Dynamic Error Analysis of CMM Based on Variance Analysis and Improved PLSR

Zhang Mei, Cheng Fang, Li Guihua


It is difficult to build an accurate model to predict the dynamic error of CMM by analyzing error sources. An innovative modeling method based on Variance Analysis and Improved Partial Least-square regression (IPLSR) is proposed to avoid analyzing the interaction of error sources and to overcome the multi-collinearity of Ordinary Least-square regression (OLSR). Among many impact factors the most influential parameters are selected as the independents of the model, by means of variance analysis.The proposed modeling method IPLSR can not only avoid the analysis of the error sources and the interactions, but can also solve the problem of multi-collinearity in OLSR. From experimental data the expository capability of this IPLSR model can be calculated as 85.624 percent, and the mean square error is 0.94μm. As comparison, the mean square values of conventional PLSR and OLSR are 1.04μm and 1.39μm, respectively. So IPLSR has higher predicting precision and better expository capability.

Full Text:


DOI: http://doi.org/10.11591/ijeecs.v12.i7.pp5342-5349


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics