基于(yu)DSP的蓄電池充放(fang)電裝置原理

1 引言

在蓄電(dian)(dian)池(chi)生產過(guo)程中，為了(le)保(bao)證產品質量，常需(xu)對(dui)成(cheng)品蓄電(dian)(dian)池(chi)進(jin)行幾(ji)次充(chong)放電(dian)(dian)處(chu)理。傳統充(chong)放電(dian)(dian)設(she)備通常采用(yong)晶閘管作(zuo)為整流(liu)逆變(bian)功率器(qi)件。裝(zhuang)置比較(jiao)復(fu)雜，交流(liu)輸(shu)(shu)進(jin)、輸(shu)(shu)出的(de)(de)功率因(yin)數較(jiao)低。對(dui)電(dian)(dian)網(wang)的(de)(de)諧(xie)波污染也比較(jiao)大。為此(ci)，設(she)計了(le)一種三相SPWM整流(liu)逆變(bian)蓄電(dian)(dian)池(chi)充(chong)放電(dian)(dian)裝(zhuang)置。它采用(yong)IGBT作(zuo)為功率變(bian)換器(qi)件。交流(liu)側以精密鎖(suo)相的(de)(de)正弦波電(dian)(dian)流(liu)實(shi)現電(dian)(dian)能(neng)變(bian)換。可獲接近于1的(de)(de)功率因(yin)數，實(shi)現對(dui)蓄電(dian)(dian)池(chi)的(de)(de)充(chong)放電(dian)(dian)處(chu)理，明顯降低了(le)對(dui)電(dian)(dian)網(wang)的(de)(de)諧(xie)波污染，滿足了(le)綠色(se)環保(bao)和節能(neng)的(de)(de)設(she)計要求。

2 系統結構及工作原理

圖1示(shi)出設(she)計的(de)(de)(de)(de)蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)生產(chan)用充(chong)(chong)(chong)放電(dian)(dian)控(kong)制(zhi)系(xi)(xi)統(tong)(tong)(tong)結構(gou)(gou)。該系(xi)(xi)統(tong)(tong)(tong)從原理(li)上(shang)可劃分為SPWM雙(shuang)向(xiang)逆(ni)變(bian)和DC／DC變(bian)換充(chong)(chong)(chong)放電(dian)(dian)兩個(ge)子(zi)系(xi)(xi)統(tong)(tong)(tong)。前者，在蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)充(chong)(chong)(chong)電(dian)(dian)時(shi)，通過三相PFC升壓(ya)控(kong)制(zhi)實(shi)(shi)現(xian)AC／DC變(bian)換。將交流(liu)電(dian)(dian)網電(dian)(dian)壓(ya)轉(zhuan)換成(cheng)蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)充(chong)(chong)(chong)電(dian)(dian)所需的(de)(de)(de)(de)直(zhi)流(liu)電(dian)(dian)壓(ya)；在蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)放電(dian)(dian)時(shi)，通過三相PFC恒壓(ya)逆(ni)變(bian)控(kong)制(zhi)實(shi)(shi)現(xian)DC／AC變(bian)換，將蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)開釋(shi)的(de)(de)(de)(de)能(neng)量(liang)回饋電(dian)(dian)網。后者，完成(cheng)逆(ni)變(bian)直(zhi)流(liu)電(dian)(dian)能(neng)與蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)電(dian)(dian)能(neng)的(de)(de)(de)(de)轉(zhuan)換，以(yi)保證(zheng)蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)充(chong)(chong)(chong)放電(dian)(dian)過程中所要求的(de)(de)(de)(de)電(dian)(dian)流(liu)、電(dian)(dian)壓(ya)和時(shi)間的(de)(de)(de)(de)控(kong)制(zhi)。各子(zi)系(xi)(xi)統(tong)(tong)(tong)采用單獨的(de)(de)(de)(de)DSP治理(li)，DSP部(bu)分以(yi)模板化直(zhi)插結構(gou)(gou)直(zhi)接(jie)插進(jin)工控(kong)機(ji)的(de)(de)(de)(de)主(zhu)板，工控(kong)機(ji)承擔整個(ge)系(xi)(xi)統(tong)(tong)(tong)的(de)(de)(de)(de)監(jian)控(kong)治理(li)。系(xi)(xi)統(tong)(tong)(tong)由1個(ge)逆(ni)變(bian)子(zi)系(xi)(xi)統(tong)(tong)(tong)和n個(ge)(實(shi)(shi)驗樣機(ji)設(she)計為15個(ge))充(chong)(chong)(chong)放電(dian)(dian)子(zi)系(xi)(xi)統(tong)(tong)(tong)組成(cheng)。系(xi)(xi)統(tong)(tong)(tong)工作時(shi)，可通過工控(kong)機(ji)編(bian)組，使后路(lu)蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)工作于充(chong)(chong)(chong)電(dian)(dian)狀(zhuang)態；n-k路(lu)工作于放電(dian)(dian)狀(zhuang)態，這樣蓄(xu)電(dian)(dian)池(chi)(chi)(chi)(chi)(chi)能(neng)量(liang)就(jiu)可直(zhi)接(jie)在系(xi)(xi)統(tong)(tong)(tong)內(nei)部(bu)進(jin)行(xing)交換，從而明顯進(jin)步了節能(neng)效果。圖2示(shi)出三相SPWM雙(shuang)向(xiang)逆(ni)變(bian)電(dian)(dian)路(lu)采用的(de)(de)(de)(de)典型(xing)電(dian)(dian)壓(ya)型(xing)結構(gou)(gou)主(zhu)電(dian)(dian)路(lu)。

三(san)相反饋電(dian)(dian)(dian)流(liu)iuf，ivf，iwf用于跟蹤由(you)DSP產生的電(dian)(dian)(dian)流(liu)給定信號(hao)，從而控(kong)制直流(liu)端電(dian)(dian)(dian)壓Ud的穩定；Ud的反饋電(dian)(dian)(dian)壓Ut的值經DSP采樣后通(tong)過電(dian)(dian)(dian)壓調節得到作用于電(dian)(dian)(dian)流(liu)內環的電(dian)(dian)(dian)流(liu)給定值。

圖(tu)3示出單相(xiang)PWM整流電路(lu)(lu)的相(xiang)量圖(tu)[2]。固然該系統采用的是三(san)相(xiang)PWM整流電路(lu)(lu)．但其工作(zuo)原(yuan)理與單相(xiang)電路(lu)(lu)相(xiang)似，只(zhi)是從(cong)單相(xiang)擴展到(dao)三(san)相(xiang)。對電路(lu)(lu)進行(xing)SPWM控制(zhi)，在橋(qiao)的交流輸進端A，B，C可得到(dao)三(san)相(xiang)橋(qiao)臂的SPWM電壓uiu，uiv，uiw。對其各相(xiang)按圖(tu)3的相(xiang)量圖(tu)進行(xing)控制(zhi)，就可使(shi)各相(xiang)電流iu，iv，iw為正弦波。且與電壓同相(xiang)位，功(gong)率因(yin)數近似為1。

由此可(ke)(ke)知，控制(zhi)uiu的(de)大小和相位(wei)δ即可(ke)(ke)控制(zhi)電流(liu)的(de)大小和流(liu)向(xiang)，從而(er)控制(zhi)功率(lv)的(de)大小和方向(xiang)。通過對Ud的(de)恒壓(ya)控制(zhi)，實(shi)現逆(ni)變器的(de)功率(lv)流(liu)向(xiang)，從而(er)實(shi)現能(neng)量(liang)的(de)自動雙向(xiang)活動。

3 電壓控制器的設計

圖(tu)4示出AD／DC逆變控(kong)制(zhi)(zhi)(zhi)框圖(tu)。該系(xi)(xi)統采(cai)用電(dian)壓(ya)、電(dian)流(liu)雙閉(bi)環控(kong)制(zhi)(zhi)(zhi)結構(gou)，其電(dian)壓(ya)控(kong)制(zhi)(zhi)(zhi)對象為(wei)直流(liu)量；電(dian)流(liu)控(kong)制(zhi)(zhi)(zhi)對象為(wei)交(jiao)流(liu)量。電(dian)壓(ya)外環采(cai)用數(shu)字算法(fa)予以(yi)實(shi)現；電(dian)流(liu)內環采(cai)用模(mo)擬電(dian)路予以(yi)實(shi)現，以(yi)確保(bao)快速進(jin)(jin)行電(dian)流(liu)控(kong)制(zhi)(zhi)(zhi)，進(jin)(jin)步系(xi)(xi)統工作的可靠性。同時(shi)，為(wei)了使(shi)誤差電(dian)流(liu)與給定相(xiang)位保(bao)持一(yi)致。電(dian)流(liu)調(diao)節器采(cai)用比(bi)例(li)控(kong)制(zhi)(zhi)(zhi)。

蓄(xu)電(dian)(dian)(dian)(dian)(dian)池充電(dian)(dian)(dian)(dian)(dian)時(shi)，輸出電(dian)(dian)(dian)(dian)(dian)壓(ya)Ud低于(yu)(yu)給(gei)定值(zhi)Ud*，則電(dian)(dian)(dian)(dian)(dian)壓(ya)調(diao)節器輸出正(zheng)的(de)(de)uc，輸進電(dian)(dian)(dian)(dian)(dian)壓(ya)Uin經過(guo)一個比例因(yin)子Ku后得到一個與(yu)(yu)Uin同相(xiang)的(de)(de)單位正(zheng)弦us，uc與(yu)(yu)us的(de)(de)乘(cheng)積作為(wei)給(gei)定電(dian)(dian)(dian)(dian)(dian)流(liu)(liu)i*，與(yu)(yu)Uin同相(xiang)，控(kong)制i跟(gen)隨i*，則能(neng)(neng)量就以單位功率(lv)(lv)因(yin)數(shu)從(cong)電(dian)(dian)(dian)(dian)(dian)網(wang)(wang)流(liu)(liu)向(xiang)蓄(xu)電(dian)(dian)(dian)(dian)(dian)池。此時(shi)，變(bian)流(liu)(liu)器工(gong)(gong)作在整流(liu)(liu)狀(zhuang)態。蓄(xu)電(dian)(dian)(dian)(dian)(dian)池放電(dian)(dian)(dian)(dian)(dian)時(shi)，Ud高(gao)于(yu)(yu)Ud*，則uc為(wei)負值(zhi)，uc與(yu)(yu)us相(xiang)乘(cheng)得到與(yu)(yu)Uin反向(xiang)的(de)(de)給(gei)定電(dian)(dian)(dian)(dian)(dian)流(liu)(liu)i*，控(kong)制i跟(gen)隨i*，能(neng)(neng)量就能(neng)(neng)以單位功率(lv)(lv)因(yin)數(shu)從(cong)蓄(xu)電(dian)(dian)(dian)(dian)(dian)池流(liu)(liu)向(xiang)電(dian)(dian)(dian)(dian)(dian)網(wang)(wang)。此時(shi)，變(bian)流(liu)(liu)器工(gong)(gong)作在逆變(bian)狀(zhuang)態。電(dian)(dian)(dian)(dian)(dian)壓(ya)外環(huan)產生(sheng)輸進給(gei)定電(dian)(dian)(dian)(dian)(dian)流(liu)(liu)i*，其幅值(zhi)表明了功率(lv)(lv)的(de)(de)大小；符號(hao)決定了功率(lv)(lv)的(de)(de)流(liu)(liu)向(xiang)；相(xiang)位決定了能(neng)(neng)量傳遞(di)的(de)(de)功率(lv)(lv)因(yin)數(shu)。電(dian)(dian)(dian)(dian)(dian)流(liu)(liu)內環(huan)使輸進電(dian)(dian)(dian)(dian)(dian)流(liu)(liu)跟(gen)蹤給(gei)定，從(cong)而實現(xian)可逆的(de)(de)單位功率(lv)(lv)因(yin)數(shu)變(bian)換。

系(xi)統采用TMS320LF2407A DSP作(zuo)為主處理器，因其有豐富(fu)的(de)(de)(de)外設和(he)較高(gao)的(de)(de)(de)運算速(su)度。由此(ci)可實現較復(fu)雜(za)的(de)(de)(de)控(kong)(kong)制(zhi)及高(gao)精度的(de)(de)(de)數據處理。在此(ci)，通過對(dui)PI控(kong)(kong)制(zhi)、IP控(kong)(kong)制(zhi)和(he)變(bian)速(su)積分PI控(kong)(kong)制(zhi)三種電壓(ya)(ya)調節器算法(fa)的(de)(de)(de)實驗得(de)出其優(you)劣(lie)，從而選擇最適合該(gai)系(xi)統的(de)(de)(de)控(kong)(kong)制(zhi)算法(fa)進行電壓(ya)(ya)調節。

(1)PI控制算法和IP控制算法

圖(tu)(tu)5a示出PI調節器結構圖(tu)(tu)。由圖(tu)(tu)可(ke)得其傳遞

比(bi)(bi)較(jiao)式(shi)(5)和(he)(he)(he)式(shi)(6)可見(jian)，兩種系(xi)統(tong)的(de)傳遞函(han)數(shu)分母相(xiang)同(tong)，故IP調(diao)(diao)節(jie)器可持有與PI相(xiang)同(tong)的(de)無靜差(cha)調(diao)(diao)節(jie)和(he)(he)(he)穩定(ding)(ding)特(te)(te)性(xing)(xing)，同(tong)時(shi)因它在(zai)傳遞函(han)數(shu)上比(bi)(bi)PI少一(yi)個零點，因此(ci)具有比(bi)(bi)PI更好的(de)高頻衰減特(te)(te)性(xing)(xing)，輕易滿足較(jiao)長采(cai)樣周期數(shu)字調(diao)(diao)節(jie)的(de)穩定(ding)(ding)性(xing)(xing)要求(qiu)，能有效抑制混(hun)迭現象。系(xi)統(tong)實(shi)驗證實(shi)，采(cai)用IP調(diao)(diao)節(jie)，調(diao)(diao)節(jie)器參數(shu)很輕易整定(ding)(ding)。可使(shi)系(xi)統(tong)達(da)到穩定(ding)(ding)、無靜差(cha)和(he)(he)(he)很小的(de)超調(diao)(diao)。不(bu)過在(zai)快速性(xing)(xing)方面將有損失。

(2)變速積分PI控制算法

在傳統的(de)PI算(suan)法中(zhong)，因積(ji)(ji)分(fen)增(zeng)益Ki為(wei)常數，在整個調(diao)(diao)節過程中(zhong)，其值不(bu)變。但系統對積(ji)(ji)分(fen)的(de)要求是(shi)偏(pian)差(cha)大(da)時，積(ji)(ji)分(fen)作(zuo)用減(jian)弱，否則(ze)會(hui)產生(sheng)超調(diao)(diao)，甚至出(chu)現積(ji)(ji)分(fen)飽(bao)和；反(fan)之則(ze)加(jia)強(qiang)，否則(ze)不(bu)能滿(man)足(zu)正確性的(de)要求。引進(jin)變速(su)積(ji)(ji)分(fen)PI控制算(suan)法能使控制性能得以滿(man)足(zu)。其基本思路是(shi)偏(pian)差(cha)大(da)時，積(ji)(ji)分(fen)累積(ji)(ji)速(su)度慢，積(ji)(ji)分(fen)作(zuo)用弱；偏(pian)差(cha)小(xiao)時，積(ji)(ji)分(fen)累積(ji)(ji)速(su)度快，積(ji)(ji)分(fen)作(zuo)用強(qiang)。為(wei)此，設置系數f[E(k)]，它(ta)是(shi)偏(pian)差(cha)E(k)的(de)函(han)數，當E(k)增(zeng)大(da)時，f[E(k)]減(jian)小(xiao)；反(fan)之則(ze)增(zeng)大(da)。每次采(cai)樣后(hou)，用f[E(k)]乘E(k)，再進(jin)行(xing)累加(jia)。f[E(k)]與E(k)的(de)關系可表示為(wei)：

在該(gai)系(xi)統中(zhong)，采用(yong)(yong)簡單的(de)(de)變速(su)積分(fen)PI控制，取A=32，B=8，當誤差(cha)(cha)大于(yu)40時，系(xi)統相當于(yu)采用(yong)(yong)純比例調節，因此響應速(su)度加快(kuai)；當誤差(cha)(cha)小于(yu)40并減小到(dao)8的(de)(de)過程中(zhong)，積分(fen)作用(yong)(yong)開始(shi)并逐漸增強，響應過程快(kuai)速(su)平(ping)滑；當誤差(cha)(cha)小于(yu)8時，完全引進積分(fen)作用(yong)(yong)，能快(kuai)速(su)有(you)效地消除靜差(cha)(cha)。該(gai)方(fang)法可(ke)有(you)效抑制系(xi)統的(de)(de)超調，同時也可(ke)兼顧系(xi)統的(de)(de)響應速(su)度。

