孫宜標(biāo), 王亞朋, 劉春芳(沈陽(yáng)工業(yè)大學(xué) 電氣工程學(xué)院, 沈陽(yáng) 110870)
電氣工程
永磁直線電機(jī)自適應(yīng)區(qū)間二型模糊滑??刂?
孫宜標(biāo), 王亞朋, 劉春芳
(沈陽(yáng)工業(yè)大學(xué) 電氣工程學(xué)院, 沈陽(yáng) 110870)
針對(duì)永磁直線同步電機(jī)伺服系統(tǒng)受到系統(tǒng)參數(shù)變化、負(fù)載擾動(dòng)而降低其性能的問(wèn)題,考慮端部效應(yīng)以及摩擦力的存在,提出一種自適應(yīng)區(qū)間二型模糊滑??刂品椒?采用自適應(yīng)區(qū)間二型模糊系統(tǒng)逼近滑??刂频刃Э刂撇糠郑褌鹘y(tǒng)的一型模糊系統(tǒng)擴(kuò)展到區(qū)間二型模糊系統(tǒng),提高系統(tǒng)面臨參數(shù)變化、不確定性擾動(dòng)時(shí)的處理能力.基于Lyapunov函數(shù)設(shè)計(jì)切換項(xiàng)增益調(diào)整的自適應(yīng)律,保證系統(tǒng)的穩(wěn)定性.仿真結(jié)果表明,該方法提高了系統(tǒng)的魯棒性,有效地削弱了系統(tǒng)的抖振.
永磁直線同步電機(jī); 滑??刂?; 區(qū)間二型模糊系統(tǒng); 一型模糊系統(tǒng); 自適應(yīng)律; Lyapunov函數(shù); 魯棒性; 抖振
永磁直線同步電機(jī)(PMLSM)的高速、高精度和高加速度性能使其在數(shù)控機(jī)床、芯片制造等高端裝備制造業(yè)中得到了廣泛應(yīng)用[1].直驅(qū)伺服與傳統(tǒng)旋轉(zhuǎn)電機(jī)構(gòu)成的伺服相比,取消了負(fù)載與電機(jī)之接的中間傳動(dòng)環(huán)節(jié),使其成為伺服驅(qū)動(dòng)技術(shù)的發(fā)展趨勢(shì).但直接驅(qū)動(dòng)的方式也使外部干擾直接作用在電機(jī)上,系統(tǒng)易受不確定擾動(dòng)的影響而降低性能.為了實(shí)現(xiàn)直驅(qū)伺服系統(tǒng)的優(yōu)勢(shì)性能,必須采用適當(dāng)?shù)目刂品椒ㄌ岣呦到y(tǒng)的魯棒性[2-3].
滑??刂?SMC)強(qiáng)魯棒性的優(yōu)點(diǎn)使其在伺服系統(tǒng)設(shè)計(jì)中受到廣泛重視,但需要解決滑模控制的抖振問(wèn)題,滑模控制與模糊控制(FC)相結(jié)合則可以削弱抖振[4-5].文獻(xiàn)[6]把一種自適應(yīng)模糊滑??刂茟?yīng)用在伺服控制系統(tǒng)中,證明了系統(tǒng)的穩(wěn)定;文獻(xiàn)[7]建立了完整二型模糊邏輯系統(tǒng),證明二型模糊系統(tǒng)比一型模糊系統(tǒng)可以更有效地處理不確定信息;文獻(xiàn)[8-10]研究了二型模糊控制器與滑模控制及其他控制方法的結(jié)合,并分析了多變量系統(tǒng)和非線性系統(tǒng)中的應(yīng)用問(wèn)題.
本文考慮永磁直線同步電機(jī)的參數(shù)變化、負(fù)載擾動(dòng)、端部效應(yīng)以及摩擦力,提出一種自適應(yīng)區(qū)間二型模糊滑模控制方法.設(shè)計(jì)了單輸入自適應(yīng)區(qū)間二型模糊系統(tǒng)逼近滑??刂频刃Э刂撇糠?,利用區(qū)間二型模糊系統(tǒng)的優(yōu)勢(shì),提高系統(tǒng)面臨參數(shù)變化、不確定性擾動(dòng)時(shí)的處理能力.基于Lyapunov函數(shù)設(shè)計(jì)切換項(xiàng)增益調(diào)整的自適應(yīng)律,保證系統(tǒng)的穩(wěn)定性,仿真結(jié)果表明了該控制方法的有效性.
