本文選題：配電系統(tǒng) + 同倫-連續(xù)方法��； 參考：《天津大學(xué)》2014年博士論文

【摘要】：近年來，，分布式發(fā)電的開發(fā)和布署呈現(xiàn)快速增長的趨勢，加上新型分布式發(fā)電技術(shù)的涌現(xiàn)，給傳統(tǒng)的配電系統(tǒng)分析、設(shè)計和控制帶來了深刻的變化。因此，發(fā)展綜合型分析工具評估大量分布式發(fā)電接入對配電系統(tǒng)的影響，進(jìn)而通過協(xié)調(diào)控制消除負(fù)面的影響具有非常重要的意義。 同倫－連續(xù)方法是一種魯棒的數(shù)值方法，成功應(yīng)用于求解物理和工程領(lǐng)域的多種問題，常用于克服牛頓－拉夫遜法等迭代方法的局部收斂性。本文采用三階段的同倫增強(qiáng)框架，開發(fā)了可擴(kuò)展的同倫－連續(xù)算法庫，發(fā)展了同倫增強(qiáng)的配電牛頓潮流和同倫增強(qiáng)的輸電潮流，并分別提出了相應(yīng)的簡單問題構(gòu)造方法，既發(fā)揮牛頓－拉夫遜法在輕度或中度負(fù)載情況下二次收斂的優(yōu)勢，也能有效克服初值問題、病態(tài)或奇異引起的不收斂，提高了潮流計算整體的收斂性。 分布式電源、特別是可再生分布式電源接入后，配電網(wǎng)的非線性行為變得更加復(fù)雜。由于分布式電源給配電三相潮流方程引入許多PV節(jié)點(diǎn)，除了傳統(tǒng)的鞍結(jié)點(diǎn)分岔以外，另外一種特殊的分岔－結(jié)構(gòu)誘導(dǎo)分岔也可能在配電系統(tǒng)中出現(xiàn)。本文對配電網(wǎng)中的局部分岔機(jī)理進(jìn)行了研究，并提出了相應(yīng)的分岔計算方法。 連續(xù)方法是追蹤一個或多個參數(shù)變化下解曲線的有效方法。本文提出一種稱為CDFLOW（Continuation Distribution Power Flow）的分析工具，它可以快速、可靠的計算參數(shù)變化下的解曲線和精確分岔點(diǎn)，幫助運(yùn)行人員充分挖掘現(xiàn)有配電網(wǎng)絡(luò)的送電潛力，以便消納更多分布式發(fā)電和給更多負(fù)荷供電，從而提高配電網(wǎng)資產(chǎn)利用率。 各種分布式電源的集成給配電網(wǎng)的運(yùn)行帶來巨大挑戰(zhàn)，特別是可再生能源分布式發(fā)電，由于它們一般在氣候適宜地區(qū)就地接入配電網(wǎng)，容易在局部電網(wǎng)引起配電線路和變壓器的過載、電壓越限和電壓穩(wěn)定問題。本文提出了考慮電壓極限、熱極限和電壓穩(wěn)定極限的配電網(wǎng)可用送到能力（AvailableDelivery Capability，ADC）問題的數(shù)學(xué)模型，并提出了精確計算ADC的數(shù)值方法。然而，確定性的可用送電能力評估忽略了配電系統(tǒng)中的不確定因素，例如分布式發(fā)電和負(fù)荷的隨機(jī)波動，本文基于分布式發(fā)電功率預(yù)測誤差隨不同時間尺度、不同風(fēng)速/光強(qiáng)變化的特點(diǎn)，提出一種依據(jù)誤差分布確定可信采樣區(qū)間的場景生成方法，使得場景的生成更精細(xì)、更有效。向前看和日前的概率ADC評估可以給出電壓越限ADC、熱極限ADC和電壓崩潰ADC的置信區(qū)間，并識別網(wǎng)絡(luò)中潛在的薄弱節(jié)點(diǎn)和支路，相對單個數(shù)值，信息更加完整。
[Abstract]:In recent years, the development and deployment of distributed power generation has shown a rapid growth trend, coupled with the emergence of new distributed generation technology, to the traditional distribution system analysis, design and control brought profound changes. Therefore, it is of great significance to develop a comprehensive analysis tool to evaluate the impact of a large number of distributed generation access on the distribution system, and then to eliminate the negative effects through coordinated control. The homotopy continuous method is a robust numerical method. It has been successfully applied to solve various problems in physics and engineering, and is often used to overcome the local convergence of iterative methods such as Newton-Raphson method. In this paper, an extensible homotopy continuous algorithm library is developed by using a three stage homotopy enhancement framework. The homotopy enhanced distribution Newtonian power flow and homotopy enhanced transmission power flow are developed, and the corresponding simple problem construction methods are presented respectively. Newton-Raphson method can not only give play to the advantage of Newton-Raphson method in the case of mild or moderate load, but also effectively overcome the initial value problem, sick or singular caused by non-convergence, and improve the global convergence of power flow calculation. Especially, the nonlinear behavior of distribution network becomes more complicated after the renewable distributed generation is connected. In addition to the traditional saddle node bifurcation, a special kind of bifurcation, structure-induced bifurcation, may also occur in the distribution system because of the introduction of many PV nodes to the three-phase power flow equation of the distribution system by distributed power generation. In this paper, the mechanism of local bifurcation in distribution network is studied, and the corresponding bifurcation calculation method is put forward. The continuous method is an effective method to trace the solution curve of one or more parameter changes. In this paper, an analytical tool called CDFLOWN continuous Distribution Power flow (CDFLOWN) is proposed. It can quickly and reliably calculate the solution curves and exact bifurcation points under the change of parameters, and help operators to fully exploit the power transmission potential of existing distribution networks. In order to absorb more distributed generation and to supply more loads, thus increasing the utilization rate of distribution network assets. The integration of various distributed power sources poses great challenges to the operation of the distribution network, especially renewable energy distributed generation, Because they are connected to the distribution network in the suitable climate area, it is easy to cause overload of distribution lines and transformers, voltage overruns and voltage stability problems in local power networks. In this paper, a mathematical model for the availability of available delivery capability (ADCC) of distribution network considering voltage limit, thermal limit and voltage stability limit is presented, and a numerical method for accurate calculation of ADC is presented. However, deterministic assessment of available power transmission capacity ignores uncertainties in distribution systems, such as random fluctuations in distributed generation and load. In this paper, the prediction error of distributed generation power varies with different time scales. Based on the characteristics of different wind speed / light intensity, a scene generation method based on the error distribution to determine the trusted sampling interval is proposed, which makes the scene generation more precise and effective. Looking forward and the probabilistic ADC evaluation before the day can give the confidence intervals of voltage overrun ADC thermal limit ADC and voltage collapse ADC and identify the potential weak nodes and branches in the network. The information is more complete than a single value.
【學(xué)位授予單位】：天津大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：TM744

