An Active Vibration Isolation Control Method Based on Voice Coil Motor

Abstract

The invention discloses an active vibration isolation control method based on a voice coil motor. The method adopts an approximate self-adaptive dynamic programming method based on a neural network, and does not need to establish an accurate dynamic model of the vibration isolation device. By measuring the input and output of the vibration isolation device value, train the neural network controller to achieve the specified performance index, and when the two networks reach the convergence condition, the required active vibration isolation controller is obtained to realize vibration isolation control. In the present invention, the sensor information used to realize the controller is acquired through a linear variable differential transformer (LVDT) type displacement sensor and an acceleration sensor. The evaluation network and behavior network in the controller are trained at the same time, and the weights are updated using the least squares gradient descent method. The invention overcomes the shortcomings of the existing conventional control system, improves the usability of active vibration isolation control, reduces the requirement for dynamic modeling of the vibration isolation device, and has certain universality.

Description

technical field [0001] The invention relates to the technical field of vibration isolation control, and more specifically, relates to an active vibration isolation control method based on a voice coil motor. Background technique [0002] Vibration exists in the vast majority of electromechanical systems and can have serious consequences if not properly isolated. For example, in the manufacture of integrated circuits, precision machinery and measuring instruments are very sensitive to vibration. Vibration will affect the machining accuracy, shorten the life of the mechanical device, and even cause serious damage to the entire system. Therefore, in order to meet the strict and special requirements of the vibration environment in precision machining, measurement and other occasions, especially the requirements for effectively suppressing low-frequency and ultra-low-frequency vibrations, reasonable vibration isolation devices and efficient vibration isolation control methods ar...

AI Technical Summary

Problems solved by technology

However, the force and displacement of the electromagnetic drive have large nonlinearity.

At the same time, considering the inherent nonlinear characteristics of the spring and damper in the active vibration isolation device, it is impossible to establish an accurate dynamic model of the active vibration isolation device based on the electromagnetic drive

[0005] Therefore, general conventional control methods and modern control methods are difficult to guarantee the control accuracy, or are too complicated and cumbersome to meet the needs of real-time control.

Images