風(fēng)電功率縱向時刻概率分析與風(fēng)電場儲能容量優(yōu)化
發(fā)布時間：2018-11-05 19:06
【摘要】：隨著能源、環(huán)境問題的日益突出,以及煤炭、石油等非可再生能源的日益枯竭,世界各國均已將可再生能源的發(fā)展提升到戰(zhàn)略高度。其中,風(fēng)能因其污染少、儲量大、不占用耕地等優(yōu)點成為最具大規(guī)模開發(fā)利用潛力的能源。近年來,隨著風(fēng)力發(fā)電技術(shù)的不斷成熟,其規(guī)模逐年擴大,裝機容量逐年增加。因此,風(fēng)力發(fā)電對電網(wǎng)的影響也日益受到關(guān)注。 風(fēng)電具有波動性、間歇性等特點,這使其面臨不確定性和難以準確預(yù)測等問題。風(fēng)電的大規(guī)模并網(wǎng)給電網(wǎng)的安全穩(wěn)定運行、電能質(zhì)量等方面帶來挑戰(zhàn)。如何平抑風(fēng)電的波動,成為重要研究課題。在此背景下,論文從風(fēng)功率波動特性、風(fēng)電場儲能容量優(yōu)化、風(fēng)功率分級后進行儲能等多方面進行研究,以期提高風(fēng)電場功率輸出的可靠性,提高風(fēng)功率的利用率,提高風(fēng)電的可調(diào)度性。論文的主要工作可以概括為以下幾個部分： 首先,提出一種新的分析風(fēng)電功率波動特性的方法,即縱向時刻概率分析法,該方法基于實測歷史數(shù)據(jù),統(tǒng)計365天或更長天數(shù)內(nèi)每天同一時刻的風(fēng)電出力,得到96個不同時刻的概率分布結(jié)果,并通過函數(shù)擬合歸納出由分段函數(shù)表達的風(fēng)電出力概率特征,在此基礎(chǔ)上實現(xiàn)對風(fēng)功率預(yù)測值的預(yù)評估。該方法不僅證明了縱向時刻概率分布特性是風(fēng)電出力的固有屬性,也為后續(xù)功率分級方法的實現(xiàn)提供了依據(jù)。 其次,為使風(fēng)電輸出最大程度滿足調(diào)度需求,引入儲能系統(tǒng),并提出考慮電池壽命和過放現(xiàn)象的風(fēng)電場儲能容量優(yōu)化計算法,該方法將放電深度及過放現(xiàn)象等造成的壽命損傷折合為運行成本,將未滿足期望輸出部分的能量折合為懲罰成本,同時考慮儲能設(shè)備的固有成本,以該三部分綜合經(jīng)濟成本最小為優(yōu)化目標,以功率約束、容量約束、電池壽命約束為約束條件,以遺傳算法為求解方法,來求解最優(yōu)的儲能容量。儲能系統(tǒng)配置這一容量后,可以從經(jīng)濟性、可靠性等方面最大程度減小風(fēng)功率波動,滿足調(diào)度需求。 再次,為減小儲能系統(tǒng)容量,降低儲能成本,提高風(fēng)電利用率,提出一種基于縱向時刻概率分析方法和區(qū)間估計理論的風(fēng)功率分級方法,并在分級方法的基礎(chǔ)上進行新的風(fēng)電場儲能容量優(yōu)化。風(fēng)功率分級是將風(fēng)電功率分為一級出力、二級出力、三級出力,其中前兩級出力可靠性較高,可直接用于風(fēng)電調(diào)度,三級出力用于優(yōu)化儲能容量。利用三級出力求取的儲能容量較小,可大大降低儲能成本,一級出力、二級出力和儲能后三級出力之和作為風(fēng)場輸出,可有效提高風(fēng)場輸出的穩(wěn)定度和利用率。 最后,在前述研究內(nèi)容的基礎(chǔ)上,以分級后加入小儲能系統(tǒng)的風(fēng)場輸出作為歷史數(shù)據(jù),進行風(fēng)功率預(yù)測,與不加儲能時利用原始風(fēng)電功率數(shù)據(jù)進行的預(yù)測相比,前者的預(yù)測精度顯著提高。這種分級后儲能的方法對于實現(xiàn)風(fēng)電的可靠調(diào)度具有現(xiàn)實意義。
[Abstract]:With the increasingly prominent energy and environmental problems, as well as the depletion of non-renewable energy such as coal and oil, countries in the world have promoted the development of renewable energy to a strategic level. Among them, wind energy becomes the most potential energy for large-scale exploitation because of its advantages of less pollution, large reserves and no occupation of cultivated land. In recent years, with the development of wind power generation technology, its scale and installed capacity increase year by year. Therefore, the influence of wind power generation on the power grid has been paid more and more attention. Wind power has the characteristics of volatility and intermittency, which makes it face uncertainty and difficult to predict accurately. The large-scale grid connection of wind power brings challenges to the safe and stable operation of power grid and power quality. How to stabilize the fluctuation of wind power has become an important research topic. Under this background, the paper studies the wind power fluctuation characteristic, the wind farm energy storage capacity optimization, the wind power classification and so on, in order to improve the reliability of the wind farm power output and the efficiency of the wind power utilization. Improve the schedulability of wind power. The main work of this paper can be summarized as follows: firstly, a new method to analyze the fluctuation of wind power is proposed, that is, the longitudinal moment probability analysis method, which is based on the measured historical data. According to the statistics of wind power output at the same time every day for 365 days or longer days, the probability distribution results of 96 different times are obtained, and the probability characteristics of wind power output expressed by piecewise function are summed up by function fitting. On this basis, the prediction of wind power can be evaluated. This method not only proves that the probability distribution characteristic of longitudinal moment is the inherent attribute of wind power generation, but also provides the basis for the realization of subsequent power classification method. Secondly, in order to maximize the output