<html xmlns="http://www.w3.org/1999/xhtml"><head><meta charset="utf-8"><meta name="generator" content="pdf2htmlEX"><meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"><link rel="stylesheet" href="https://csdnimg.cn/release/download_crawler_static/css/base.min.css"><link rel="stylesheet" href="https://csdnimg.cn/release/download_crawler_static/css/fancy.min.css"><link rel="stylesheet" href="https://csdnimg.cn/release/download_crawler_static/10281785/raw.css"><script src="https://csdnimg.cn/release/download_crawler_static/js/compatibility.min.js"></script><script src="https://csdnimg.cn/release/download_crawler_static/js/pdf2htmlEX.min.js"></script><script>try{pdf2htmlEX.defaultViewer = new pdf2htmlEX.Viewer({});}catch(e){}</script><title></title></head><body><div id="sidebar" style="display: none"><div id="outline"></div></div><div id="pf1" class="pf w0 h0" data-page-no="1"><div class="pc pc1 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://csdnimg.cn/release/download_crawler_static/10281785/bg1.jpg"><div class="t m0 x1 h2 y1 ff1 fs0 fc0 sc0 ls0 ws0">华侨大学</div><div class="t m0 x2 h2 y2 ff1 fs0 fc0 sc0 ls0 ws0">硕士学位论文</div><div class="t m0 x3 h2 y3 ff1 fs0 fc0 sc0 ls0 ws0">参数模糊自整定PID控制器的研制</div><div class="t m0 x2 h2 y4 ff1 fs0 fc0 sc0 ls0 ws0">姓名：宁海峰</div><div class="t m0 x4 h2 y5 ff1 fs0 fc0 sc0 ls0 ws0">申请学位级别：硕士</div><div class="t m0 x5 h2 y6 ff1 fs0 fc0 sc0 ls0 ws0">专业：@</div><div class="t m0 x6 h2 y7 ff1 fs0 fc0 sc0 ls0 ws0">指导教师：王永初</div><div class="t m0 x1 h2 y8 ff1 fs0 fc0 sc0 ls0 ws0">20060601</div></div><div class="pi" data-data='{"ctm":[1.611639,0.000000,0.000000,1.611639,0.000000,0.000000]}'></div></div></body></html>
<div id="pf2" class="pf w0 h0" data-page-no="2"><div class="pc pc2 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://csdnimg.cn/release/download_crawler_static/10281785/bg2.jpg"><div class="t m0 x7 h3 y9 ff2 fs1 fc0 sc1 ls1 ws0">论 文 摘 要 </div><div class="t m0 x8 h3 ya ff2 fs1 fc0 sc1 ls0 ws0"> </div><div class="t m0 x9 h4 yb ff2 fs2 fc0 sc0 ls0 ws0">在实际的工业控制中，PID<span class="_ _0"> </span>控制依然是最主要的一种控制方式，所以如何简单、</div><div class="t m0 x8 h4 yc ff2 fs2 fc0 sc0 ls0 ws0">有效的实现<span class="_ _0"> </span>PID<span class="_ _1"> </span>参数的整定有着重要的工业意义；<span class="_ _2"></span><span class="ls2">对于工业过程中常出现的非线性、</span></div><div class="t m0 x8 h4 yd ff2 fs2 fc0 sc0 ls0 ws0">时变的系统来讲，如何实现<span class="_ _0"> </span>PID<span class="_ _1"> </span>参数的自整定显得犹为重要。 </div><div class="t m0 x9 h4 ye ff2 fs2 fc0 sc0 ls3 ws0">PID<span class="_"> </span>的参数整定多依赖于人工经验和具体<span class="ls4">的对象模型，整定结果往往不能令人</span></div><div class="t m0 x8 h4 yf ff2 fs2 fc0 sc0 ls0 ws0">满意。另外，即使<span class="_ _3"> </span>PID<span class="_ _3"> </span>参数调整的很好，当控制对象参数变化后，系统的性能必然</div><div class="t m0 x8 h4 y10 ff2 fs2 fc0 sc0 ls5 ws0">也会受到影响。