| Because of the advantages of low cost, no pollution, simple operation, ultrasonic testing methods are widely used in the quality inspection of automotive resistance spot welding. In order to improve the testing efficiency and accuracy, and to evaluate the quality of spot welding better, it is necessary to carry out research about vehicle spot welding automatic UT(ultrasonic testing) equipment and the relevant digital processing technology. Finally this paper achieved online automatic detection of spot welding quality.This paper first introduced the principle of ultrasonic testing of spot welding and the development of spot welding automatic UT system. The design of automatic UT system mainly includes the hardware design, software design and the design of detection actuator. The automatic UT system is the carrier of the acquisition of digital ultrasonic signal and implementation of the technique in this paper. Then according to the characteristics of ultrasonic signal of spot welding, this paper made research on the parameters estimation method of ultrasonic echoes and the automatic evaluation method of spot wel ding quality. The theory and method in the paper were validated by experiment and test. The main achievements of this paper include the following two aspects:1) In the process of spot welding automatic UT, due to the size of spot welding is thin, the echoes of defects and the echoes of spot welding interface are extremely easy to overlap, the ultrasonic signal is probably an overlapping multi-echoes signal with noise. By establishing linear convolution model and nonlinear parameterized model of multi-echoes, a parameters estimation method of ultrasonic echoes was proposed in this paper, which combining with minimum entropy deconvolution and the basic idea of EM algorithm. The method has realized the separation of overlapped signals and the estimation of parameters such as time of flight, number of echoes, center frequency and amplitude. The effectiveness and advantages of the proposed method were verified through simulation signal and experiment. This method can effectively extract the signal characteristics of ultrasonic echoes. The conditions for highly accurate detection of spot welding were provided.2) In order to realize the automatic evaluation of spot welding quality, a simple and effective and highly accurate re al-time automatic evaluation method of spot welding quality should be studied, according to the characteristics of the ultrasonic echoes of spot welding. A spot welding quality automatic evaluation methodology was proposed in this paper based on analysis o f ultrasonic signal characteristics of various spot welding and definition of time-frequency characteristics of ultrasonic echoes. Through the aforementioned parameter estimation method and the analysis method of signal energy, the time-frequency characteristics of ultrasonic echoes were extracted. Then a spot welding quality evaluation criterion which is advantageous for online automatic detection was established based on relevant testing flow. Practical detection results indicated that the UT system based on the method of this paper can automatically identify seven types of spot welding quality quickly and effectively, so this system can preliminarily be applied to online automatic detection of spot welding quality. |
PDF Full Text Request