4 實驗結果

利用(yong)(yong)PI，IP和變(bian)速(su)(su)積(ji)分(fen)PI數字電(dian)壓調(diao)節(jie)器(qi)的(de)逆變(bian)子(zi)系統對(dui)該設計(ji)方案(an)進(jin)行(xing)了大量實驗(yan)。結果可見，采(cai)用(yong)(yong)變(bian)速(su)(su)積(ji)分(fen)PI數字電(dian)壓調(diao)節(jie)器(qi)的(de)綜合性能(neng)優(you)于前兩種算法(fa)(fa)。圖6示(shi)出采(cai)用(yong)(yong)PI調(diao)節(jie)、IP調(diao)節(jie)，以及(ji)變(bian)速(su)(su)積(ji)分(fen)PI調(diao)節(jie)時用(yong)(yong)100M-Tektronix TDS220存儲示(shi)波器(qi)獲取的(de)一組直(zhi)流(liu)母線電(dian)壓Ud的(de)實驗(yan)對(dui)比(bi)波形。逆變(bian)器(qi)起(qi)動(dong)時Ud由150V升至200V。由圖6可見。3種調(diao)節(jie)器(qi)在(zai)無靜差調(diao)節(jie)方面的(de)性能(neng)相同(tong)，而(er)IP的(de)上升時間明(ming)顯(xian)大于另(ling)外兩種算法(fa)(fa)；在(zai)抑(yi)制超(chao)(chao)調(diao)及(ji)高頻噪(zao)聲(sheng)誘發振(zhen)蕩方面，變(bian)速(su)(su)積(ji)分(fen)PI法(fa)(fa)有(you)著明(ming)顯(xian)的(de)上風(feng)，PI系統的(de)起(qi)動(dong)超(chao)(chao)調(diao)超(chao)(chao)過20V，IP系統的(de)超(chao)(chao)調(diao)不到10V，而(er)變(bian)速(su)(su)積(ji)分(fen)PI系統則(ze)無超(chao)(chao)調(diao)。無振(zhen)蕩，能(neng)很快進(jin)進(jin)穩定狀態：在(zai)抗(kang)干擾(rao)性能(neng)方面，變(bian)速(su)(su)積(ji)分(fen)PI系統也具有(you)同(tong)樣(yang)的(de)特點。

5 結論

先容(rong)的(de)逆變器采(cai)用了直流(liu)母(mu)線電(dian)壓(ya)的(de)恒壓(ya)數(shu)字調節(jie)，可方便(bian)地實現電(dian)網能量和(he)蓄電(dian)池能量的(de)雙向活動，精密鎖相(xiang)的(de)SPWM控(kong)(kong)(kong)制(zhi)(zhi)可獲得接近于1的(de)功率(lv)因(yin)數(shu)，理論分(fen)析和(he)系(xi)統實驗表明，在DSP控(kong)(kong)(kong)制(zhi)(zhi)采(cai)樣周期即是交流(liu)電(dian)源周期的(de)交流(liu)控(kong)(kong)(kong)制(zhi)(zhi)系(xi)統中，采(cai)用變速(su)積(ji)分(fen)PI調節(jie)更易(yi)獲得小超調、無振蕩(dang)、無靜差的(de)控(kong)(kong)(kong)制(zhi)(zhi)性(xing)能指標。該設(she)計系(xi)統可攜帶15路3A蓄電(dian)池組(每(mei)組12V蓄電(dian)池15節(jie)串(chuan)聯)進行充放電(dian)子系(xi)統工(gong)作(zuo)，每(mei)路工(gong)作(zuo)由工(gong)控(kong)(kong)(kong)機編(bian)程(cheng)獨立控(kong)(kong)(kong)制(zhi)(zhi)。通過對充電(dian)組和(he)放電(dian)組的(de)公道配(pei)置，可獲得明顯的(de)節(jie)能效果(guo)。