• wismail
    了解作者
  • PDF
    开发工具
  • 767KB
    文件大小
  • rar
    文件格式
  • 0
    收藏次数
  • 1 积分
    下载积分
  • 0
    下载次数
  • 2018-11-13 23:39
    上传日期
Evaluation of Fuzzy logic
05584571.rar
  • 05584571.pdf
    824.2KB
内容介绍
<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://static.pudn.com/base/css/base.min.css"> <link rel="stylesheet" href="https://static.pudn.com/base/css/fancy.min.css"> <link rel="stylesheet" href="https://static.pudn.com/prod/directory_preview_static/622b74eb3d2fbb0007fc7f66/raw.css"> <script src="https://static.pudn.com/base/js/compatibility.min.js"></script> <script src="https://static.pudn.com/base/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://static.pudn.com/prod/directory_preview_static/622b74eb3d2fbb0007fc7f66/bg1.jpg"><div class="t m0 x1 h2 y1 ff1 fs0 fc0 sc0 ls0 ws0">Evaluation of Fuzzy-based M<span class="ls1 ws1">aximum Power Tracking in </span></div><div class="t m0 x2 h2 y2 ff1 fs0 fc0 sc0 ls0 ws2">Wind Energy Conversion Systems (WECS) <span class="ff2 ls2 ws3">&#58883;</span></div><div class="t m0 x3 h3 y3 ff3 fs1 fc0 sc0 ls3 ws4">Maher Azzouz<span class="_ _0"></span>, Abdel-lati<span class="_ _0"></span>f Elshafei, and <span class="_ _0"></span>Hasan Emara </div><div class="t m0 x4 h4 y4 ff4 fs2 fc0 sc0 ls4 ws3">Abstract&#8212;<span class="ff1 ls5 ws5">Intelligent control techniques c<span class="_ _1"></span>an play a pivotal </span></div><div class="t m0 x4 h4 y5 ff1 fs2 fc0 sc0 ls6 ws6">role in improvin<span class="_ _0"></span>g the performance an<span class="_ _0"></span>d the efficiency of win<span class="_ _0"></span>d </div><div class="t m0 x4 h4 y6 ff1 fs2 fc0 sc0 ls7 ws7">energy conversion sy<span class="_ _1"></span>stems (WECS). A ne<span class="_ _1"></span>w scheme </div><div class="t m0 x4 h4 y7 ff1 fs2 fc0 sc0 ls8 ws8">composed of two fuzzy systems <span class="ls9 ws9">is proposed here to achiev<span class="_ _0"></span>e </span></div><div class="t m0 x4 h4 y8 ff1 fs2 fc0 sc0 lsa wsa">this goal. The first fuzzy system<span class="lsb"> mimics the hill climb search </span></div><div class="t m0 x4 h4 y9 ff1 fs2 fc0 sc0 lsc wsb">technique to yield the maximum pow<span class="_ _1"></span>er as a set point. The </div><div class="t m0 x4 h4 ya ff1 fs2 fc0 sc0 lsd wsc">second fuzz<span class="_ _0"></span>y system is an ad<span class="_ _0"></span>aptive PI-like controll<span class="_ _0"></span>er that </div><div class="t m0 x4 h4 yb ff1 fs2 fc0 sc0 lsa wsd">tracks the set point. Simula<span class="ls5 wse">tion results show that the </span></div><div class="t m0 x4 h4 yc ff1 fs2 fc0 sc0 ls2 wsf">proposed algorithm can signific<span class="lsd ws10">antly improve<span class="_ _0"></span> the efficien<span class="_ _0"></span>cy. </span></div><div class="t m0 x5 h3 yd ff3 fs3 fc0 sc0 lse ws3">I. <span class="fs1 ls2">I</span><span class="lsf">NTRODUCTION<span class="_ _1"></span> </span></div><div class="t m0 x6 h3 ye ff3 fs1 fc0 sc0 ls10 ws11">IND en<span class="_ _1"></span>ergy sour<span class="_ _1"></span>ces hav<span class="_ _1"></span>e enjoye<span class="_ _1"></span>d an incr<span class="_ _1"></span>eased </div><div class="t m0 x6 h3 yf ff3 fs1 fc0 sc0 ls11 ws12">interest due both to th<span class="_ _1"></span>e technological </div><div class="t m0 x4 h3 y10 ff3 fs1 fc0 sc0 ls12 ws13">enhancement<span class="_ _0"></span> of wind turbines and t<span class="_ _0"></span>o a significant<span class="_ _0"></span> </div><div class="t m0 x4 h3 y11 ff3 fs1 fc0 sc0 ls13 ws14">decrease of the wind <span class="_ _0"></span>power costs. </div><div class="t m0 x4 h3 y12 ff3 fs1 fc0 sc0 ls14 ws15">To enhance the economical e<span class="_ _0"></span><span class="ls15 ws16">fficiency of WECS, it is </span></div><div class="t m0 x4 h3 y13 ff3 fs1 fc0 sc0 ls16 ws17">critical to devel<span class="_ _0"></span>op advanced co<span class="_ _0"></span>ntrol met<span class="_ _0"></span>hods to extract<span class="_ _0"></span> </div><div class="t m0 x4 h3 y14 ff3 fs1 fc0 sc0 ls17 ws18">maxim<span class="_ _0"></span>um power output<span class="_ _0"></span> of wind<span class="_ _0"></span> turbines<span class="_ _0"></span> at variabl<span class="_ _0"></span>e wind </div><div class="t m0 x4 h3 y15 ff3 fs1 fc0 sc0 ls18 ws19">speeds [1<span class="_ _1"></span>], [2]. </div><div class="t m0 x7 h3 y16 ff3 fs1 fc0 sc0 ls14 ws1a">The mechanical out<span class="_ _0"></span>put power at a gi<span class="_ _0"></span>ven wind speed is </div><div class="t m0 x4 h3 y17 ff3 fs1 fc0 sc0 ls19 ws1b">drastically affected by the turb<span class="_ _1"></span>ine&#8217;s tip speed ratio (TSR), </div><div class="t m0 x4 h3 y18 ff3 fs1 fc0 sc0 ls1a ws1c">which is defined as the ratio of turb<span class="_ _1"></span>ine rotor tip speed to </div><div class="t m0 x4 h3 y19 ff3 fs1 fc0 sc0 ls1b ws1d">wind speed. <span class="_ _0"></span>At a give<span class="_ _0"></span>n wind spee<span class="_ _0"></span>d, the ma<span class="_ _0"></span>ximum turbine<span class="_ _0"></span> </div><div class="t m0 x4 h3 y1a ff3 fs1 fc0 sc0 ls1c ws1e">energy conversion efficiency <span class="ls1d ws1f">occurs at an<span class="_ _1"></span> optimal T<span class="_ _1"></span>SR. </span></div><div class="t m0 x4 h3 y1b ff3 fs1 fc0 sc0 ls1e ws20">Therefo<span class="_ _1"></span>re, as<span class="_ _1"></span> wind spee<span class="_ _1"></span>d chang<span class="_ _1"></span>es, the<span class="_ _1"></span> turbine&#8217;<span class="_ _1"></span>s rotor </div><div class="t m0 x4 h3 y1c ff3 fs1 fc0 sc0 ls16 ws21">speed needs to c<span class="_ _0"></span>hange accordi<span class="_ _0"></span>ngly in order t<span class="_ _0"></span>o maintai<span class="_ _0"></span>n </div><div class="t m0 x4 h3 y1d ff3 fs1 fc0 sc0 ls1f ws22">the optimal TSR and thus to extract the maximum power </div><div class="t m0 x4 h3 y1e ff3 fs1 fc0 sc0 ls20 ws23">from the available wind res<span class="_ _0"></span>ources. </div><div class="t m0 x7 h3 y1f ff3 fs1 fc0 sc0 ls3 ws24">Previous research <span class="_ _0"></span>has focused on t<span class="_ _0"></span>hree types of<span class="_ _0"></span> </div><div class="t m0 x4 h3 y20 ff3 fs1 fc0 sc0 ls21 ws25">maxim<span class="_ _0"></span>um wind po<span class="_ _0"></span>wer extra<span class="_ _0"></span>ction m<span class="_ _0"></span>ethods,<span class="_ _0"></span> namely<span class="_ _0"></span> TSR </div><div class="t m0 x4 h3 y21 ff3 fs1 fc0 sc0 ls22 ws26">control, power signal feedback (PSF) control and hill-</div><div class="t m0 x4 h3 y22 ff3 fs1 fc0 sc0 ls23 ws27">climb searching (HCS) [3].TSR control regulates the wind </div><div class="t m0 x4 h3 y23 ff3 fs1 fc0 sc0 ls24 ws28">turbine