【參考文獻(xiàn)】

相關(guān)期刊論文 前7條

1 王成山;鄭海峰;謝瑩華;陳愷;;計及分布式發(fā)電的配電系統(tǒng)隨機(jī)潮流計算[J];電力系統(tǒng)自動化;2005年24期

2 余貽鑫;李鵬;孫強(qiáng);賈宏杰;;電力系統(tǒng)潮流可行域邊界拓?fù)湫再|(zhì)及邊界算法[J];電力系統(tǒng)自動化;2006年10期

3 陳禮義,戴宏偉,張琦鵬;一種大范圍收斂的電力系統(tǒng)潮流算法——同倫延拓法[J];電力系統(tǒng)及其自動化學(xué)報;1993年01期

4 王毅;侯俊賢;馬世英;鄭超;許曉菲;楊志剛;;用于調(diào)度計劃安全穩(wěn)定校核的潮流數(shù)據(jù)自動整合調(diào)整方法[J];電網(wǎng)技術(shù);2010年04期

5 于爾鏗，潘毅，王憲榮，柳焯;電力系統(tǒng)潮流收斂性的實用性改進(jìn)[J];電網(wǎng)技術(shù);1995年01期

6 梅生偉;朱建全;;智能電網(wǎng)中的若干數(shù)學(xué)與控制科學(xué)問題及其展望[J];自動化學(xué)報;2013年02期

7 劉廣一,胡錫龍,于爾鏗,夏祖治;電力系統(tǒng)病態(tài)潮流計算的新算法[J];中國電機(jī)工程學(xué)報;1991年S1期

本文編號：1988000