模糊控制作为智能控制一个新<span class="ls6">型领域，已经广泛应用到工业生产过</span></div><div class="t m0 x8 h4 y11 ff2 fs2 fc0 sc0 ls5 ws0">程中，模糊控制事先不需要获知对象的精确数<span class="ls6">学模型，而是基于人类的思维以及生</span></div><div class="t m0 x8 h4 y12 ff2 fs2 fc0 sc0 ls5 ws0">产经验，用语言规则描述控制过程，并根据规<span class="ls6">则去调整控制算法或控制参数。本论</span></div><div class="t m0 x8 h4 y13 ff2 fs2 fc0 sc0 ls2 ws0">文分析了工程中常用的<span class="_ _4"> </span>PID<span class="_ _4"> </span>参数整定方法，通过研究将模糊控制与<span class="_ _4"> </span>PID<span class="_ _4"> </span>不同的结合</div><div class="t m0 x8 h4 y14 ff2 fs2 fc0 sc0 ls5 ws0">方法，实现了无须精确确定对象模型，只须将<span class="ls6">操作人员和专家长期实践积累的经验</span></div><div class="t m0 x8 h4 y15 ff2 fs2 fc0 sc0 ls5 ws0">知识用控制规则模型化，然后用模糊推理在线<span class="ls6">辨识对象特征参数，实时改变控制策</span></div><div class="t m0 x8 h4 y16 ff2 fs2 fc0 sc0 ls0 ws0">略，便可对<span class="_ _0"> </span>PID<span class="_ _1"> </span>参数实现最佳调整。 </div><div class="t m0 x9 h4 y17 ff2 fs2 fc0 sc0 ls0 ws0">在研究模糊技术与<span class="_ _3"> </span>PID<span class="_ _3"> </span>控制相结合的基础上，提出了微分剥离法和将不完全微</div><div class="t m0 x8 h4 y18 ff2 fs2 fc0 sc0 ls0 ws0">分引入法这两种参数模糊自整定<span class="_ _4"> </span>PID<span class="_ _4"> </span>控制器，并设计出一种参数模糊自整定<span class="_ _4"> </span>PID<span class="_ _4"> </span>控</div><div class="t m0 x8 h4 y19 ff2 fs2 fc0 sc0 ls0 ws0">制器。该控制器硬件系统选用<span class="_ _3"> </span>MCS-51<span class="_ _3"> </span>系列单片机，完成数据采集、参数整定、LED</div><div class="t m0 x8 h4 y1a ff2 fs2 fc0 sc0 ls0 ws0">显示、声光报警等功能，软件采用单片机常用的编程语言<span class="_ _3"> </span>C51<span class="_ _3"> </span>编制。论文中结合典</div><div class="t m0 x8 h4 y1b ff2 fs2 fc0 sc0 ls5 ws0">型工业控制过程的例子对不同的设计结构进行<span class="ls6">了仿真试验。仿真结果表明，对在实</span></div><div class="t m0 x8 h4 y1c ff2 fs2 fc0 sc0 ls0 ws0">际复杂工业对象中存在的非线性、时变性等特点，所设计的模糊自整定<span class="_ _3"> </span>PID<span class="_ _3"> </span>控制策</div><div class="t m0 x8 h4 y1d ff2 fs2 fc0 sc0 ls5 ws0">略具有较好的控制效果和品质，具有一定的自<span class="ls6">适应能力，能够对对象变化以及外来</span></div><div class="t m0 x8 h4 y1e ff2 fs2 fc0 sc0 ls0 ws0">的扰动做出及时调整，保证整个系统的平稳运行。 </div><div class="t m0 x9 h4 y1f ff2 fs2 fc0 sc0 ls0 ws0"> </div><div class="t m0 x9 h4 y20 ff2 fs2 fc0 sc0 ls0 ws0"> </div><div class="t m0 x8 h5 y21 ff2 fs3 fc0 sc1 ls7 ws0">关键词：<span class="fs2 sc0 ls0">自整定<span class="_ _0"> </span>PID<span class="_ _1"> </span>控制器 模糊控制 单片机 Matlab<span class="_ _0"> </span>仿真 </span></div><div class="t m0 x8 h4 y22 ff2 fs2 fc0 sc0 ls0 ws0"> </div><div class="t m0 x8 h4 y23 ff2 fs2 fc0 sc0 ls0 ws0"> </div><div class="t m0 x8 h4 y24 ff2 fs2 fc0 sc0 ls0 ws0"> </div></div><div class="pi" data-data='{"ctm":[1.611639,0.000000,0.000000,1.611639,0.000000,0.000000]}'></div></div>