rotor speed to maintain an op<span class="_ _1"></span>timal TSR [4]. As </div><div class="t m0 x4 h3 y24 ff3 fs1 fc0 sc0 ls12 ws29">shown in Fig. <span class="_ _0"></span>1, both the wi<span class="_ _0"></span>nd speed and <span class="_ _0"></span>turbine speed </div><div class="t m0 x4 h3 y25 ff3 fs1 fc0 sc0 ls25 ws2a">need to be measured for TSR calculation, and t<span class="_ _0"></span>he optimal </div><div class="t m0 x4 h3 y26 ff3 fs1 fc0 sc0 ls26 ws2b">TSR must be give<span class="_ _0"></span>n to the contr<span class="_ _0"></span>oller. The wi<span class="_ _0"></span>nd speed </div><div class="t m0 x4 h3 y27 ff3 fs1 fc0 sc0 ls13 ws2c">measurement [5], adds to system cost and prese<span class="_ _0"></span>nts </div><div class="t m0 x4 h3 y28 ff3 fs1 fc0 sc0 ls27 ws2d">difficulties in practical implementations. Furthermore, </div><div class="t m0 x4 h3 y29 ff3 fs1 fc0 sc0 ls20 ws2e">this technique can&#8217;t adapt itself for the param<span class="_ _0"></span>eters&#8217; </div><div class="t m0 x4 h3 y2a ff3 fs1 fc0 sc0 ls15 ws2f">changes of the wind turbine <span class="ls22 ws30">which lead to change t<span class="_ _0"></span>he </span></div><div class="t m0 x4 h3 y2b ff3 fs1 fc0 sc0 ls28 ws31">optimal TSR. PSF cont<span class="_ _0"></span>rol requires the <span class="_ _0"></span>knowledge of the </div><div class="t m0 x4 h3 y2c ff3 fs1 fc0 sc0 ls29 ws32">maximum<span class="_ _0"></span> power curve <span class="_ _0"></span>of th<span class="ls2a ws33">e wind turbine [6<span class="_ _1"></span>]. </span></div><div class="t m0 x4 h3 y2d ff3 fs1 fc0 sc0 ls24 ws34">The lack of adaptation results in<span class="_ _1"></span> power losses in case of </div><div class="t m0 x4 h3 y2e ff3 fs1 fc0 sc0 ls25 ws35">aerodynamic changes, such <span class="ls2b ws36">a<span class="_ _0"></span>s air density changes. </span></div><div class="t m0 x4 h3 y2f ff3 fs1 fc0 sc0 ls2 ws3"> </div><div class="t m0 x4 h5 y30 ff3 fs3 fc0 sc0 ls2c ws37"> <span class="_ _2"> </span>Maher Azzouz is a teaching assistant of Electric Power and<span class="_ _1"></span> </div><div class="t m0 x4 h5 y31 ff3 fs3 fc0 sc0 ls2d ws38">Machines Department, Faculty of Engi<span class="ls2e ws39">neering, Cairo <span class="_ _0"></span>University, Egypt, </span></div><div class="t m0 x4 h5 y32 ff3 fs3 fc0 sc0 ls2f ws3a">Giza.E-mail Address: <span class="fc1 ls30 ws3">maher.abdelkhalek@hotmail.com.</span></div><div class="t m0 x8 h5 y33 ff3 fs3 fc0 sc0 ls2 ws3"> </div><div class="t m0 x4 h5 y34 ff3 fs3 fc0 sc0 ls31 ws3b"> <span class="_ _3"> </span>Abdel-latif Elshafei is a professor of Electric Power a<span class="_ _1"></span>nd Machines </div><div class="t m0 x4 h5 y35 ff3 fs3 fc0 sc0 ls32 ws3c">Department, Faculty of Engineering,<span class="_ _0"></span> Cairo University, Egypt, Giza. E-</div><div class="t m0 x4 h5 y36 ff3 fs3 fc0 sc0 ls33 ws3d">mail address: <span class="fc1 ls31 ws3">elshafei@eng.cu.edu.eg.</span></div><div class="t m0 x9 h5 y37 ff3 fs3 fc0 sc0 ls2 ws3"> </div><div class="t m0 x4 h5 y38 ff3 fs3 fc0 sc0 ls31 ws3e"> <span class="_ _4"> </span>Hasan Emara is an associate professor of Electric Power and </div><div class="t m0 x4 h5 y39 ff3 fs3 fc0 sc0 ls2d ws38">Machines Department, Faculty of Engi<span class="ls2e ws39">neering, Cairo <span class="_ _0"></span>University, Egypt, </span></div><div class="t m0 x4 h5 y3a ff3 fs3 fc0 sc0 ls2e ws3f">Giza. E-mail address: <span class="fc1 ls34 ws3">hmrashad@ieee.o<span class="_ _1"></span>rg.</span></div><div class="t m0 xa h5 y3b ff3 fs3 fc2 sc0 ls35 ws3"> </div><div class="t m0 xb h3 y3c ff3 fs1 fc0 sc0 ls20 ws40">To overcome the pre<span class="_ _0"></span>vious<span class="ls2b ws41"> drawbacks, HCS control </span></div><div class="t m0 xc h3 y3d ff3 fs1 fc0 sc0 ls16 ws42">has been propose<span class="_ _0"></span>d to continuo<span class="_ _0"></span>usly search for t<span class="_ _0"></span>he peak </div><div class="t m0 xc h3 y3e ff3 fs1 fc0 sc0 ls36 ws43">output power of the wind tu<span class="_ _0"></span>rbine [7], [8]. HCS contr<span class="_ _0"></span>ol </div><div class="t m0 xc h3 y3f ff3 fs1 fc0 sc0 ls23 ws44">works well only when the wind turb<span class="_ _1"></span>ine inertia is very </div><div class="t m0 xc h3 y40 ff3 fs1 fc0 sc0 ls37 ws45">small so that the turbine-<span class="ws46">speed reacts to wind spee<span class="_ _0"></span>d </span></div><div class="t m0 xc h3 y41 ff3 fs1 fc0 sc0 ls1c ws47">almost &#8220;instantaneously&#8221;. For <span class="ls19 ws48">large inertia wind turbines, </span></div><div class="t m0 xc h3 y42 ff3 fs1 fc0 sc0 ls14 ws49">the system output power is <span class="ls25 ws4a">interlaced wi<span class="_ _0"></span>th<span class="ls2 ws4b"> the turbine </span></span></div><div class="t m0 xc h3 y43 ff3 fs1 fc0 sc0 ls20 ws4c">mechanical power and rate of <span class="ls14 ws4d">cha<span class="_ _0"></span>nge in the mechanically </span></div><div class="t m0 xc h3 y44 ff3 fs1 fc0 sc0 ls26 ws4e">stored energy,<span class="_ _0"></span> which ofte<span class="_ _0"></span>n renders the HCS m<span class="_ _0"></span>ethod </div><div class="t m0 xc h3 y45 ff3 fs1 fc0 sc0 ls15 ws4f">ineffective [9]. </div><div class="t m0 xd h5 y46 ff3 fs3 fc0 sc0 ls2 ws3"> </div><div class="t m0 xe h5 y47 ff3 fs3 fc0 sc0 ls38 ws50">Fig. 1. Block diagram of TSR control.</div><div class="t m0 xf h3 y48 ff3 fs1 fc0 sc0 ls2 ws3"> </div><div class="t m0 x10 h3 y49 ff3 fs3 fc0 sc0 ls39 ws51">Fig. 2. Block diagram of PSF control.<span class="fs1 ls2 ws3"> </span></div><div class="t m0 x11 h5 y4a ff3 fs3 fc0 sc0 ls2 ws3"> </div><div class="t m0 xb h3 y4b ff3 fs1 fc0 sc0 ls13 ws52">The method in [9] is base<span class="_ _0"></span>d on an Advanced Hill </div><div class="t m0 xc h3 y4c ff3 fs1 fc0 sc0 ls3a ws53">Climb Search (AHCS) that m<span class="_ _0"></span>aximizes the power by </div><div class="t m0 xc h3 y4d ff3 fs1 fc0 sc0 ls29 ws54">detecting the i<span class="_ _0"></span>nverter out<span class="_ _0"></span>put power (P<span class="_ _0"></span>out) and the inve<span class="_ _0"></span>rter </div><div class="t m0 xc h3 y4e ff3 fs1 fc0 sc0 ls11 ws55">dc-link voltage. During<span class="_ _1"></span> its training mode, the algorithm </div><div class="t m0 xc h3 y4f ff3 fs1 fc0 sc0 ls16 ws56">continually recor<span class="_ _0"></span>ds and updates operat<span class="_ _0"></span>ing param<span class="_ _0"></span>eters into </div><div class="t m0 xc h3 y50 ff3 fs1 fc0 sc0 ls3b ws57">its programmable lookup<span class="_ _1"></span> table for its intelligent memory </div><div class="t m0 xc h3 y51 ff3 fs1 fc0 sc0 ls14 ws58">feature. Since this m<span class="_ _0"></span>ethod is <span class="ls3c ws59">trainable with its intelligent<span class="_ _0"></span> </span></div><div class="t m0 xc h3 y52 ff3 fs1 fc0 sc0 ls1a ws5a">memory, it can adapt to<span class="_ _1"></span> a turbine. As a result, it is <span class="_ _1"></span>a </div><div class="t m0 xc h3 y53 ff3 fs1 fc0 sc0 ls3d ws5b">solution to th<span class="_ _1"></span>e customization problems of man<span class="_ _1"></span>y turbines. </div><div class="t m0 xc h3 y54 ff3 fs1 fc0 sc0 ls3e ws5c">Also, this algorithm does not<span class="_ _0"></span> require mechanical sensors </div><div class="t m0 xc h3 y55 ff3 fs1 fc0 sc0 ls2 ws5d">(like anemometers), t<span class="_ _0"></span>his lowers its cost and eliminates its </div><div class="t m0 xc h3 y56 ff3 fs1 fc0 sc0 ls1c ws5e">associated practical issues. Ho<span class="ls3f ws5f">wever, it can be seen in [9] </span></div><div class="t m0 xc h3 y57 ff3 fs1 fc0 sc0 ls40 ws60">that the algorithm is relativel<span class="_ _1"></span>y slow and complex as it h<span class="_ _1"></span>as </div><div class="t m0 xc h3 y58 ff3 fs1 fc0 sc0 ls1b ws61">three different m<span class="_ _0"></span>odes of operati<span class="_ _0"></span>on [10]. </div><div class="t m0 xb h3 y59 ff3 fs1 fc0 sc0 ls37 ws62">Fuzzy logic provides a <span class="ls41 ws63">convenie<span class="_ _1"></span>nt meth<span class="_ _1"></span>od for </span></div><div class="t m0 xc h3 y5a ff3 fs1 fc0 sc0 ls40 ws64">constructing a maximum power poin<span class="_ _1"></span>t tracking algorithms </div><div class="t m0 xc h3 y5b ff3 fs1 fc0 sc0 ls1b ws65">[7], [11]. <span class="_ _0"></span>We develop a<span class="_ _0"></span> new adaptiv<span class="_ _0"></span>e fuzzy logic<span class="_ _0"></span> </div><div class="t m0 xc h3 y5c ff3 fs1 fc0 sc0 ls12 ws66">technique for m<span class="_ _0"></span>aximum power p<span class="_ _0"></span>oint trackin<span class="_ _0"></span>g (MPPT) </div><div class="t m0 xc h3 y5d ff3 fs1 fc0 sc0 ls27 ws67">under different aerodynamic conditions. Also, the </div><div class="t m0 xc h3 y5e ff3 fs1 fc0 sc0 ls3d ws68">adaptation feature allows th<span class="_ _1"></span>e algorithm to be robust under </div><div class="t m0 xc h3 y5f ff3 fs1 fc0 sc0 ls27 ws69">wind turbine parameters uncertainties. </div><div class="t m0 xb h3 y60 ff3 fs1 fc0 sc0 ls42 ws6a">This paper al<span class="_ _0"></span>so proposes a new curre<span class="_ _0"></span>nt regulator <span class="_ _0"></span>of </div><div class="t m0 xc h3 y61 ff3 fs1 fc0 sc0 ls3f ws6b">the power track<span class="_ _1"></span>ing loop of variable-<span class="_ _1"></span>speed WECS based </div><div class="t m0 xc h3 y62 ff3 fs1 fc0 sc0 ls43 ws6c">on an adaptive<span class="_ _0"></span> fuzzy system<span class="_ _0"></span>. The proposed<span class="_ _0"></span> controller is<span class="_ _0"></span> </div><div class="t m0 xc h3 y63 ff3 fs1 fc0 sc0 ls24 ws6d">initialized using th<span class="_ _1"></span>e rule-base of the standard PI-like </div><div class="t m0 x4 h6 y64 ff3 fs4 fc0 sc0 ls44 ws3">w </div><div class="c x12 y65 w2 h7"><div class="t m0 x0 h8 y66 ff5 fs5 fc0 sc0 ls2 ws3">978-1-4244-8126-2/10/$26.00 &#169;2010 IEEE</div></div></div><div class="pi" data-data='{"ctm":[1.568627,0.000000,0.000000,1.568627,0.000000,0.000000]}'></div></div> </body> </html>
评论
    相关推荐