【摘要】：常規(guī)能源的不斷枯竭導致以風能、光伏等可再生能源發(fā)電的DG技術(shù)得到了迅速的發(fā)展。而DG并入電網(wǎng)滲透率過高時,會影響電網(wǎng)的可靠與穩(wěn)定運行。目前對電能質(zhì)量問題的研究基本上都采用先發(fā)后治的策略,無法保證微電網(wǎng)的持續(xù)穩(wěn)定運行。微電網(wǎng)并網(wǎng)變流器的出現(xiàn)能實現(xiàn)電能質(zhì)量問題的跟蹤補償,是保障微電網(wǎng)安全穩(wěn)定運行的一種手段�；诖�,本文提出將并網(wǎng)變流器升級為具有態(tài)勢感知和態(tài)勢利導功能的先進變流器,對涉及電能質(zhì)量變化的各類信息元素進行采集、理解與預測,力求及時準確地掌握所需要素信息的變化態(tài)勢,通過變化態(tài)勢的呈現(xiàn),再對微電網(wǎng)電能質(zhì)量的控制決策進行有利方向的引導,從而可實現(xiàn)動態(tài)靈活調(diào)整和控制。為實現(xiàn)微網(wǎng)系統(tǒng)電能質(zhì)量態(tài)勢感知與微網(wǎng)變流器主動運行決策,本文研究了IEC61850 90-7標準,OpenADR通信規(guī)范、需求響應(yīng)技術(shù)、馬爾科夫決策過程以及微網(wǎng)潮流計算、奇異值分解等相關(guān)知識。基于需求響應(yīng)技術(shù)與微網(wǎng)潮流計算等知識能從不同角度實現(xiàn)對微電網(wǎng)電能質(zhì)量的變化態(tài)勢進行感知;基于馬爾科夫決策過程的建模方法是采用動態(tài)規(guī)劃的思想,通過確定變流器主動運行中的狀態(tài)、行為、狀態(tài)轉(zhuǎn)移概率矩陣和收益矩陣來實現(xiàn)對決策過程建模;基于IEC 61850 90-7標準的分布式系統(tǒng)的信息建模方法可使得微網(wǎng)系統(tǒng)中各設(shè)備具有自描述能力,滿足系統(tǒng)各設(shè)備之間的互操作。以上述技術(shù)為基礎(chǔ),本文提出了兩種分布式發(fā)電主動運行決策方法,分別從基于電網(wǎng)電價以及微網(wǎng)潮流兩個角度去研究。前者通過對電網(wǎng)發(fā)布電價進行分析、理解以實現(xiàn)對微網(wǎng)電能質(zhì)量的需求狀態(tài)的感知,后者通過微網(wǎng)連續(xù)潮流計算的以及奇異值分解法實現(xiàn)對微網(wǎng)電能質(zhì)量的需求狀態(tài)的感知;前者是通過馬爾科夫決策過程模型實現(xiàn)分布式發(fā)電變流器運行模式最優(yōu)決策,而后者是提出了兩個階段的分布式發(fā)電主動運行決策,第一階段是基于奇異值分解的方法實現(xiàn)對微網(wǎng)的潮流Jacobian矩陣分解,找到電壓薄弱節(jié)點,來通過變流器的運行決策實現(xiàn)對電壓薄弱節(jié)點的調(diào)節(jié);第二階段是基于微網(wǎng)潮流預測所得電能質(zhì)量需求信息,變流器基于馬爾科夫決策過程進行運行模式的決策。
[Abstract]:The depletion of conventional energy has led to the rapid development of DG technology which uses renewable energy, such as wind energy and photovoltaic. When the permeability of DG is too high, the reliable and stable operation of power grid will be affected. At present, the research on power quality is based on the strategy of first sending and then treating, which can not guarantee the continuous and stable operation of microgrid. The appearance of microgrid grid-connected converter can realize the tracking compensation of power quality problem and is a means to ensure the safe and stable operation of microgrid. Based on this, this paper proposes to upgrade the grid-connected converter to an advanced converter with situational awareness and situational guidance, and to collect, understand and predict all kinds of information elements related to the change of power quality. This paper tries to grasp the change situation of the required element information in time and accurately, through the presentation of the change situation, and then guide the power quality control decision of microgrid in a favorable direction, so that the dynamic and flexible adjustment and control can be realized. In order to realize the situation perception of power quality and the active operation decision of microgrid converter, the communication specification, demand response technology, Markov decision process and micro-grid power flow calculation of IEC61850 90-7 standard are studied in this paper. Singular value decomposition and other related knowledge. Based on the knowledge of demand response technology and micro-grid power flow calculation, the change situation of power quality of micro-grid can be perceived from different angles, and the modeling method based on Markov decision-making process is based on the idea of dynamic programming. The decision-making process is modeled by determining the state, behavior, state transition probability matrix and income matrix in the active operation of the converter. The information modeling method of distributed system based on IEC 61850 90-7 standard can make each device in microgrid system have self-description ability to satisfy the interoperability among the devices in the system. Based on the above technology, this paper presents two kinds of distributed generation active operation decision methods, which are studied from two aspects of grid price and micro-grid power flow respectively. The former realizes the perception of the demand state of the micro-grid power quality through the analysis of the electricity price released by the grid, and the latter realizes the perception of the demand state of the micro-grid power quality by the micro-grid continuous power flow calculation and the singular value decomposition method. The former uses Markov decision process model to realize the optimal decision of distributed power converter operation mode, while the latter puts forward two stages of distributed generation active operation decision. The first stage is to realize the power flow Jacobian matrix decomposition of the microgrid based on singular value decomposition (SVD) and find the weak voltage node to adjust the voltage weak node through the operation decision of the converter. The second stage is based on the power quality information obtained from the micro-grid power flow prediction, and the converter is based on Markov decision-making process to make the decision of the operation mode.
【學位授予單位】：湘潭大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TM732