PMLSM的電磁推力表達(dá)式[2]為
(1)
式中:τn為極距;pn為極對(duì)數(shù);ψf為永磁體在動(dòng)子繞組直軸上的磁鏈分量;id、iq、Ld、Lq分別為d、q軸的電流、電感.
面裝式PMLSM中,Ld=Lq,則式(1)可表示為
(2)
式中,kf為電磁推力系數(shù).
PMLSM的機(jī)械運(yùn)動(dòng)方程為
(3)
式中:l為動(dòng)子位移;M為動(dòng)子和負(fù)載質(zhì)量;Bv為粘滯摩擦系數(shù);d(t)為外部干擾,d(t)=Ffric+Frip+Fl;Ffric為摩擦力;Frip為端部效應(yīng)推力波動(dòng);Fl為負(fù)載阻力.
令狀態(tài)量x=[x1,x2]T=[l,v]T,u=iq為輸入控制量,則系統(tǒng)的狀態(tài)方程為
(4)
定義位移跟蹤誤差為
(5)
2.1 積分滑??刂破髟O(shè)計(jì)
建立積分滑模面模型為
(6)
式中:k1和k2為非零正常數(shù).
(7)
根據(jù)式(4)、(7)可得滑??刂频牡刃Э刂茷?/p>
(8)
但是ueq在系統(tǒng)受到擾動(dòng)時(shí)控制目標(biāo)難以實(shí)現(xiàn),采用區(qū)間二型模糊系統(tǒng)逼近ueq可以解決該問(wèn)題.
2.2 自適應(yīng)區(qū)間二型模糊滑??刂?/p>
由于PMLSM伺服系統(tǒng)是一個(gè)對(duì)控制器要求較高的具有不確定性的較強(qiáng)非線性系統(tǒng).采用模糊滑模控制理論設(shè)計(jì)控制器,可以降低對(duì)系統(tǒng)模型的依賴程度,可以削弱抖振并具有較強(qiáng)魯棒性,適用于非線性系統(tǒng).但普通一型模糊系統(tǒng)由于自身的局限,處理不確定性能力有限,所以使用區(qū)間二型模糊系統(tǒng)[11-12]與滑模控制相結(jié)合設(shè)計(jì)控制器,可以更好地處理不確定性擾動(dòng).
一個(gè)二型模糊系統(tǒng)包括模糊器、規(guī)則庫(kù)、推理機(jī)、降型器以及解模糊器五部分,如圖1所示.
圖1 二型模糊系統(tǒng)Fig.1 Type-2 fuzzy logic system
考慮單輸入的模糊系統(tǒng)不會(huì)出現(xiàn)模糊規(guī)則庫(kù)指數(shù)式的增大,區(qū)間二型模糊系統(tǒng)模糊規(guī)則定義為
圖2 區(qū)間二型模糊集Fig.2 Interval type-2 fuzzy set
(9)
區(qū)間二型模糊高斯隸屬函數(shù)由可調(diào)整不確定中心值[m1,m2]與可調(diào)整標(biāo)準(zhǔn)方差σ構(gòu)成.
基于乘機(jī)推理機(jī)與單值模糊器,經(jīng)過(guò)集合中心(center-of-sets,COS)降型得到降階集為
Ycos(Y1,Y2,…,Y7,A1,A2,…,A7)=(yl,yr)=
(10)
重心法解模糊后的清晰輸出為
(11)
(12)
(13)
將式(12)和式(13)代入式(11)可得
(14)
ufz(s,α)=αTξ
(15)
式中,α=[α1,α2,…,α7]T.
根據(jù)模糊逼近理論,存在一個(gè)最優(yōu)模糊系統(tǒng)ufz(s,α*)來(lái)逼近ueq,即
ueq=ufz(s,α*)+ε=α*Tξ+ε
(16)
(17)
(18)
式中,uvs為滑模控制的切換項(xiàng).
2.3 切換項(xiàng)增益的自適應(yīng)律設(shè)計(jì)
(19)
(20)
由式(6)和式(8)可得
(21)
由式(18)和式(21)可得
(22)
定義Lyapunov函數(shù)為
(23)
(24)
(25)
uvs=E(t)sgn(s(t))
(26)
則式(24)變?yōu)?/p>
-E(t)|s(t)|λ+|ε||s(t)|λ=
-(E(t)-|ε|)|s(t)|λ≤0
(27)
通常切換項(xiàng)增益E(t)難以確定,往往通過(guò)經(jīng)驗(yàn)確定,如果E(t)值過(guò)大,系統(tǒng)會(huì)產(chǎn)生較大抖振;反之,系統(tǒng)受到擾動(dòng)后不能恢復(fù)到原來(lái)的狀態(tài).