<div id="pf3" class="pf w0 h0" data-page-no="3"><div class="pc pc3 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://csdnimg.cn/release/download_crawler_static/10281785/bg3.jpg"><div class="t m0 x8 h6 y9 ff3 fs2 fc0 sc0 ls8 ws0"> <span class="ff4 fs1 ls0">ABSTRACT </span></div><div class="t m0 x9 h7 y25 ff4 fs2 fc0 sc0 ls0 ws0"> </div><div class="t m0 xa h8 y26 ff3 fs2 fc0 sc0 ls9 ws1">In the real industry process,<span class="ls2 ws2"> PID Control is still one of<span class="lsa"> the important ways. So how </span></span></div><div class="t m0 x8 h8 y27 ff3 fs2 fc0 sc0 lsb ws3">to find a simple and ef<span class="_ _5"></span>fective way to achieve adjustment is meaningful. T<span class="_ _6"></span>o those </div><div class="t m0 x8 h8 y28 ff3 fs2 fc0 sc0 lsc ws4">nonlinearity and time-varying sys<span class="_ _5"></span>tem, it is significant to achieve PID param<span class="_ _5"></span>eters </div><div class="t m0 x8 h8 y29 ff3 fs2 fc0 sc0 lsd ws0">auto-tuning. </div><div class="t m0 x9 h8 y2a ff3 fs2 fc0 sc0 lsc ws5">The parameters’<span class="_ _6"></span> adjustment of the PID m<span class="_ _5"></span>a<span class="lsa ws6">inly depends on the human’<span class="_ _7"></span>s experience </span></div><div class="t m0 x8 h8 y2b ff3 fs2 fc0 sc0 lsa ws7">and the object model, the result is norma<span class="_ _5"></span><span class="ls9 ws8">lly dissatisfying. Fuzzy control which was </span></div><div class="t m0 x8 h8 y2c ff3 fs2 fc0 sc0 lsb ws9">outstanding in the field of the intelligent cont<span class="lse">roller has already applied widely in industry </span></div><div class="t m0 x8 h8 y2d ff3 fs2 fc0 sc0 lse wsa">production process. Fuzzy control do not need<span class="lsc wsb"> to know accurate mathem<span class="_ _5"></span>atic model of </span></div><div class="t m0 x8 h8 y2e ff3 fs2 fc0 sc0 lsf wsc">object. But it is based on human thinking <span class="ls10 wsd">and producing experience; and it applies </span></div><div class="t m0 x8 h8 y2f ff3 fs2 fc0 sc0 ls11 wse">language rule to describe controlling proce<span class="wsf">ss and bases rule to m<span class="_ _5"></span>odify the controlling </span></div><div class="t m0 x8 h8 y30 ff3 fs2 fc0 sc0 ls12 ws10">arithmetic and parameter<span class="_ _6"></span>.<span class="ff5 ls0 ws0"> </span><span class="ls10 ws11">The paper studies the methods <span class="lsd ws12">of combing fuzzy control and </span></span></div><div class="t m0 x8 h8 y31 ff3 fs2 fc0 sc0 ls11 ws13">PID control, achieves the best parameter tu<span class="lse ws14">ning which only use the experience knowledge </span></div><div class="t m0 x8 h8 y32 ff3 fs2 fc0 sc0 lse ws15">of manipulator and fuzzy contro<span class="ws16">l rule of expert, and use the fuzzy reasoning iden<span class="_ _5"></span>tifying </span></div><div class="t m0 x8 h8 y33 ff3 fs2 fc0 sc0 ls13 ws17">the characteristic param<span class="_ _5"></span>eter of object on-line,<span class="_ _5"></span><span class="ls10 ws18"> changing control rule in time, needn’t the </span></div><div class="t m0 x8 h8 y34 ff3 fs2 fc0 sc0 lsc ws19">identify the model. </div><div class="t m0 x9 h8 y35 ff3 fs2 fc0 sc0 ls14 ws1a">On the basis of this, work out a paramete<span class="lsa ws1b">rs’<span class="_ _6"></span> fuzzy auto-tuning PID controller<span class="_ _7"></span>. The </span></div><div class="t m0 x8 h8 y36 ff3 fs2 fc0 sc0 ls10 ws1c">firmware of the product based on the MCS-51 series’<span class="_ _6"></span> single chip mi<span class="lse ws1d">croprocessor<span class="_ _7"></span>. It can </span></div><div class="t m0 x8 h8 y37 ff3 fs2 fc0 sc0 ls11 ws1e">accomplish data sampling, param<span class="_ _5"></span>eters adjustment, LED display<span class="_ _6"></span>, sound and light alarming </div><div class="t m0 x8 h8 y38 ff3 fs2 fc0 sc0 lsb ws1f">etc. The C51 language is used in the product’<span class="_ _7"></span>s software designing. By the example of </div><div class="t m0 x8 h8 y39 ff3 fs2 fc0 sc0 lsd ws20">industry process, simulation prove it have better ef<span class="_ _7"></span>fect and quality<span class="_ _7"></span>. It also has strong </div><div class="t m0 x8 h8 y3a ff3 fs2 fc0 sc0 lsb ws21">self-adaptation to the complex nonlinear<span class="_ _7"></span>, time-vari<span class="lse ws22">ety object. It can regulate the system in </span></div><div class="t m0 x8 h8 y3b ff3 fs2 fc0 sc0 ls12 ws23">time when the object changes or something interfere the system<span class="_ _5"></span>. <span class="_"> </span> <span class="_"> </span> <span class="_"> </span> <span class="_"> </span> <span class="_"> </span> <span class="_"> </span> <span class="_"> </span> <span class="_"> </span> <span class="_"> </span> </div><div class="t m0 xa h8 y3c ff3 fs2 fc0 sc0 ls0 ws0"> </div><div class="t m0 xa h8 y3d ff3 fs2 fc0 sc0 ls0 ws0"> </div><div class="t m0 x8 h9 y3e ff4 fs3 fc0 sc0 ls15 ws0">Keywords:<span class="ff3 fs2 lsf ws24"> <span class="_"> </span> <span class="_"> </span>Auto-tuning PID controller<span class="_ _7"></span>, Fuzzy Control, Single chip microprocessor<span class="_ _7"></span>, </span></div><div class="t m0 xb h9 y3f ff4 fs3 fc0 sc0 ls0 ws0"> <span class="_"> </span><span class="ff3 fs2 ls16 ws25">Matlab simulation</span><span class="ff3 fs4"> </span></div></div><div class="pi" data-data='{"ctm":[1.611639,0.000000,0.000000,1.611639,0.000000,0.000000]}'></div></div>
<div id="pf4" class="pf w0 h0" data-page-no="4"><div class="pc pc4 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://csdnimg.cn/release/download_crawler_static/10281785/bg4.jpg"><div class="t m0 x8 ha y40 ff6 fs4 fc0 sc0 ls0 ws0">华侨大学硕士学位论文<span class="ls17"> </span></div><div class="t m0 x8 hb y41 ff7 fs5 fc0 sc0 ls18 ws0"> 1</div><div class="t m0 xc h6 y9 ff8 fs1 fc0 sc1 ls19 ws0">第一章<span class="ff9 sc0 ls0"> <span class="_"> </span></span>绪论<span class="ff9 sc0 ls0"> </span></div><div class="t m0 x8 h9 y42 ff9 fs3 fc0 sc0 ls1a ws0">1.1 <span class="_"> </span><span class="ff8 sc1 ls7">引言</span><span class="ls0"> </span></div><div class="t m0 xd h4 yb ff7 fs2 fc0 sc0 lsa ws0">PID<span class="_ _8"> </span><span class="ff8 ls1b">控制器是在工业过程控制中最常见的一种控制调节器，广泛应用于化工、机</span></div><div class="t m0 x8 h4 yc ff8 fs2 fc0 sc0 ls2 ws0">械、<span class="_ _9"></span>冶金和轻工等工业过程控制系统中。<span class="_ _9"></span>有一些文献陈述了当前的应用状况</div><div class="t m0 xe hc y43 ff7 fs6 fc0 sc0 ls1c ws0">[1,2]</div><div class="t m0 xf h4 yc ff8 fs2 fc0 sc0 ls1d ws0">。日<span class="_ _a"> </span>本</div><div class="t m0 x8 h4 yd ff8 fs2 fc0 sc0 ls0 ws0">电子测量仪表协会在<span class="_ _3"> </span><span class="ff7">1989<span class="_ _4"> </span></span>年对过程控制做的调查报告，该报告表明<span class="_ _3"> </span><span class="ff7">90%</span><span class="ls1e">以上的控制</span></div><div class="t m0 x8 h4 ye ff8 fs2 fc0 sc0 ls0 ws0">回路具有<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>结构。<span class="_ _b"></span>另外一篇有关加拿大造纸厂的统计报告表明典型的造纸厂一般有</div><div class="t m0 x8 h4 yf ff7 fs2 fc0 sc0 ls0 ws0">2000<span class="_ _4"> </span><span class="ff8">多个控制回路，其中<span class="_ _4"> </span></span>97%<span class="ff8">以上是<span class="_ _c"> </span></span><span class="lsa">PID<span class="_ _4"> </span></span><span class="ff8">控制，而且仅仅有<span class="_ _4"> </span></span>20%<span class="ff8">的控制回路工作比</span></div><div class="t m0 x8 h4 y10 ff8 fs2 fc0 sc0 ls0 ws0">较满意。控制回路性能普遍差的原因中参数整定不合适的占<span class="_ _0"> </span><span class="ff7">30%<span class="_ _d"></span></span>，阀门问题占<span class="_ _0"> </span><span class="ff7">30%<span class="_ _d"></span></span>，</div><div class="t m0 x8 h4 y11 ff8 fs2 fc0 sc0 ls0 ws0">而另外<span class="_ _0"> </span><span class="ff7">20%<span class="_ _d"></span></span>的控制器性能差有多种原因，<span class="_ _e"></span>如传感器问题、<span class="_ _e"></span>采样频率的选择不当以及滤</div><div class="t m0 x8 h4 y12 ff8 fs2 fc0 sc0 ls0 ws0">波器的问题等。<span class="_ _5"></span>在已安装的过程控制器中<span class="_ _0"> </span><span class="ff7">30%</span>是处于手动状态，<span class="ff7">20%</span>的控制回路采用</div><div class="t m0 x8 h4 y13 ff8 fs2 fc0 sc0 ls0 ws0">厂家整定的参数，即控制器制造商预先设定的参数值，<span class="ff7">30%</span><span class="ls1f">的控制回路由于阀门和传</span></div><div class="t m0 x8 h4 y14 ff8 fs2 fc0 sc0 ls0 ws0">感器的问题导致控制性能较差。<span class="ff7"> </span></div><div class="t m0 xd h4 y15 ff8 fs2 fc0 sc0 ls0 ws0">因此，<span class="ff7 ls20">PID<span class="_ _8"> </span></span><span class="ls1b">控制器虽然在工业过程控制中普遍应用，但是获得的控制效果并不十</span></div><div class="t m0 x8 h4 y16 ff8 fs2 fc0 sc0 ls0 ws0">分理想。<span class="_ _b"></span>同时由于<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制器特别适用于过程的动态性能是良性的而且控制性能要求</div><div class="t m0 x8 h4 y17 ff8 fs2 fc0 sc0 ls0 ws0">不高的情况，<span class="_ _7"></span>但随着现代工业的发展，<span class="_ _7"></span>人们面临的被控对象越来越复杂，<span class="_ _5"></span>对于控制系</div><div class="t m0 x8 h4 y18 ff8 fs2 fc0 sc0 ls0 ws0">统的精度性能和可靠性的要求越来越高，<span class="_ _9"></span>这对<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制技术提出了严峻的挑战。<span class="_ _9"></span>只有</div><div class="t m0 x8 h4 y19 ff8 fs2 fc0 sc0 ls0 ws0">和先进控制策略相结合，<span class="_ _6"></span>才能保证<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制技术永不过时，<span class="_ _e"></span>而它也正是向高精度、<span class="_ _6"></span>高</div><div class="t m0 x8 h4 y1a ff8 fs2 fc0 sc0 ls0 ws0">性能、智能化的方向在逐步发展。