Adaptive data fusion method based on Kalman filter and its application in piston shape and position detection

Abstract

In order to meet the needs of production, this paper studies the detection data processing algorithm for the angle of the piston shape and position, the groove width and the runout of the groove bottom. An adaptive data fusion method based on Kalman filtering is proposed. This algorithm can use the adaptive method to calculate the different weight coefficients of each data point according to the standardized sampling data and the smoothed data points obtained by the cubic B-spline approximate fitting algorithm. This paper obtain the fusion data; then calculate the groove angle, groove width and groove bottom runout of the piston according to the conversion formula. Finally, experimental data verification is carried out, which shows the effectiveness and good robustness of the data processing of this algorithm. At the same time, a piston shape and position detection system was developed. The system rotates a circle at 90 degrees per second to obtain the measurement data. After the system software and hardware processing, the groove angle detection accuracy can reach ±2 minutes, the groove width and the groove bottom. The detection accuracy of the runout amount is within ±2 microns, and the measurement items in the measurement time and accuracy side meet the customer’s technical document requirements and achieve satisfactory results.