RResearch and development of algorithm for determination of absolute conductivity value without calibration by eddy current techniques

通过确定材料的载荷极限，无损检测方法对于技术部件的安全可靠运行至关重要。一种广泛使用的程序是导电材料的涡流检测。除了检测材料缺陷外，还可以非破坏性地进行复杂材料的表征。材料的特性由它们对电学特性的影响决定。电磁场在材料中产生涡流，涡流的大小受材料参数(电导率、磁导率、介电常数、几何形状)的影响。这些影响可以通过确定线圈的阻抗来测量。阻抗的这些值取决于频率和提离，例如线圈和待测材料之间的距离。到目前为止，用涡流测量电导率的方法是基于经验方法。用不同电导率的参比样品，测定了一条校准曲线。这种校准只对一个设备、一个频率有效，而且大多数情况下只适用于非常有限的升空范围。同时，也存在这样一个问题，即通常没有足够的精确定义和良好分级的参考样本来确定这些曲线。通过仿真结果，找到了一种新的理论方法。这是使直接测定电导率成为可能的另一种方法。在这种情况下，它直接来自物理定律，而不再来自经验校准曲线。因此，它不限于特定的测量频率。此外，提离值可以直接分配给测量的阻抗值。这种方法使电导率层析成像能够像分层深度剖面法那样进行。对材料内部的场分布进行积分应得到电导值。求出了积分值与电导率之间的关系。这在数学上是非常苛刻的，因为它导致了一个逆问题。传播体积的电特性必须根据表面某一点上的测量值来确定。假定电导率只在深度(z方向、波传播方向)上变化，而在平面上具有相同的值。通过频率改变材料在某一深度改变侵彻深度，对计算的积分有不同的影响。这样，就必须开发出一种确定作为深度函数的电导率的算法。在薄层的情况下，似乎可以同时确定层的厚度和电导率，另外还可以确定衬底的近似电导率。这一新的方法必须进行详细的研究，并为实际使用做好准备。使用改进的硬件不仅测量接收线圈中的电压，而且还测量励磁线圈中的电流，从而能够计算测量装置的阻抗。

Non-destructive testing methods are essential for safe and reliable operation of technical components by determination of load limits of the material. A widely-used procedure is eddy current testing of conductive materials. Besides detection of material defects it is possible to perform complex material characterization non-destructively. Material properties are determined by their influence on electrical properties. The electromagnetic field induces eddy currents in the material whose values are influenced by the material parameters (conductivity, permeability, permittivity, geometry). These influences can be measured by determining the impedance of the coil. These values of the impedance depend on the frequency and the lift off, e.g. the distance between the coil and the material to be tested. Until now the measurement of conductivity by eddy current is based on a empirical approach. Using some reference samples with different conductivity a calibration curve is determined. This calibration is valid only for one device, one frequency and mostly only for a very limited range of lift off. At the same time there is the problem there are often not enough exactly defined and well-graduated reference samples for determining those curves. A new theoretical approach was found by simulation results. It is another way that makes possible direct determining of conductivity. In this case it is derived directly from the physical laws and no longer from the empirical calibration curve. Thereby it is not limited to a specific measurement frequency. Additionally the lift off value can be assigned directly to the measured impedance value. This method enables carrying out a conductivity tomography as sliced depth profiling. Integration over the field distribution inside the material should result the conductivity value. The relationship between the value of the integral and the conductivity has to been found. This is mathematically very demanding because it leads to an inverse problem. The electrical properties of the propagation volume have to be determined from the measured values on a point of the surface. It is assumed that the conductivity changes only in the depth (z direction, direction of wave propagation), whereas it has in plane the same value. Changing the penetration depth by frequency change the material in a certain depth has a different influence on the calculated integral. In this way an algorithm for determining conductivity as a function of depth has to be developed. In case of thin layers it seems to be possible to determine the thickness and the conductivity of the layer at the same time and additionally the approximate conductivity of the substrate. This new approach has to be examined in detail and prepared for the practical use. A modified hardware is used measuring not only the voltage in the receiving coil but additionally the current in the exciting coil making possible the calculation of the impedance of the measurement set.