<span class="ff7"> </span></div><div class="t m0 x8 h9 y44 ff9 fs3 fc0 sc0 ls21 ws26">1.2 PID<span class="_"> </span><span class="ff8 sc1 ls7 ws0">控制器的历史及存在问题</span><span class="ls0 ws0"> </span></div><div class="t m0 x8 h4 y45 ff7 fs2 fc0 sc0 lsc ws0">1) <span class="_ _f"> </span><span class="ff8 ls0">控制器发展经历了三个阶段：<span class="ff7"> </span></span></div><div class="t m0 x8 h4 y46 ffa fs2 fc0 sc0 ls22 ws0"> <span class="_ _10"></span><span class="ff8 ls0">十七世纪中叶由于机器工业的发展，<span class="_ _6"></span>对控制提出了要求。<span class="_ _7"></span>反馈的方法首先被提出，</span></div><div class="t m0 x10 h4 y47 ff8 fs2 fc0 sc0 ls23 ws0">在研究气动和电动记录仪的基础上发现了比例和积分作用，它们的主要的调节对</div><div class="t m0 x10 h4 y48 ff8 fs2 fc0 sc0 ls0 ws0">象是火炉的温度和蒸汽机的阀门位置等。调节方式类似于<span class="_ _11"> </span><span class="ff7 ls24">Bang-Bang<span class="_ _11"> </span></span><span class="ls1e">继电控制，</span></div><div class="t m0 x10 h4 y49 ff8 fs2 fc0 sc0 ls0 ws0">精度比较低控制器的形式是<span class="_ _0"> </span><span class="ff7">P<span class="_"> </span></span>和<span class="_ _1"> </span><span class="ff7 ls20">Pi</span>。<span class="ff7"> </span></div><div class="t m0 x8 h4 y4a ffa fs2 fc0 sc0 ls22 ws0"> <span class="_ _10"></span><span class="ff8 ls0">上个世纪<span class="_ _c"> </span><span class="ff7">20<span class="_ _c"> </span></span>年代至<span class="_ _c"> </span><span class="ff7 ls25">40<span class="_ _c"> </span></span>年代，泰勒仪器公司的发现了微分作用，微分作用的发现</span></div><div class="t m0 x10 h4 y22 ff8 fs2 fc0 sc0 ls23 ws0">具有重要的意义，它能直观地实现对慢系统的控制，对该系统的动态性能能够进</div><div class="t m0 x10 h4 y23 ff8 fs2 fc0 sc0 ls0 ws0">行调节，与先期提出的比例和积分作用成为主要的调节部件。<span class="ff7"> </span></div><div class="t m0 x8 h4 y24 ffa fs2 fc0 sc0 ls22 ws0"> <span class="_ _10"></span><span class="ff7 ls0">1942<span class="_"> </span><span class="ff8">年和<span class="_ _0"> </span></span>1943<span class="_"> </span><span class="ff8">年，<span class="_ _10"></span>泰勒仪器公司的<span class="_ _1"> </span><span class="ff7 ls14">Zielger<span class="_"> </span></span>和<span class="_ _0"> </span><span class="ff7 lsf">Nichols<span class="_"> </span></span>等人分别在开环和闭环的情</span></span></div></div><div class="pi" data-data='{"ctm":[1.611639,0.000000,0.000000,1.611639,0.000000,0.000000]}'></div></div>
<div id="pf5" class="pf w0 h0" data-page-no="5"><div class="pc pc5 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://csdnimg.cn/release/download_crawler_static/10281785/bg5.jpg"><div class="t m0 x8 ha y40 ff6 fs4 fc0 sc0 ls0 ws0">华侨大学硕士学位论文<span class="ls17"> </span></div><div class="t m0 x8 hb y41 ff7 fs5 fc0 sc0 ls18 ws0"> 2</div><div class="t m0 x10 h4 y4b ff8 fs2 fc0 sc0 ls23 ws0">况下，用实验的方法分别研究了比例、积分和微分这三部分在控制中的作用，首</div><div class="t m0 x10 h4 y4c ff8 fs2 fc0 sc0 ls0 ws0">次提出了<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制器参数整定的问题</div><div class="t m0 x11 hc y4d ff7 fs6 fc0 sc0 ls1c ws0">[3,4]</div><div class="t m0 x12 h4 y4c ff8 fs2 fc0 sc0 ls0 ws0">。