Application of Hybrid PSO and LS-SVR Intelligent Recognition of Dual Linear Axis Dynamic Synchronization Error Characteristics

The synchronization error of feed axis has direct relation with contour error in interpolation motions. Recognition of synchronization error of dual linear motor driven axis with gantry frame has an important role in compensating the error to avoid the effect of contour precision. In the contribution, a combination of Particle Swarm Optimization (PSO) algorithm and Least Squares Support Vector Regression (LS-SVR) machine technique for intelligent recognition of dynamic synchronization error in dual linear axis feed process is presented. The 2d-time function of laser interferometer is applied in acquisition of the dynamic synchronization error data as recognition learning patterns. In recognition of various error characteristics during different process in the feed motion, dual important parameters of LS-SVR, support vector number and error punishment factor, are tuned with PSO in an intelligent way. To demonstrate the procedure of the proposed approach, an illustration is discussed in detail. The result shows that the combination technique can decrease support number 38.5% and effectively recognize the characteristics with 9.04μm error under 40m/min feed rate motion condition.