The dissertation is devoted to the solution of the scientific and technical task of improving the electric capacitance method of non-destructive testing of materials through the development of systems structures for increasing the sensor sensitivity and speed of the device. It is also devoted for studying the parameters and operating modes of the sensor to ensure the maximum sensitivity of the measuring scheme, the influence of the geometric shape and dimensions of the sensor on its sensitivity.
In this work it was make an overview of the problem of electrical non-destructive testing in the manufacture of nanomodified intelligent polymer composite materials, taking into account the need for point measurement of dielectric material permeability and control of the depth of intelligent sensors introduction in the melt of polymer material in real time with the possibility of only one-way access to the object of control.
The electric capacitance method is one of the most versatile and easy to use. Be this method it is possible to solve a significant range of research tasks. However, by this time, the electric capacitance method was used only for the integral evaluation of the parameters of the object of control. Using the method for point control is limited by a significant control time, which is due to the use of Wheatstone bridge schemes for measuring small capacitance values with high accuracy. This is explained by the fact that such schemes require multi-step algorithms, which leads to a significant increase in the measurement time. When scanning object of control in many points the time consuming becomes unacceptably large. Besides, for this method there are no analytical methods for finding capacitance of the asymmetric sensors and sensors of complex form requiring the use of computer methods for calculating capacitance. Also, the issue of combating electromagnetic interference is still unresolved.
The theoretical substantiation of the electric capacitance method and object of control is carried out. The method of measurement and criteria for estimating the sensor operation is proposed. The recommendations for determining an optimal parameters of the work and minimum possible dimensions of the sensor for the implementation of the defectoscopy are obtained.
The optimal frequencies are determined for the application of the method and a method for determining these frequencies when changing the parameters of the scheme is proposed. With the help of simulation, the dependence of the sensor capacity on the characteristic size of the electrode was determined, which allowed setting the minimum possible value of the electrode, which is may be measuring the change in capacitance, that was 6x6 mm.
It was investigated the parameters of control process and the limits of method application for conducting and non-conducting samples, in particular: thickness of the control object, working gap, relative permittivity of the sample.
Simulation of sensor parameters, namely the distribution of sensitivity, allowed to compare sensors of different geometric shapes and to determine the optimal design parameters for the sensitivity criterion. The design of the capacitive sensor is proposed, which ensures achievement of the highest sensitivity. The proposed sensor has a 50% greater control depth and 63% greater sensitivity than with standard sensors. Thus, the proposed sensor combines the advantages of previously studied sensors and is best for use in nondestructive testing.
The method of increasing noise immunity is developed, which allows to register with high accuracy the change not only of the phase shift of the measuring signal, but also the change in the amplitude of the measuring signal, which provides an increase in the accuracy of the measurement. On the developed method were obtained the Ukrainian patents for the invention.
Experimental studies have been carried out, which confirmed the efficiency of the method for solving defectoscopy problems. The measurement error of the amplitude and phase of the signal, which is acceptable for control, is calculated.
The hardware error has been determined, and the simulation in the MATLAB software complex has been carried out in order to determine the random error when measuring the developed scheme.
