| Piezoelectric vibration feeder is widely used in automated production lines, because of its smart bulk, low noise, especially the low energy consumption. But the variation of parts weight lead to a comparatively large variation in moving speed. To control moving speed we need to acquire feedback signal, in ordinary way an external sensor is necessary which will change the feeder’s structure.To avoid this, we used bilaterally reversible piezoelectric effect and obtained moving speed signal from piezoelectric actuator itself. In this way we achieve a close-loop control for moving speed. We deduce the observer model according to the third class of piezoelectric equations. This model have two input, one is the driving voltage Vin, and the other is the sampling voltage Vout that stands for the working current of piezoelectric ceramic, which could be acquired by adding a resistor to the driving port; the model outputs are moving speed v and parts weight F. An item of constant stress was introduced in the observer model to represent an initial stress applied to the piezoelectric actuators by feeder’s structure.From this model we found the relationship among moving speed v, the increment of driving voltage ΔVin and the sampling voltage Vout in a fixed resonant frequency band. And there are only two parameters need to be identified. To make parameter identification, record the moving speed when varying parts weight at different driving voltage.The speed control experiments were carried on a0.5kg feeder. The experimental results show that the self-sensing control algorithm can regulate the moving speed effectively, also can make a real-time calculation of parts weight, which is meaningful for the automatic management of piezoelectric vibration feeder. If a compensation for hysteresis and creep is added in the observer model, the accuracy of the model will be improved, which will make the observer model suitable for high accuracy requirements occasions. |
PDF Full Text Request