<span class="ff7"> </span></div><div class="t m0 x8 h4 y4e ff7 fs2 fc0 sc0 lsc ws0">2) <span class="_ _f"> </span><span class="ff8 ls0">控制器存在的主要问题：<span class="ff7"> </span></span></div><div class="t m0 xd h4 y4f ff7 fs2 fc0 sc0 lsa ws0">PID<span class="_ _8"> </span><span class="ff8 ls1b">控制器的结构简单，容易被理解和实现，应用中不需要精确的系统模型的预</span></div><div class="t m0 x8 h4 y50 ff8 fs2 fc0 sc0 ls0 ws0">先知识，<span class="_ _12"></span>因而<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制器成为应用最广泛的控制器。<span class="_ _12"></span>但是人们对<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制器的认识和</div><div class="t m0 x8 h4 y51 ff8 fs2 fc0 sc0 ls0 ws0">改进远没有完成，<span class="_ _6"></span>到目前为止<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制的机理、<span class="_ _e"></span>使用范围、<span class="_ _6"></span>鲁棒性等问题还没有彻底</div><div class="t m0 x8 h4 y52 ff8 fs2 fc0 sc0 ls0 ws0">全面的分析研究。事实上，<span class="ff7 lsa">PID<span class="_"> </span></span>控制器并非万能的，它存在其固有的缺点</div><div class="t m0 x13 hc y53 ff7 fs6 fc0 sc0 ls1c ws0">[5,6]</div><div class="t m0 x14 h4 y52 ff8 fs2 fc0 sc0 ls0 ws0">：<span class="ff7"> </span></div><div class="t m0 x8 h4 y54 ffa fs2 fc0 sc0 ls22 ws0"> <span class="_ _10"></span><span class="ff7 lsa">PID<span class="_ _8"> </span><span class="ff8 ls26">对系统基本线性和动态特性不随时间变化的系统能较好的控制，而很多工业</span></span></div><div class="t m0 x10 h4 y55 ff8 fs2 fc0 sc0 ls0 ws0">过程是非线性或时变的。<span class="ff7"> </span></div><div class="t m0 x8 h4 y56 ffa fs2 fc0 sc0 ls22 ws0"> <span class="_ _10"></span><span class="ff7 lsa">PID<span class="_ _8"> </span><span class="ff8 ls26">参数必须根据过程的动态特性整定的很好。如果过程的动态特性变化，例如</span></span></div><div class="t m0 x10 h4 y57 ff8 fs2 fc0 sc0 ls27 ws0">可能由负载的变化引起<span class="ls4">系统动态特性变化，<span class="ff7 ls20">PID<span class="_ _8"> </span></span><span class="ls28">参数要重新整定。实际应用中，</span></span></div><div class="t m0 x10 h4 y58 ff7 fs2 fc0 sc0 lsa ws0">PID<span class="_"> </span><span class="ff8 ls0">参数的整定很困难。<span class="ff7"> </span></span></div><div class="t m0 x8 h4 y59 ffa fs2 fc0 sc0 ls22 ws0"> <span class="_ _10"></span><span class="ff7 lsa">PID<span class="_"> </span><span class="ff8 ls0">在控制强耦合及结构不确定的复杂过程时总显得无能为力。<span class="ff7"> </span></span></span></div><div class="t m0 x10 h4 y5a ff7 fs2 fc0 sc0 lsa ws0">PID<span class="_ _3"> </span><span class="ff8 ls0">参数自整定技术是为了处理<span class="_ _11"> </span></span>PID<span class="_ _3"> </span><span class="ff8 ls0">参数整定这个问题而产生的，现在自动整定</span></div><div class="t m0 x8 h4 y5b ff8 fs2 fc0 sc0 ls0 ws0">的<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制器己是商业单回路控制器和分布控制系统的一个标准。<span class="_ _e"></span><span class="ff7 lsa">PID<span class="_"> </span><span class="ff8 ls0">参数整定与自</span></span></div><div class="t m0 x8 h4 y5c ff8 fs2 fc0 sc0 ls0 ws0">整定的方法很多，<span class="_ _6"></span>但往往难以实现或不很理想，<span class="_ _7"></span>在精度与速度的折衷及对象的使用范</div><div class="t m0 x8 h4 y5d ff8 fs2 fc0 sc0 ls0 ws0">围上常常难以令人满意。