基于LS-SVM非线性内模控制在焊缝跟踪中的运用

Non-linear uncertain systems based on dynamic compensation inverse of LS-SVM internal model control

针对单纯的模糊控制器在焊接机器人的焊缝跟踪中的控制精度欠佳、自适应性不强等问题,设计了一种新的用于焊缝跟踪的LS-SVM非线性内模控制器。通过样本数据建立系统固定的LS-SVM逆模型,与系统串联成精确的伪线性系统,对伪线性系统采用鲁棒性强的内模控制。仿真结果表明该方法具有很好的跟踪结果。

For the non-linear uncertain systems and the traditional non-linear internal model control in the control deficiencies,this paper presents a Least Squares Support Vector Machines（LS-SVM） internal model control based on dynamic compensation inverse for nonlinear uncertain systems method.At the same time,in the introduction of LS-SVM to establish the inverse model,there will be Model Free Adaptive Control（MFAC） method as an additional controller.Both forward and backward models will be automatically modified online when they deviate from the plant.The simulation results show that the method proposed in this paper has good robustness of time-varying uncertainties remain better tracking results,with better real-time performance,robustness,and online calibration.