(28)
定義估計(jì)誤差為
(29)
則重新定義Lyapunov函數(shù)為
(30)
式中,β1和β2為正常數(shù),則
(31)
(32)
則式(31)變?yōu)?/p>
|ε||s(t)|λ-E|s(t)|λ=(|ε|-E)|s(t)|λ≤0
(33)
自適應(yīng)區(qū)間二型模糊滑??刂葡到y(tǒng)的結(jié)構(gòu)如圖3所示.
圖3 自適應(yīng)區(qū)間二型模糊滑模控制系統(tǒng)Fig.3 Adaptive interval type-2 fuzzy sliding mode control system
表1 一型模糊和區(qū)間二型模糊隸屬函數(shù)Tab.1 type-1 fuzzy and interval type-2fuzzy membership functions
圖4 位移階躍響應(yīng)曲線Fig.4 Step response curve of displacement
圖5 參數(shù)變化前后位移階躍響應(yīng)曲線Fig.5 Step response curve of displacement beforeand after parameter variation
圖6 自適應(yīng)一型模糊滑??刂破鞯妮敵鰅q曲線Fig.6 Output iq curve of adaptive type-1fuzzy sliding mode controller
圖7 自適應(yīng)區(qū)間二型模糊滑模控制器的輸出iq曲線Fig.7 Output iq curve of adaptive interval type-2fuzzy sliding mode controller
圖8 切換增益估計(jì)值的變化曲線Fig.8 Change of switch gain estimation value
由圖4可以看出,AT-1FSMC相對(duì)AIT-2FSMC受到負(fù)載干擾后對(duì)系統(tǒng)影響較大,恢復(fù)時(shí)間較長(zhǎng),說(shuō)明AIT-2FSMC相對(duì)AT-1FSMC有較強(qiáng)的魯棒性;由圖5可知參數(shù)變化對(duì)系統(tǒng)影響較小,控制器的設(shè)計(jì)對(duì)系統(tǒng)精確數(shù)學(xué)模型的依賴程度??;由圖6可知AT-1FSMC的輸出iq存在抖振,數(shù)值在-0.02~0.07 A之間波動(dòng);而由圖7可知,AIT-2FSMC輸出幾乎無(wú)抖振,波動(dòng)大小在0.005 A以下,AIT-2FSMC可以更有效地削弱抖振.
針對(duì)永磁直線同步電機(jī)伺服系統(tǒng)易受參數(shù)變化、負(fù)載擾動(dòng)等不確定性擾動(dòng)的問(wèn)題,提出一種自適應(yīng)區(qū)間二型模糊滑??刂品椒?基于PMLSM數(shù)學(xué)模型設(shè)計(jì)積分滑模面,采用模糊規(guī)則數(shù)為7,單輸入單輸出的自適應(yīng)區(qū)間二型模糊系統(tǒng)逼近滑模控制的等效控制部分,并基于Lyapunov函數(shù)設(shè)計(jì)自適應(yīng)律保證系統(tǒng)的穩(wěn)定性.仿真結(jié)果表明,設(shè)計(jì)的控制器對(duì)系統(tǒng)參數(shù)變化和負(fù)載擾動(dòng)具有較強(qiáng)的魯棒性,且有效地削弱抖振.
[1] 李澤源,張文農(nóng),張奕黃,等.基于預(yù)估觀測(cè)器的二自由度速度控制器設(shè)計(jì) [J].控制與決策,2015,30(11):2067-2072.
(LI Ze-yuan,ZHANG Wen-nong,ZHANG Yi-huang,et al.Design of two-degree-of-freedom velocity controller based on predictive observer [J].Control and Decision,2015,30(11):2067-2072.)
[2] 孫宜標(biāo),閆峰,劉春芳.基于μ理論的永磁直線同步電機(jī)魯棒重復(fù)控制 [J].中國(guó)電機(jī)工程學(xué)報(bào),2009,29(30):52-57.
(SUN Yi-biao,YAN Feng,LIU Chun-fang.Robust repetitive control for permanent-magnet linear synchronous motor based on μ theory [J].Proceedings of the CSEE,2009,29(30):52-57.)
