A segmentation scheme with SVR-based prediction in stroke-sensing cylinder

作者：Liu, Lu; Guo, Yanqing^*; Fu, Yongling; Li, Chuangchuang; Guo, Liang

来源：Cluster Computing, 2019, 22(2): S2875-S2887.

DOI：10.1007/s10586-017-1645-2

• Stroke-sensing cylinder
• Signal segmentation
• Tangent subdivision
• Support vector regression

Signal processing in stroke-sensing system of cylinder has been addressed by researchers from very different fields. However, the low accuracy of subdividing technique results in the difficulties in signal segmentation. This paper proposes methodology analyzes the measuring principle and subdivision error of the displacement detection. A segmentation scheme based on support vector regression (SVR) algorithm is developed to efficiently compensate the complexity and nonlinearity in tangent subdivision, and it enables accurate modeling and predicting of angle variation by applying SVR algorithm. Further, the regression model improves the accuracy by resolving the segmentation interval. Due to the significances presented in this work, the detection error decreases from 1.59 to 0.286%, when processed with the proposed segmentation algorithm. Experimental results are statistically analyzed, which makes it a promising basis for the realization of exact position detection.

• 出版日期2019-3
• 单位中北大学; 北京航空航天大学