Model-Following Controller Based on Neural Network for Variable Displacement Pump

スポンサーリンク

概要

• 論文の詳細を見る

• The variable displacement axial piston pump (VDAPP) is inherently nonlinear, time variant and subjected to load disturbance. The controls of flow and pressure of VDAPP are achieved by changing the swashplate angle. The swashplate actuators are controlled by an electro-hydraulic proportional valve (EHPV). It is reasonable for swashplate angle of a VDAPP to employ neural network based on adaptive control. In this study, the nonlinear model of the VDAPP with a three-way electro-hydraulic proportional valve is proposed, and a neural network model-following controller is designed to control the swashplate swivel angle. The time response for the swashplate angle is analyzed by simulation and experiment, and a favorable model-following characteristic is achieved. The proposed neural controller can conduct nonlinear control in VDAPP, enhance adaptability and robustness, and improve the performance of the control system.

• 一般社団法人日本機械学会の論文
• 2003-03-15

著者

• Chang Chuan-wei

  Department Of Mechanical Engineering Chung Yuan Christian University

• Liu Yuan-liang

  Department Of Mechanical Engineering Chung Yuan Christian University

• CHU Ming-Hui

  Department of Mechanical Engineering, Tung Nan Technology College

• KANG Yuan

  Department of Mechanical Engineering, Chung Yuan Christian University

• CHANG Yih-Fong

  Department of Mechanical Engineering, Chung Yuan Christian University

• Kang Yuan

  Department Of Mechanical Engineering Chung Yuan Christian University

• Chu Ming-hui

  Department Of Mechanical Engineering Tung Nan Technology College

• Chang Yih-fong

  Department Of Mechanical Engineering Chung Yuan Christian University

• Kang Y

  Department Of Mechanical Engineering Chung Yuan Christian University

関連論文

• Model-Following Controller Based on Neural Network for Variable Displacement Pump
• The Minimization Method of Measuring Errors for Balancing Asymmetrical Rotors

スポンサーリンク