Nonlinear Error Compensation of Capacitive Angular Encoders Based on Improved Particle Swarm Optimization Support Vector Machines

Abstract

Rotary encoders are widely applied in a variety of industrial fields. However, as the exist of the installation, processing and demodulation circuits errors, the test result of the encoder is superimposed with periodic nonlinear errors and the encoder needs compensation to achieve high measurement accuracy. Traditional methods including the least square method (LSM) and back propagation artificial neural network (BP-ANN), are not capable of addressing nonlinear errors. Thus, a novel method based on improved particle swarm optimization (IPSO) and support vector machines (SVM) is proposed to provide better compensation. The proposed method incorporates the SVM method into the design of the compensation model, and the IPSO algorithm is applied to tune the SVM parameters. To validate the algorithm, four sets of data were obtained from encoders with different numbers of segments. The experimental results show that the IPSO-SVM algorithm has a better prediction precision and the nonlinear standard deviation of 180 petal-shaped numbers has dropped from 0.08 • to 0.0005 • after compensation over 0 • to 360 • measurement range. Based on the results, the proposed IPSO-SVM model provided more accurate compensation on the nonlinear errors to the capacitive angular encoders than other method.

Prediction and optimization of employee turnover intentions in enterprises based on unbalanced data

“…The traditional Support Vector Machine (SVM) can lead to the resulting hyperplane being more biased toward minority class, making them misclassified as majority class. Therefore, how to improve the recognition rate and overall performance of minority class in unbalanced data is the important research topic in the field of machine learning [10][11][12].…”