Precision positioning of LPM(linear piezoelectric motor) table system using adaptive wavelet and back propagation neural network control
碩士 === 明志科技大學 === 機電工程研究所 === 101 === Since the piezoelectrically actuated system has nonlinear and time-varying behavior, it is difficult to establish an accurate dynamic model for a model-based sensing and control design. In this thesis, two kinds of model-free adaptive neural network controller i...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2013
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/21336243186065314641 |
| Summary: | 碩士 === 明志科技大學 === 機電工程研究所 === 101 === Since the piezoelectrically actuated system has nonlinear and time-varying behavior, it is difficult to establish an accurate dynamic model for a model-based sensing and control design. In this thesis, two kinds of model-free adaptive neural network controller including back propagation neural network (BPNN) and wavelet neural network (WNN) are proposed to overcome the dead-zone and hysteresis nonlinearities of the linear piezoelectric motor (LPM) system. The important advantages of these approaches are to achieve the controllers design without knowledge of the system dynamic model. The proposed controllers are implemented on a LPM-actuated table while the motion control performances of the system are investigated. Experimental verifications including regulating and tracking control are performed firstly to assure the reliability of the proposed control schemes. A real dispensing is then conducted to validate the feasibility of the LPM-actuated drop-on-demand droplet generator. The results demonstrate satisfactory robustness and accuracies.
|
|---|