Abstract: In order to improve the roundness measurement accuracy of large diameter revolving parts under rotating state, an optimization model is established based on the influences of positions of three measuring points on the measurement accuracy, and an optimization strategy of position of three points is presented on the basis of particle swarm and genetic hybridization algorithm. The influence of sensor position on roundness extraction accuracy is analyzed, and the optimization function of measuring point position based on the error transfer coefficients is established by using the outer contour measurement model and roundness extraction technology. A hybrid algorithm of a new particle population is generated through particle swarm optimization and genetic complement, and the circumferential angles of the sensors are obtained under the given harmonic orders. The circumferential layout angles of the measuring point under the condition of the given harmonic orders (160,240, and 480) are calculated, and the error transfer coefficients before and after optimizing the position of measuring points are compared. The accuracy of roundness extraction was tested with the virtual and real revolving parts. The results show that the error transfer coefficient after position optimization is less than that before optimization, and the extracted roundness value is closer to the true roundness value with accuracy of about 0.010 mm. The proposed optimization strategy can be used to improve the measurement accuracy of the revolving part roundness, and provides an effective detection means for manufacturing high precision and large diameter revolving parts. Key

Key words: largediameterrevolvingpart, measuringpointoptimization, measurementaccuracy, three-pointmethod, particleswarmoptimization, geneticalgorithm, roundness