| Resistance spot welding (RSW) is applied widely in sheet metal joining for its high efficiency, low cost and high degree of automation. In general, a typical steel car body is assembled by 300 to 400 sheet metals through 4000 to 6000 spot welds, and both the automobile production and sales volume of China is over 18 million in 2010. Therefore, it is necessary to ensure the resistance spot weld quality for car's safety and cost. However, it is difficult to monitor the resistance spot weld quality directly for the nugget formation process is unobservable and instantaneous. Thus, the resistance spot weld quality is monitored only using kinds of electric and physical parameters of welding process which is closely related with the welding quality, and the electrode displacement is considered one of the ideal monitoring parameters. However, the monitoring method of electrode displacement is not practical to industrial applications for the inconvenient installation and high cost of the displacement sensor. Most researchers do not pay attention to the mechanical vibrations and consider them as interference or useless signal to remove in RSW. However, a few researchers considered the mechanical vibrations of the welding machine and the fluctuations on the electrode displacement curve are induced by the alternating welding current and related with the nugget formation process, but they did not study it systematically. Thus, in this paper, the resistance spot weld quality was monitored using accelerometer from the study of electrode vibration. The aim of this paper is to develop a new monitoring method of resistance spot weld quality and resolve the industrial applications for the typical monitoring method of the thermal expansion displacement.At first, the monitoring system of electrode vibration was developed and the resistance spot weld process parameters, including welding voltage, welding current, electrode displacement and accelerations, were collected. Next, the mechanism of the electrode displacement fluctuation was analyzed form the consideration of the fluctuation phenomenon of electrode displacement, and the electrode displacement fluctuation characteristics were extracted and its changing law was studied experimentally. Then, from consideration of the electrode vibration, the electrode vibration signals were monitored using both displacement sensor and dual accelerometer during RSW process, and the electrode vibration signals were used to monitor resistance spot weld quality. At last, the lower electrode vibration signals were used to monitor the resistance spot weld quality for their obvious amplitude value compared with the upper electrode, and the classification estimation of the spot welds were performed using HMMs. The main research works of this paper are shown as follows.1. RSW monitoring system for electrode vibration signalsThe main tasks of RSW monitoring system include the hardware configuration of RSW monitoring and controlling system, sensing devices and measurement techniques for monitoring welding process parameters, the programs of RSW monitoring and open-loop control system developed using LabVIEW and their tests. The results showed that the accuracy, reliability and operability of the RSW monitoring system are good.2. Mechanism analysis and experimental study of the electrode displacement fluctuation characteristics during resistance spot weldingThe mechanism of the displacement fluctuation was analyzed by collecting the welding current and electrode displacement during welding process and extracted the incremental pulse expansion of the electrode displacement, which was named as the displacement fluctuation peak. Through calculation and experiments to obtain the primary dynamic resistance and metallurgical photos, the relationship of the displacement fluctuation peak with the weld nugget growth during resistance spot welding process was analyzed. The whole process can be divided into four successive stages, and five characteristic points of the displacement fluctuation peak curve were selected to online predict the weld quality. The variation in the displacement fluctuation peak curve under different welding currents and electrode force was studied, and the results showed that the displacement fluctuation characteristics can be used to effectively evaluate the weld quality.3. Monitoring method of resistance spot weld quality using dual accelerometerAlthough a displacement sensor can be used to detect the electrode displacement fluctuations, it is not suitable for industrial applications for its inconvenient installation and high cost. In this paper, acceleration signals from the upper and lower electrodes were measured using a dual accelerometer, and the amplitude curve of electrode vibration was obtained through a series of data process. Comparison of the amplitude curves of electrode vibration monitored using a displacement sensor and the dual accelerometer was performed, and the results showed that they are consistent. The effect of welding current on an electrode vibration curve monitored using the dual accelerometer was studied, and the results showed that electrode vibrations can be used to monitor weld quality, such as no weld, sticking weld, small weld, good weld and expulsion. Moreover, the monitoring of resistance spot weld quality using electrode vibration signals under abnormal welding conditions, such as shunting, small edge distance and greasy surface, was also performed.4. On-line classification estimation of resistance spot weld quality using signal accelerometer and HMMBoth upper and lower electrodes had vibrations during resistance spot weld process, and the amplitude value of the lower electrode is more obvious. Therefore, the resistance spot weld quality can be monitored using the vibration amplitude of the lower electrode which is linear correlation with the total vibration amplitude monitored form upper and lower electrodes by improving the band-pass filter parameters for acceleration signals. The vibration amplitude curves of lower electrode under different welding current were obtained through experiments, and“no weld”,“sticking weld”,“small weld”,“good weld”and“expulsion”appeared consequently with the increase of welding current. When the expulsion occurred, the vibration amplitude of lower electrode jumped greatly. Therefore, the expulsion can be monitored through setting a threshold value. The HMMs for“false weld”(including“no weld”and“sticking weld”),“small weld”and“good weld”were modeled using the vibration amplitude of lower electrode as observations and nugget diameter as states. Through pattern identification for testing samples, the results showed that the classification estimation of spot welds using HMMs is effective and reliable.In this study, by absorbing and referring the exist achievements, the mechanism of electrode displacement fluctuation was analyzed, the displacement fluctuation peak was extracted and the change mechanism of the displacement fluctuation peak during nugget growth process was studied experimentally. Moreover, in order to overcome the installation defects of displacement sensor, the electrode vibration was monitored using accelerometer, and the resistance spot weld quality monitored using electrode vibration signals was studied systematically. This research will provide references for monitoring resistance spot weld quality using accelerometer method. |
PDF Full Text Request