【正文】
. 7 20 . 8 40 . 9 20 . 9 80 . 1 40 . 30 . 4 40 . 5 80 . 7 20 . 8 40 . 9 20 . 9 85101520250 . 1 40 . 30 . 4 40 . 5 8R o o t L o c u sR e a l A x i sImaginary Axis ( 3)作出單位階躍輸入下的系統(tǒng)響應,分析系統(tǒng)單位階躍響應的性能指標。 程序如下: num0=[143 10]。 den0=[ 165 10]。 [y x t]=step(num0,den0)。t1=length(t)。yss=y(t1)。[ym,tm]=max(y)。 singma=100*(ymyss)/yss %計算超調量 singma = n=1。while y(n)*yss。n=n+1。end %計算上升時間 m=1。while y(m)*yss。m=m+1。end risetime=t(m)t(n) risetime = 38 while y(t1)amp。y(t1)*yss。t1=t11。end %計算調節(jié)時間 stime=t(t1) stime = plot(t,y) grid on 0 5 10 15 20 25 30 35 4000 . 20 . 40 . 60 . 811 . 21 . 4 由圖形可以看到系統(tǒng)是穩(wěn)定的。 上升時間 Tr= ,超調量 ?%=, Ts =。 ( 4)繪出系統(tǒng)開環(huán)傳函的 bode 圖,利用頻域分析方法分析系統(tǒng)的頻域性能指標(相角裕度和幅值裕度,開環(huán)振幅)。 39 程序如下: num=[143 10]。den=[ 165 1 0]。 bode(tf(num,den))。grid。[gm,pm,pf,gf]=margin(num,den) gm = pm = pf = gf = 1 5 0 1 0 0 5 0050100150Magnitude (dB)104103102101100101102103 2 7 0 2 2 5 1 8 0 1 3 5 9 0Phase (deg)B o d e D i a g r a mF r e q u e n c y ( r a d / s e c ) 從而我們知道幅值穩(wěn)定裕度 gm=(若化為對數則為 ),相角穩(wěn)定裕度 pm=, 相角 穿越頻率 pf = ,幅值 穿越頻率 gf =。由 幅值穩(wěn)定裕度為 10dB,相角穩(wěn)定裕度為500。 可知系統(tǒng)是穩(wěn)定的, 且滿足題目中的要求,設計結束。 5. 2Multisim電路設計仿真 方式 使用 matlab 軟件中的 Simulink 仿真:其單位階躍相應如下 40 對應的階躍響應圖如下: 校正后閉環(huán)傳遞函數可以看成一個積分環(huán)節(jié)與 四 個慣性環(huán)節(jié)及 兩個微分環(huán)節(jié)。 模擬電路如下:該圖是在 pretel 99se 中繪制完成的。 41 42 第 6 章 設計總結 本章 主要是講一下本次設計的小結。 6. 1 總結 上述的設計任務是:有一未校正系統(tǒng), 開環(huán)傳遞函數為 )2)(1()( ??? sss KsG ,要求 Kv 為 10/s,相位裕量為 500,增益裕量大于等于 10 分貝, 試 設計一個校正裝置。 根據設計任務和設計要求 本人 從多方面查找資料和學習相關的知識,在查找資料,學習相關知識和設計過程可分以下幾點: ( 1) 根據 設計任務和要求并學習教科書中第 五章 《線性系統(tǒng)的 頻域分析》和第六章《線性系統(tǒng)的校正方法》的 內容; ( 2) 在課本的理論知識的基礎上,學習 Matlab 軟件,主要是自動控制在 Matlab 軟件中的運用部分,以及學習電路仿真軟件( Multisim 軟件)。 ( 3) 通過上述軟件描繪幅相特性,根軌跡,波特圖等,進行分析,并通過 Matlab 軟件中的 Simulink 動態(tài)仿真工具進行仿真; ( 4) 途中遇到問題, 積極向老師請教和 討論該如何進行設計。 6. 2 心得 剛開始接到這題目時幾乎無從下手,這主要是因為 自己選的設計課 題 和其他同學的一點 很 不相似,沒有共同的要求,很難和其他同學一起溝通。不過,經過自己的 努力,從多方面查找資料,和向老師請教以及復習課本中的相關知識和學習 Matlab 和 Multisim 軟件,終于順利完成。與平時所做的實驗都是按照實驗指導書的說明很機械的完成相比,這次 畢業(yè)設計給了我很大的思考空間,在設計過程訓練了我 的自學能力,并也開始學著在給定任務情況下該何如查找資料,何如在設計過程的時間內能更好地分配所要學習的內容服務于設計的需要,而不會沒有主次之分。 43 整個設計過程從學習相關知識到寫報告每一個步驟都在經過 自己的 思考。我發(fā)現許多相關的知識在書上都能找到,因此只要根據課本設計出來是沒多大困難的。 44 附英文文獻: 1. Some Open Problems in Matrix Theory Arising in Linear Systems and Control ABSTRACT Control theory has long provided a rich source of motivation for developments in matrix theory. Accordingly, we discuss some open problems in matrix theory arising from theoretical and practical issues in linear systems theory and feedback control. The problems discussed include robust stability, matrix exponentamp。 induced norms, stabilizability and pole assignability, and nonstandard matrix equations. A substantial number of references are included to acquaint matrix theorists with problems and trends in this application area. 1. INTRODUCTION Feedback control theory has long provided a rich source of motivation for developments in matrix theory. The purpose of this paper is to discuss several open problems in matrix theory that arise from theoretical and practical issues in feedback control theory and the associated area of linear systems theory. Many of these problems are remarkably simple to state, are of intense interest in control theory and applications, and yet remain unsolved. Besides leading to the resolution of these problems, it is hoped that this paper will help to stimulate increased interaction between matrix and control theorists. Accordingly, the paper includes some brief tutorial discussions and provides motivation for these problems. The problems we discuss are divided into five topics, 45 namely, robust stability, matrix exponentials, induced norms, stabilizability and pole assignability, and nonstandard matrix equations. It is important to note that these problems are not my own, but have originated in a variety of control and matrixtheory applications and are due to a multitude of researchers. 2. ROBUST STABILITY A fundamental problem in the analysis of linear systems is the following: Given a collection of matrices ,determine a subset such that if every element of is stable (that is, each of its eigenvalues has negative real part), then every element of J is also stable. This problem arises when the modeling data are uncertain and guarantees of stability are desired. A related problem involves a set of polynomials rather than a set of matrices. Consider, for example, the set of polynomials: where, for i = 0, ? , n 1, the lower and upper coefficient bounds are given. In this case the rather remarkable result of Kharitonov states that every element of is stable if every element of is stable, where is the subset of consisting of the following four polynomials: 46 where the 4cyclic pattern of the coefficients is repeated for successively decreasing powers of s. Thus, to determine whether every polynomial in is stable, it suffices to check only these four polynomials. Kharitonov’s result has generated considerable interest an