[3] 劉春芳,胡雨薇.磁懸浮平臺(tái)電磁懸浮系統(tǒng)的模糊滑??刂?[J].沈陽(yáng)工業(yè)大學(xué)學(xué)報(bào),2015,37(6):607-612.
(LIU Chun-fang,HU Yu-wei.Fuzzy sliding mode control of electromagnetic system for magnetic levitation platform [J].Journal of Shenyang University of Technology,2015,37(6):607-612.)
[4] 張碧陶,皮佑國(guó).永磁同步電機(jī)伺服系統(tǒng)模糊分?jǐn)?shù)階滑??刂?[J].控制與決策,2012,27(12):1776-1780.
(ZHANG Bi-tao,PI You-guo.Fractional order fuzzy sliding mode control for permanent magnet synchronous motor servo drive [J].Control and Decision,2012,27(12):1776-1780.)
[5] Soltanpour M R,Otadolajam P,Khooban M H.Robust control strategy for electrically driven robot manipulators:adaptive fuzzy sliding mode [J].Science Mea-surement Technology,2014,9(3):322-334.
[6] Wai R J,Lin C M,Hsu C F.Adaptive fuzzy sliding-mode control for electrical servo drive [J].Fuzzy Sets and Systems,2004,143(2):295-310.
[7] Mendel J M.Uncertain rule-based fuzzy logic systems:introduction and new directions [M].Upper-Saddle River:Prentice-Hall,2001.
[8] Sakalli A,Kumbasar T,Yesil E,et al.Analysis of the performances of type-1,self-tuning type-1 and interval type-2 fuzzy PID controllers on the magnetic levitation system [C]//IEEE International Conference on Fuzzy Systems.Beijing,China,2014:1859-1866.
[9] Lin T C,Chen C L.Uncertain nonlinear time delay systems fast and large disturbance rejection based on adaptive interval type-2 fuzzy PI control [C]//IEEE International Conference on Fuzzy Systems.Beijing,China,2014:647-653.
[10]Chen C S,Wang Y S.Robust inverse control for PMLSM drives using self-adaptive interval type-2 neural fuzzy network [C]//International Automatic Control Conference.Nantou,China,2013:383-388.
[11]Kumbasar T.Robust stability analysis and systematic design of single-input interval type-2 fuzzy logic controllers [J].IEEE Transactions on Fuzzy Systems,2016,24(3):675-694.
[12]Mendel J M,Liu X.Simplified interval type-2 fuzzy logic systems [J].IEEE Transactions on Fuzzy Systems,2013,21(6):1056-1069.
Adaptiveintervaltype-2fuzzyslidingmodecontrolforpermanentmagnetlinearmotor
SUN Yi-biao, WANG Ya-peng, LIU Chun-fang
(School of Electric Engineering, Shenyang University of Technology, Shenyang 110870, China)
Aiming at the problem that the performance of permanent magnet linear synchronous motor servo system is weakened due to the parameter variation and load disturbance of the system, an adaptive interval type-2 fuzzy sliding mode control method, where the end effect and the existence of friction were taken into consideration, was proposed. An adaptive interval type-2 logic system was adopted to approximate the equivalent control part of sliding mode control, and thus, the traditional type-1 fuzzy logic system was extended to the interval type-2 fuzzy logic system. In addition, the processing ability of the system was improved when the system was confronted with the parameter variation and uncertain disturbance. Based on the Lyapunov function, the adaptive law of switch item gain adjustment was designed to guarantee the stability of the system. The simulated results indicate that the proposed method can improve the robustness of the system and effectively weaken the chattering of the system.
permanent magnet linear synchronous motor; sliding mode control; interval type-2 fuzzy logic system; type-1 fuzzy logic system; adaptive law; Lyapunov function; robustness; chattering
2016-09-18.
國(guó)家自然科學(xué)基金資助項(xiàng)目(50805098).
孫宜標(biāo)(1970-),男,安徽巢湖人,副教授,博士,主要從事交流伺服系統(tǒng)、魯棒控制及非線性系統(tǒng)等方面的研究.
* 本文已于2017-08-01 12∶23在中國(guó)知網(wǎng)優(yōu)先數(shù)字出版. 網(wǎng)絡(luò)出版地址: http:∥www.cnki.net/kcms/detail/21.1189.T.20170801.1223.010.html
10.7688/j.issn.1000-1646.2017.06.01
TP 273
A
1000-1646(2017)06-0601-06
(責(zé)任編輯:景 勇 英文審校:尹淑英)