Research On On-line Analysis Of Resistance Spot Welding Quality Based On Dynamic Resistance

Resistance spot welding is a welding forming method which is widely used invarious fields of industry. It has some merits such as short welding time, low cost,high efficiency, simple operation, high degree of mechanization and automation andso on. Lean production has promoted enterprises to put forward the higherrequirements of quality control and monitoring in spot welding process. There are twomainly traditional method for quality monitoring of spot weld, cycle sampling fordestructive testing and nondestructive testing. However, cycle sampling fordestructive testing leads to the cost of welding production and the waste of weldingmaterial increasing. Due to its own characteristics, the method of materialnondestructive testing cannot be widely applied in the quality monitoring of spotwelding. Therefore, developing a method which can be used for on-line analysis ofresistance spot welding quality is very significant.The formation and growth process of weld nugget was conducted under thecompletely closed condition, and there are also some casual factors which influencethe quality of welding joint. Therefore, it is relatively difficult to monitor spot weldingquality directly. It must depend on monitoring the dynamic information which canreflect spot welding quality to better solve this problem. The analysis of study of spotwelding process showed that the dynamic resistance has a very close relationship withjoint quality and can decent reflect the formation and growth process of weld nugget.By researching the spot welding process of the SUS301-HT stainless steel, fivecharacteristics has been extracted from the parameters such as welding current,voltage between electrodes, and dynamic resistance. These characteristics are theeffective value of welding current, the average value of electrode voltage, the averagevalve of dynamic resistance, the minimum and maximum value of dynamic resistance.The tensile shear strength of spot welding joint has been regard as evaluation index ofthe spot welding quality. The analysis result showed that the five characteristics haveclose relationship with the shear strength of welds. A BP neural network model based on dynamic resistance whose input vectors are the five characteristics and outputvector is the shear strength of welds has been developed to finish the prediction ofspot welding quality. The system realize the online quality analysis function of spotwelding with the auxiliary information,such as the effective value of welding currentand the actual output value of electrode pressure.Based on advanced sensor technology and data collection technology, a set ofdata acquisition and processing hardware system was designed for the real-timedynamic information of spot welding process synchronous data acquisition in thispaper. Rogowski coil and integrator were used to collect welding current of spotwelding process, twisted pair electrode leads and isolating circuit module were usedto collect voltage between the two electrodes, and gas pressure transmitter was used tocollect actual pressure which output by spot welding equipment. The industrialcomputer get electrical signals which were switched and sent by the data acquisitioncard.In this paper, an on-line analysis software system for spot welding quality whichinclude function of data acquisition, data processing, quality analysis, data storage,information display and data query has been developed. Database has also beendesigned to support the spot welding quality analysis and intelligent management. Thesoftware system realize the function of on-line intelligent analysis of spot weldingquality. To test the accuracy and reliability of the online analytical system based ondynamic resistance, corresponding experiment was designed in this paper. The resultof experiment shows that the monitoring system has high precision and ability ofanti-interference.