of measurement accuracy are with important practical significance.In view of the above problems,this paper has done the following work: 1)For the two problems of electrical tomography that “underdetermined problem” and “soft field effect” which coused the image reconstruction cannot be improved,this paper through the exploration of data fusion to mining the deep coupling relations of measured data to improve the reconstruction effect.In the view fo the characteristics of available electrical tomography system has 8 different excitation modes,fully utilize all the modes and fusion all useful data,so as to improve the reconstruction effect of the two-dimensional ET image.2)Aiming at the problem of 3D reconstruction of electrical tomography,import the volume rendering algorithm which always used in computed tomography,and through the fast fuzzy C-means clustering algorithm to handle the 2D slice of electrical tomography and give the optimal classification by using clusteringvalidity index,so as to achieve the purpose of faster 3D reconstruction.The experimental results show that the proposed method is effective in 3D reconstruction of electrical tomography,and can further improve the effect of CT 3D reconstruction.3)Low power,portable flowmeter is the development direction of the electromagnetic flowmeter,this paper examines the market existing battery powered electromagnetic flowmeter with low power consumption,to reduce power consumption measurement as the basic requirements,the interval measurement based on the scenario of the traffic detection technology is analyzed and studied.With traffic prediction theory,put forward Calman filtering prediction,Markov forecast model,and used to predict the result of the fusion of fuzzy integral,through comparative analysis of traffic prediction simulation and actual prediction results,verify the accuracy of the improved flow detection.