Inverter resistance welding parameter error prediction quality control method

Abstract

The invention discloses an inverter resistance welding parameter error prediction quality control method. A dsPIC33 digital signal controller is adopted as an operation processor, a welding parameter prediction algorithm is operated on the processor, and a welding control algorithm is simultaneously operated; the output current of secondary side rectifying is collected through a current sensor, the electrode displacement amount is detected through a displacement sensor, and the electrode voltage is detected through a voltage sensor; the parameters obtained through the sensors are compared with preset expected values, a parameter error prediction value sequence is built, a simplified prediction model containing the three parameters is built, a welding parameter error prediction value of the next sampling moment is calculated through recursion through a least squares support vector machine prediction method, then the error prediction value is sent to the controller before the control amount is sent out, the controller is regulated in advance according to the error prediction value to control output, and then quality control is achieved.

Description

technical field [0001] The invention relates to an inverter resistance welding quality monitoring and control method, in particular to a quality control method for error prediction of welding parameters such as welding current, electrode displacement, and electrode voltage used in high-power inverter resistance welding. Background technique [0002] The inverter constant current resistance welding machine has changed the traditional power frequency chopper control method, and adopted the inverter control process to increase its working frequency to the kilohertz level and the response time to the millisecond level, which is more efficient than the traditional power frequency resistance welding machine Greatly improved, so the response speed is fast, the control precision is high, the welding performance is good, the effective power is high, and the power saving is more than 30% compared with the traditional welding, and there is no pollution and no impact on the power grid. ...

AI Technical Summary

Problems solved by technology

[0005] But at present, the detection of welding parameter error is input to the controller as a feedback amount after the control amount of the control system is sent out, which causes the lag of the error parameter in the control, and the feedback of the welding parameter error can only negatively indicate that it has been completed However, if the error of these welding parameters can be predicted before the control quantity is issued, the controller can make compensation for the control quantity in advance, and effective quality control of resistance welding can be realized

[0006] However, as far as the patent applicant knows, there is currently no publicized method for predicting the errors of welding parameters such as inverter resistance welding welding current, electrode displacement, and electrode voltage.

Images