Visualizing the signal of the metal detector and increasing the efficiency of the search

Abstract

In this thesis, the design and analysis of multi-frequency metal detectors that allow the classification of metallic objects is explained. The working method of the designed detector is similar to the continuous wave metal detector, but it works in the range of 1 kHz to 100 kHz instead of a single frequency. As the amount of distinctive information from the target object increases, classification becomes easier. In the theoretical analysis part of the thesis, the working principles of metal detectors are explained. In addition, the working methods of pulse induction metal detectors and continuous wave metal detectors are explained in detail. A basic induction circuit model has been created to perform electromagnetic analysis in frequency space. Since it works at low frequencies, the equations are made easier and their solutions are made by using the quasi-static approach. In the hardware design part, the appropriate inductance value of the coils in the 1 kHz - 100 kHz band range is calculated. The induction reset of the double-D coil, which is suitable for the design, is done. Afterwards, it is isolated from capacitive effects and external EM noises with metallic fabric. The weak signals received with the phase inverting differential amplifier are amplified. In the data analysis part, a remote computer-controlled test setup was set up in the laboratory environment. Signal generator and oscilloscope were controlled via computer with LabVIEW® program. The phase and amplitude data of the signal received with the LabVIEW® interface are recorded in the file. With the MATLAB® program, coherent and quadrature signal segments were created from the phase and amplitude data. The graphs were drawn according to the phase, amplitude, coherent and quadrature data of the signal taken from the test objects, and their analyzes were made.