增益模糊自整定的增量式神经元非模型控制及其应用  

MODEL FREE INCREMENTAL NEURON CONTROLLER WITH FUZZY SELF-TUNING GAIN AND ITS APPLICATIONS

作　　者：王宁[1] 张建明[1] 王树青[1] 

机构地区：[1]浙江大学工业控制技术国家重点实验室先进控制研究所,杭州310027

出　　处：《模式识别与人工智能》2000年第3期305-308,共4页Pattern Recognition and Artificial Intelligence

基　　金：国家自然科学基金;浙江省自然科学基金

摘　　要：车削系统的切削力控制是机械加工自动化的关键环节,目前仍未能有效地解决切削过程中切削深度变化时控制系统的稳定性等问题。本文提出了增益模糊自整定的增量式神经元非模型控制方法,通过在线学习来调节控制系统的增益,使切削力在切削深度大范围变化时保持恒定,达到改善切削过程的动态品质、增强控制系统稳定性的目的。使用该方法对某车削系统进行了仿真实验研究,结果表明这种新颖的控制方法非常简单,具有很强的鲁棒性、抗干扰性和满意的控制品质。The constant turning force adaptive control system design is the key link of machining automation. By now, the control stability while cutting a workpiece at varied cutting depth is still an existing problem. In this paper, a model-free incremental neuron control method with fuzzy self-tuning gain is proposed for the cutting process. In order to enhance the stability of control system and improve the dynamic performance of the system, the proposed control method tries to keep the cutting force being constant by changing the control system gain on-line when a cutting tool cuts at various cutting depths or a spindle operates at different speeds. The simulation results of a cutting process control show that very strong robustness, good disturbance resistibility and satisfied performances can be obtained. The new control method is very simple.

关 键 词：增量式神经元 非模型控制 增益模糊自整定 车削 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置]