Inertia Estimation And Smooth Start For The Servo Motor

In this thesis,the problems of inertia estimation and smooth start for the servo motor under unknown load are studied.The main contents are as follows:(1)The background and research significance of the problem of inertia estimation are introduced,and the development status of this technique at home and abroad is expounded.(2)The types of vibration during motor starting process are summarized,and they are sorted into the first type of vibration and the second type of vibration.The causes of various vibrations are analyzed,and the suppression methods for different situations are introduced.It is pointed out that the S-curve acceleration and deceleration can suppress the first type of vibration,and the high precision estimation of moment of inertia helps to suppress the second type of vibration.This thesis combines the S-curve acceleration and deceleration with the algorithm of moment of inertia estimation to help the servo motor to start smoothly without the first and second kinds of vibration at the same time.(3)The inertia estimation algorithm based on recursive least squares is improved.Tracking differentiator is introduced into the pre-processing of input data of estimator.A kind of recursive least squares algorithm combined with tracking differentiator is formed.Comparing with traditional algorithm,this algorithm enhances the anti-jamming ability and improves the estimation accuracy.(4)Aiming at the problem of inertia estimation under the condition of dragging large inertia load,the calculation of angular acceleration of the rotor is avoided through reasonable deformation of the mechanical motion equation of servo system.As a result,the noise of input data is reduced,the calculation process is simplified,and a kind of inertia estimation algorithm based on recursive least squares and integral torque is formed.Comparing to traditional algorithm,this algorithm has the ability of estimating large moment of inertia accurately.(5)The simulation model of servo system is built in MATLAB/Simulink,and the simulation experiments show the correctness of the algorithms introduced in this thesis.Engineering experiments are carried out on the self-developed servo driver to verify the engineering feasibility of the algorithms.