Study on Measurement and Control of Dynamic Behavior of Subnano-mechanism

亚纳米机构动态行为测量与控制研究

• 批准号: 06452168
• 负责人: SHIMOKOHBE Akira
• 金额: $ 4.74万
• 依托单位: Tokyo Institute of Technology
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for General Scientific Research (B)
• 财政年份: 1994
• 资助国家: 日本
• 起止时间: 1994 至 1995
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-06452168/
• 关键词: Positioning; PID control; Fuzzy control; neural network control; Optimal control; Coarse and fine positioning mechanism; Ball screw; Piezoelectric actuator; ファジイ制御; フィルタリング; 位置決め分解能; 位置決め分解能の評価; 転がり案内; 電磁モータ; ばね特性; ヒステリシス特性

The aim of this research project was to develop the coarse and fine mechanism system with a long working range on the order of 10 millimeters and high positioning resolution for realize more than sub-nanometers for realize sub-nanometer machining and measuring on the order of 10 millimeters. In the project, first, the experimental sub-nanomechanism which consists of the coarse and the fine mechaisms was made and modeled for the aim and then the best fit control methods for the mechanism were discussed. PID,Fuzzy, Neural network and Linear quadratic control method were selected for the discussion of the best fit one. In order to evaluate these control method, the specification of basic control design was determined and the robustness of the positioning systems with their controller were examined theoretically and experimentally. The results were the following ;(1) The controller based on PID algorithm for the sub-nanometer mechanism was satisfied the specification of the control design and had the good robustness to the change of the step height.(2) The controller based on Fuzzy algorithm for the mechanism was satisfied the specification and the good robustness to the change of the weight of the movable table and the addition of the constant load.(3) In case of the controller based on Neural network algorithm, the number and the kind of the weight of the neuron changed by the back propagation algorithm influenced on the performance of the system. In the experiments, the limit of the number of the weight changed is necessary for high precision positioning.(4) Though the linear quadratic controller designed for the fine positioning mechanism can control the table precisely, the controller for the coarse and fine positioning mechanism causes the vibration of its table near the reference position.

本研究项目的目的是研制具有10毫米量级的大工作范围和高定位分辨率的粗精机构系统，以实现亚纳米级以上的加工和10毫米级的测量。在该项目中，首先建立了由粗细机构组成的实验亚纳米机构，并以此为目标进行了建模，然后讨论了机构的最佳匹配控制方法。选择了PID、模糊、神经网络和线性二次型四种控制方法，讨论了最优控制方法。为了评价这些控制方法，确定了基本控制设计的指标，并从理论和实验上检验了定位系统及其控制器的鲁棒性。结果表明：(1)基于PID算法的亚纳米机构控制器满足控制设计要求，对步进高度变化具有较好的鲁棒性。(2)基于模糊算法的机构控制器满足设计要求，对动台重量变化和恒定负载的增加具有较好的鲁棒性。(3)基于神经网络算法的控制器，通过BP算法改变神经元的权值个数和权值的种类对系统性能有影响。在实验中，为了实现高精度的定位，有权重变化次数的限制是必不可少的。(4)针对精密定位机构设计的线性二次型控制器虽然可以实现对工作台的精确控制，但粗精定位机构的控制器会导致工作台在基准位置附近产生振动。

项目成果

下河辺明,佐藤海二ほか: "最適制御による粗微動位置決め機構の高精度制御" 日本機械学会第73期通常総会講演会講演論文集. (発表予定). (1996)

Akira Shimokawabe、Kaiji Sato 等人：“利用最优控制对粗、精定位机构进行高精度控制”日本机械工程学会第 73 届普通大会论文集（待发表）。

Akira SHIMOKOHBE,Kaiji SATO,et al.: ""Control of Coarse and Fine Mechanism-Positioning Characteristics Comparison between PID and Fuzzy Control"" Prepr.of JSPE. (to be published). (1996)

Akira SHIMOKOHBE、Kaiji SATO等人：“粗精机构控制-PID与模糊控制的定位特性比较”JSPE准备。

Akira SHIMOKOHBE,Kaiji SATO,et al.: ""High Precision Positioning of the Coarse and Fine Positioning Mechanism with Linear Quadratic Control Method"" Prepr.of JSME. (to be published). (1996)

Akira SHIMOKOHBE、Kaiji SATO等：“线性二次控制法粗精定位机构的高精度定位”，JSME预编。

下河辺明,佐藤海二ほか: "粗微動位置決め機構の制御に関する研究-各種制御法による特性比較-" 1996年度精密工学会春季学術講演会講演論文集. (発表予定). (1996)

Akira Shimokawabe、Kaiji Sato 等：“粗精定位机构的控制研究 - 使用各种控制方法的特性比较 -”1996 年日本精密工程学会春季学术会议论文集（待发表）。 ）

下河辺 明: "精密位置決め制御とセンサ" 1994年度精密工学会秋季大会シンポジウム資料. 31-36 (1994)

Akira Shimokawabe：“精密定位控制和传感器”1994年日本精密工程学会秋季会议研讨会资料31-36（1994）。