<span class="_ _9"></span>因此，<span class="_ _9"></span>在<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参数的整定及自整定技术方面还有待进一步的</div><div class="t m0 x8 h4 y5e ff8 fs2 fc0 sc0 ls0 ws0">深入研究。<span class="ff7"> </span></div><div class="t m0 x8 h9 y5f ff9 fs3 fc0 sc0 ls21 ws26">1.3 PID<span class="_"> </span><span class="ff8 sc1 ls7 ws0">控制器的发展</span><span class="ls0 ws0"> </span></div><div class="t m0 x10 h4 y60 ff8 fs2 fc0 sc0 ls0 ws0">自<span class="_ _0"> </span><span class="ff7 lsa">Ziegler-Nichols<span class="_"> </span></span>在<span class="_ _0"> </span><span class="ff7 ls29">1942<span class="_"> </span></span>年提出<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span><span class="ls2">参数整定方法以来，<span class="_ _6"></span>许多技术己经被应用于</span></div><div class="t m0 x8 h4 y61 ff7 fs2 fc0 sc0 lsa ws0">PID<span class="_"> </span><span class="ff8 ls0">控制器的手动和自动整定中。<span class="_ _7"></span>根据研究方法的划分，<span class="_ _7"></span>可分为基于频域的<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参数</span></div><div class="t m0 x8 h4 y62 ff8 fs2 fc0 sc0 ls0 ws0">整定方法和基于时域的<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参数整定方法；<span class="_ _12"></span>根据发展阶段的划分，<span class="_ _12"></span>可分为常规<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参</div><div class="t m0 x8 h4 y63 ff8 fs2 fc0 sc0 ls0 ws0">数整定方法和智能<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参数整定方法；<span class="_ _7"></span>按照被控对象个数来划分，<span class="_ _7"></span>可分为单变量<span class="_ _1"> </span><span class="ff7 lsa">PID</span></div><div class="t m0 x8 h4 y64 ff8 fs2 fc0 sc0 ls0 ws0">参数整定方法和多变量<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参数整定方法，<span class="_ _e"></span>前者包括现有大多数整定方法，<span class="_ _9"></span>后者是最</div><div class="t m0 x8 h4 y65 ff8 fs2 fc0 sc0 ls0 ws0">近研究的热点及难点；<span class="_ _9"></span>按控制量的组合形式来划分，<span class="_ _9"></span>可分为线性<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参数整定方法和</div><div class="t m0 x8 h4 y66 ff8 fs2 fc0 sc0 ls0 ws0">非线性<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>参数整定方法，<span class="_ _12"></span>前者适用于经典<span class="_ _1"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>调节器，<span class="_ _12"></span>后者适用于由非线性跟踪微</div><div class="t m0 x8 h4 y67 ff8 fs2 fc0 sc0 ls0 ws0">分器和非线性组合方式生成的非线性<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制器。<span class="ff7"> </span></div><div class="t m0 x15 h4 y68 ff8 fs2 fc0 sc0 ls0 ws0">近年来，<span class="_ _9"></span>随着人工智能技术的发展，<span class="_ _9"></span>智能控制得到迅速发展，<span class="_ _9"></span>而将智能技术和常</div><div class="t m0 x8 h4 y69 ff8 fs2 fc0 sc0 ls0 ws0">规<span class="_ _0"> </span><span class="ff7 lsa">PID<span class="_"> </span></span>控制方法有机的融合在一起，<span class="_ _b"></span>利用人工智能的方法将操作人员的调整经验作为</div></div><div class="pi" data-data='{"ctm":[1.611639,0.000000,0.000000,1.611639,0.000000,0.000000]}'></div></div>