【參考文獻】

相關(guān)期刊論文 前10條

1 王守相;梁棟;葛磊蛟;;智能配電網(wǎng)態(tài)勢感知和態(tài)勢利導關(guān)鍵技術(shù)[J];電力系統(tǒng)自動化;2016年12期

2 彭寒梅;曹一家;黃小慶;黃超;;無平衡節(jié)點孤島運行微電網(wǎng)的連續(xù)潮流計算[J];中國電機工程學報;2016年08期

3 劉念;余星火;張建華;;網(wǎng)絡(luò)協(xié)同攻擊:烏克蘭停電事件的推演與啟示[J];電力系統(tǒng)自動化;2016年06期

4 鐘新文;;我國水電裝機突破3億千瓦 世界第一[J];廣西電業(yè);2015年11期

5 段瑤;張步涵;劉怡芳;;基于快速隨機潮流的電力系統(tǒng)安全風險評估[J];中國電機工程學報;2014年22期

6 王力科;楊勝春;曹陽;鄢蜜f ;嚴春華;樊陳;;需求響應(yīng)國際標準體系架構(gòu)研究[J];中國電機工程學報;2014年22期

7 許曉慧;辛建波;范瑞祥;陳麗娟;;基于OpenADR 2.0的需求響應(yīng)信息交互機制研究及仿真驗證[J];華東電力;2014年07期

8 李軼鵬;;DR市場中的OpenADR技術(shù)[J];江西電力;2014年03期

9 彭寒梅;曹一家;黃小慶;;基于BFGS信賴域算法的孤島微電網(wǎng)潮流計算[J];中國電機工程學報;2014年16期

10 彭寒梅;曹一家;黃小慶;羅聰;;基于組合抽樣的含分布式電源隨機潮流計算[J];電力自動化設(shè)備;2014年05期

相關(guān)會議論文 前1條

1 蔡驥然;周振宇;祁兵;伍軍;;需求響應(yīng)通信架構(gòu)和數(shù)據(jù)模型研究[A];2013年中國電機工程學會年會論文集[C];2013年

相關(guān)碩士學位論文 前9條

1 李奇;電能質(zhì)量變化趨勢預測與預警方法研究及其應(yīng)用[D];華南理工大學;2015年

2 王超群;微網(wǎng)電壓穩(wěn)定的全節(jié)點雅可比矩陣奇異值分解法[D];深圳大學;2015年

3 閆興文;微電網(wǎng)電能質(zhì)量分析與控制技術(shù)研究[D];太原理工大學;2014年

4 黨美琳;智能電網(wǎng)需求響應(yīng)信息交換接口研究[D];華北電力大學;2014年

5 白天;基于點的POMDP值迭代算法研究與設(shè)計[D];南京大學;2013年

6 李婧靚;微網(wǎng)的隨機潮流計算研究[D];華中科技大學;2011年

7 石軻;基于馬爾可夫決策過程理論的Agent決策問題研究[D];中國科學技術(shù)大學;2010年

8 羅存;可再生能源分布式發(fā)電站微機監(jiān)控系統(tǒng)的研究與實現(xiàn)[D];華北電力大學（河北）;2009年

9 王珊珊;基于JADE的多Agent系統(tǒng)仿真研究[D];華中科技大學;2006年

，

本文編號：2250264