Fuzzy PID control for impact force of high speed wire bonding process

The Z-axis motion acceleration of the wire bonding machine is very high (>100g) and may cause serious mechanical vibration and large impact force resulting in IC chip damage, optimal control of the impact force is one of the key factors to improve the bonding quality. How to reduce the vibration energy of the Z-axis motion in a very short time and reduce the force fluctuation is the main concern of this paper. Considering the effects of the mechanical motion, vibration and environment factor to the bonding control system, this paper proposed an improved fuzzy PID algorithm to control the overshot and fluctuation of the impact force. This algorithm inputs the deviation (e) of the system output and the variation rate(ec) of the system to the fuzzy controller, and outputs the fine-tuning value(ΔKp, ΔKi, ΔKd) of the parameters of the proportional, integral and differential gain. With the online adjustment of the PID parameters the system can rapidly adapt of any change. For the control scheme of the system, we adopted a Kalman filtering switch algorithm to predict the right switch condition for the fuzzy module control, which is based on the impedance current variation of the z-axis motor. The fuzzy PID control algorithm is simulated and compared with the conventional PID control algorithm through Matlab tool. The control algorithm is implemented and tested through bonding experiments. The experimental results indicated that the Z-axis impact force was reduced significantly and the force fluctuation was controlled within 3%, which will greatly improve the bonding quality of the wire bonding machine with high-speed motion.