Resistance spot welding (RSW) is one of the most widely used sheet metal joining methods in modern manufacturing industry. Its process is a metal melting and liquid nugget growth process through the input of electrical power. There are a number of elements which affect the weld quality. Hence, real time control of the RSW process plays a critical role in obtaining welds of a satisfactory quality. Currently, ensuring weld quality through process control has been and remains a major challenge and worthwhile goal during the welding manufacturing process. Though many previous works have focused on online quality estimation or control, these methods have some drawbacks during practical application and need to be improved.

The objective of this work is to develop an online quality control system for RSW, which can be used to generate welds of a satisfactory quality. The selected quality criterion should be a predetermined reference condition and also an input to the system. For obtaining welds of satisfactory quality, the most important thing is obtaining the qualitative information in real time. Because the duration of one welding operation is very short and no effective sensor can be used to provide the internal information of the welds in real time, the dynamic resistance curve, which has a particular variation tendency during the welding process, is employed to analyze the nugget formation and growth process. According to analysis of the curve, the time when the first melting point of the welding process appears can be found, because the property of the weld undergoes an obvious change at this point, which can be monitored via the dynamic resistance curve with a higher resolution than that used during the use of the traditional method. Using the first melting point, the heat energy which contributes to nugget formation and growth can be more precisely obtained because the heat energy which is used for eliminating the contact resistance is abandoned. Hence, the model of the relation between the heat energy and nugget diameter, which is used for estimation of weld quality in this thesis, can be established. This model is composed of a segmented polynomial function of heat energy absorbed by the weld from the time when the first melting point appears for each welding current, and is easy to be realized in practical applications. To obtain a consistent mathematical model for the specific workpiece, a constant current control (CCC) strategy should be used because it is the only control strategy that can assure constant nugget formation and growth. The operation of the RSW electrical system is to input a firing angle for the corresponding silicon-controlled rectifier (SCR) during each control cycle. The first thing is to ensure that each firing angle is safe;​ hence, power factor angle should be obtained online because each firing angle must be larger than or at least equal to it. Also, the power factor can be obtained correspondingly. Then, a nonlinear mathematical model of the relation between each firing angle and the welding current can be established. Combining a proportional-derivative (PD) controller, a new nonlinear controller to implement the CCC strategy is proposed. It can achieve better performance than that achieved using a traditional PID controller.

In this dissertation, contributions were made to the development of:​ online measurement of the power factor angle and power factor, a nonlinear controller to implement CCC strategy, online measurement of dynamic resistance with a high resolution and algorithm to detect the time when the first melting point appears, and a mathematical model of the relation between heat energy and nugget diameter. Based on these contributions, this newly developed online quality control system can obtain satisfactory performance in producing welds with predetermined nugget diameter. Finally, experimental results validated that the effectiveness of the proposed control system. The error between predefined nugget diameter and actual measurement of nugget diameter is very small. Moreover, the integrated control system is easily realizable. Therefore, the proposed quality control system may have a profound impact on how the welding process being conducted if it is be implemented in actual industrial production. As a result, welds of high quality can be produced by RSW with efficient usage of the energy. Though the system is developed using single-phase AC RSW, this original research can be used in three-phase MFDC as well, this is because they have the same working principle only except in the mode of input of electrical power.