【摘要】：進(jìn)入21世紀(jì)后,大范圍開發(fā)利用化石能源爆發(fā)的能源危機(jī)、環(huán)境危機(jī)日趨嚴(yán)峻,基于傳統(tǒng)化石能源發(fā)展的電力工業(yè)面臨重大挑戰(zhàn),對(duì)新能源的探索已經(jīng)成為能源變革領(lǐng)域中新興的研究熱點(diǎn)。電動(dòng)汽車(Electric Vehicle,EV)作為一種新能源汽車,具有能源利用效率高、清潔環(huán)保等特點(diǎn),是未來交通工具的主要發(fā)展方向。在轉(zhuǎn)型智能交通的大趨勢(shì)下,車輛入電網(wǎng)(Vehicle-to-Grid,V2G)系統(tǒng)正在成為新的研究熱點(diǎn)。在智能電網(wǎng)的技術(shù)支持下,V2G可以實(shí)現(xiàn)電動(dòng)汽車、充電站與電網(wǎng)之間的雙向交互。隨著電動(dòng)汽車高速市場(chǎng)化推廣,其充饋電行為將對(duì)電網(wǎng)帶來的影響不容小覷。電動(dòng)汽車用戶不合理的充電行為將會(huì)造成電網(wǎng)“峰上加峰”,甚至形成新的用電高峰。而通過V2G合理控制用戶充電行為,并且激勵(lì)用戶積極參與饋電,可有效優(yōu)化電網(wǎng)負(fù)荷曲線,提高電網(wǎng)利用率和穩(wěn)定性。在這樣的時(shí)代背景下,本文將通過設(shè)計(jì)合理的電價(jià)機(jī)制,激勵(lì)電動(dòng)汽車用戶積極參與V2G,合理規(guī)劃其充饋電行為,減緩對(duì)電網(wǎng)穩(wěn)定性的影響,主要包括:首先,本文基于最小二乘支持向量機(jī)(least squares support vector machine,LS-SVM)的數(shù)學(xué)方法,結(jié)合改進(jìn)的蟻群算法共同作用實(shí)現(xiàn)對(duì)沒有電動(dòng)汽車接入時(shí)的電網(wǎng)日負(fù)荷的有效預(yù)測(cè),得到的電網(wǎng)負(fù)荷曲線將作為V2G場(chǎng)景中的背景負(fù)荷,不僅是引導(dǎo)電動(dòng)汽車合理規(guī)范充饋電行為的指導(dǎo)方向,同時(shí)更是合理制定V2G充饋電電價(jià)的基礎(chǔ)之一。其次,本文基于充電需求迫切程度將電動(dòng)汽車用戶區(qū)分為彈性需求與剛性需求用戶,并引入效用函數(shù)對(duì)用戶滿意度進(jìn)行衡量;同時(shí)基于電網(wǎng)背景負(fù)荷曲線對(duì)電網(wǎng)供電代價(jià)進(jìn)行量化,最終從時(shí)隙公平最大化的角度建立優(yōu)化模型。在這個(gè)模型中充電電價(jià)將以動(dòng)態(tài)滑窗形式實(shí)時(shí)制定并發(fā)布,用以引導(dǎo)EV充電行為,并與單一充電電價(jià)下EV隨機(jī)充電模式進(jìn)行對(duì)比,驗(yàn)證了V2G充電定價(jià)機(jī)制對(duì)電網(wǎng)負(fù)荷曲線的平滑作用。最后,在V2G充電定價(jià)機(jī)制的基礎(chǔ)上,對(duì)單一電價(jià)下EV隨機(jī)充電負(fù)荷曲線進(jìn)行修正,提出一種V2G饋電定價(jià)機(jī)制,并通過Matlab進(jìn)行仿真,驗(yàn)證了合理V2G充饋電定價(jià)機(jī)制能實(shí)現(xiàn)對(duì)電網(wǎng)負(fù)荷曲線的有效平滑,提高運(yùn)營穩(wěn)定性。
[Abstract]:After entering the 21st century, the energy crisis of large-scale exploitation and utilization of fossil energy, the environmental crisis is becoming more and more serious, and the electric power industry based on the development of traditional fossil energy is facing great challenges. The exploration of new energy has become a new research hotspot in the field of energy reform. Electric vehicle (Electric Vehicle,EV), as a new energy vehicle, has the characteristics of high energy efficiency, clean and environmental protection, etc. Under the trend of transforming intelligent transportation, Vehicle-to-Grid,V2G system is becoming a new research hotspot. With the support of smart grid, V2G can realize the two-way interaction between electric vehicle, charging station and power grid. With the high speed marketization of electric vehicles, the effect of charging and feeding behavior on the power grid will not be underestimated. The unreasonable charging behavior of electric vehicle users will cause the power grid to "add peak", and even form a new peak. By using V2G to reasonably control the charging behavior of users and to encourage users to actively participate in the feeders, the load curve of power network can be effectively optimized, and the utilization ratio and stability of power network can be improved. In this context, this paper will design a reasonable electricity price mechanism, encourage electric vehicle users to actively participate in V2G, reasonably plan their charging and feeding behavior, mitigate the impact on the stability of power grid, mainly include: first, Based on the mathematical method of least squares support vector machine (least squares support vector machine,LS-SVM) and combined with the improved ant colony algorithm, this paper realizes the effective forecasting of the daily load of power grid without electric vehicle access. The grid load curve will be used as the background load in the V2G scene, which is not only the guiding direction to guide the electric vehicle to regulate the charging behavior reasonably, but also one of the bases for the rational formulation of the V2G charging and feed price. Secondly, based on the urgency of charging demand, the users of electric vehicles are divided into flexible demand and rigid demand, and the utility function is introduced to measure the user satisfaction. At the same time, the cost of power supply is quantified based on the background load curve, and the optimization model is established from the angle of maximizing the fairness of time slot. In this model, the charging price will be set and published in the form of dynamic sliding window in real time to guide the charging behavior of EV, and compared with the random charging mode of EV under a single charge price. The effect of V2G charging pricing mechanism on load curve smoothing is verified. Finally, on the basis of V2G charging pricing mechanism, the random charging load curve of EV is modified under a single charge price, and a V2G feed pricing mechanism is proposed and simulated by Matlab. The reasonable V2G charging and feed pricing mechanism can effectively smooth the load curve and improve the operation stability.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：U492.8;TM73

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李含怡;趙彩虹;陳笑;胡駿;陳子奇;;電動(dòng)汽車充放電模式對(duì)電網(wǎng)日負(fù)荷的影響[J];南京師范大學(xué)學(xué)報(bào)(工程技術(shù)版);2015年03期

2 李明洋;鄒斌;;電動(dòng)汽車充放電決策模型及電價(jià)的影響分析[J];電力系統(tǒng)自動(dòng)化;2015年15期

3 李龍;魏靖;黎燦兵;曹一家;宋軍英;方八零;;基于人工神經(jīng)網(wǎng)絡(luò)的負(fù)荷模型預(yù)測(cè)[J];電工技術(shù)學(xué)報(bào);2015年08期

4 王義軍;李殿文;高超;張洪赫;;基于改進(jìn)的PSO-SVM的短期電力負(fù)荷預(yù)測(cè)[J];電測(cè)與儀表;2015年03期

5 汪海燕;黎建輝;楊風(fēng)雷;;支持向量機(jī)理論及算法研究綜述[J];計(jì)算機(jī)應(yīng)用研究;2014年05期

6 唐舟進(jìn);任峰;彭濤;王文博;;基于迭代誤差補(bǔ)償?shù)幕煦鐣r(shí)間序列最小二乘支持向量機(jī)預(yù)測(cè)算法[J];物理學(xué)報(bào);2014年05期

7 高亞靜;王辰;呂孟擴(kuò);梁海峰;;計(jì)及車主滿意度的電動(dòng)汽車最優(yōu)峰谷分時(shí)電價(jià)模型[J];電力自動(dòng)化設(shè)備;2014年02期

8 陳學(xué)有;文明浩;陳衛(wèi);楊波;;電動(dòng)汽車接入對(duì)電網(wǎng)運(yùn)行的影響及經(jīng)濟(jì)效益綜述[J];陜西電力;2013年09期

9 王晶;陳駿宇;藍(lán)愷;;基于實(shí)時(shí)電價(jià)的微網(wǎng)PSO最優(yōu)潮流算法研究[J];電力系統(tǒng)保護(hù)與控制;2013年16期

10 李成偉;劉俊勇;魏震波;;基于博弈論的電動(dòng)汽車放電電價(jià)研究[J];華東電力;2013年06期

相關(guān)博士學(xué)位論文 前1條

1 楊劍峰;蟻群算法及其應(yīng)用研究[D];浙江大學(xué);2007年

相關(guān)碩士學(xué)位論文 前4條

1 周玲芳;智能電網(wǎng)條件下的用戶側(cè)實(shí)時(shí)電價(jià)機(jī)制研究[D];華北電力大學(xué);2015年

2 王曉涵;電動(dòng)汽車充放電行為建模及V2G研究[D];廣西大學(xué);2014年

3 寇凌峰;電動(dòng)汽車大規(guī)模接入對(duì)電網(wǎng)的影響分析[D];華北電力大學(xué)（北京）;2011年

4 唐燕影;基于灰色理論的電力負(fù)荷預(yù)測(cè)模型研究及系統(tǒng)實(shí)現(xiàn)[D];南昌大學(xué);2010年

，

本文編號(hào)：2404935