English 
日本語 
한국어 
Deutsch 
Español 
italiano 
русский 
português 
العربية 
tiếng việt 
ไทย 
中文
NEWS
In modern industrial manufacturing, eddy current testing technology has become an important tool for ensuring the surface quality of forged steel rolls. Eddy current roll surface detection is based on...
Eddy current is generated when a conductor is placed in a varying magnetic field. Due to the presence of the induced eddy current electric field around the varying magnetic field, the induced electric field acts on the free charges inside the conductor, causing the charges to move and generate eddy currents.
Eddy current Testing (ET) is a non-destructive testing method that uses electromagnetic induction principle. It involves passing AC current through a coil generating a changing magnetic field. When the coil is placed close to the tested material's surface, the changing magnetic field induces eddy currents in the material surface which in turn generate a magnetic field in a direction opposite to the original magnetic field. This, in part, cancels the original magnetic field, leading to changes in the coil's resistance and inductance.
If there is a defect in the metal material, it will change the intensity and distribution of the eddy current field, which will in turn change the impedance of the coil, enabling the detection of the existence of defects.
Eddy test technology is a non-destructive testing method that evaluates the performance of conductive materials and their products by detecting changes in eddy currents without causing any damage.
In industrial production, eddy current inspection is one of the main means of controlling the quality of various metal materials and some non-metallic materials (such as graphite, carbon fiber composites, etc.).
Compared with other non-destructive testing methods, eddy current testing is more easily automated and is particularly effective in detecting defects in shaped materials such as pipes, rods, and wires.
Signal processing in eddy current testing requires improving the signal-to-noise ratio and resistance to interference to achieve signal recognition, analysis, and diagnosis, and to obtain the best signal characteristics and results.
Common feature extraction methods include Fourier descriptors, principal component analysis, and wavelet transforms.
Fourier descriptors are commonly used methods to extract feature values. Its advantages include not being affected by the probe speed, and the impedance map can be reconstructed using this method. The more the number of sampling points, the closer the reconstruction curve is to the original curve. However, this method is only sensitive to the shape of the curve, and is not sensitive to the zero-point and gain of the eddy current detector, nor does it change with changes in curve rotation, translation, size transformation and starting point selection.
Principal component analysis is a method of describing signal features using the eigenvalues and eigenvectors of the autocorrelation matrix of the test signal. This method has strong discrimination capability for similar defects.
Wavelet transform is an advanced signal time-frequency analysis method. When the multi-resolution analysis of wavelet transform is applied to the analysis of eddy current testing signals, the signal is processed by different wavelet coefficients, and then reconstructed. This signal, which has been processed by wavelet transforms, has a significantly improved signal-to-noise ratio.
Artificial neural network
The input vector of the artificial neural network is the signal's feature parameters. Correctly selecting and extracting signal's feature parameters is the key to the success of using neural network intelligent discrimination. The combined neural network model, using hierarchical discrimination method, reduces the input variable dimension from N2 to N, simplifying the network structure and greatly increasing the training speed and accuracy of defect recognition.
Neural networks can achieve defect classification with the advantage of high recognition accuracy, being effective even for incomplete or unclear data.
Information fusion technology
Information fusion involves multi-level processing of information from different sources, including detection, correlation, relevance, estimation, and synthesis, to obtain a unified estimation of the measured object.