of wind power to meet the demand of dispatching, the energy storage system is introduced, and the optimal calculation method of energy storage capacity of wind farm considering battery life and over-discharge phenomenon is proposed. In this method, the life damage caused by discharge depth and overdischarge phenomenon is reduced to the operating cost, and the energy which is not satisfied with the expected output is converted into the penalty cost, and the inherent cost of the energy storage equipment is considered at the same time. The optimal energy storage capacity is solved by taking the minimum comprehensive economic cost as the optimization objective, the power constraint, the capacity constraint, the battery life constraint as the constraint conditions and the genetic algorithm as the solution method. After the energy storage system is configured with this capacity, the fluctuation of wind power can be minimized to the greatest extent from the aspects of economy and reliability, and the dispatching demand can be satisfied. Thirdly, in order to reduce the capacity of energy storage system, reduce the cost of energy storage and improve the utilization rate of wind power, a wind power classification method based on longitudinal time probability analysis and interval estimation theory is proposed. The new wind farm energy storage capacity is optimized based on the classification method. Wind power classification is to divide wind power into first output, second output and third output, among which the first two are of high reliability and can be directly used in wind power dispatching, and the third is used to optimize energy storage capacity. The cost of energy storage can be greatly reduced by using the small storage capacity of the three-stage output, and the sum of the first-order output and the three-stage output after the storage can be taken as the output of the wind field, which can effectively improve the stability and utilization ratio of the output of the wind field. Finally, on the basis of the above research, the wind field output of the small energy storage system is used as the historical data to predict the wind power, compared with the prediction using the original wind power data without the energy storage. The prediction accuracy of the former is improved significantly. This method of energy storage after classification has practical significance for the reliable dispatching of wind power.
【學(xué)位授予單位】：山東大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TM614
本文編號：2313082
[Abstract]:With the increasingly prominent energy and environmental problems, as well as the depletion of non-renewable energy such as coal and oil, countries in the world have promoted the development of renewable energy to a strategic level. Among them, wind energy becomes the most potential energy for large-scale exploitation because of its advantages of less pollution, large reserves and no occupation of cultivated land. In recent years, with the development of wind power generation technology, its scale and installed capacity increase year by year. Therefore, the influence of wind power generation on the power grid has been paid more and more attention. Wind power has the characteristics of volatility and intermittency, which makes it face uncertainty and difficult to predict accurately. The large-scale grid connection of wind power brings challenges to the safe and stable operation of power grid and power quality. How to stabilize the fluctuation of wind power has become an important research topic. Under this background, the paper studies the wind power fluctuation characteristic, the wind farm energy storage capacity optimization, the wind power classification and so on, in order to improve the reliability of the wind farm power output and the efficiency of the wind power utilization. Improve the schedulability of wind power. The main work of this paper can be summarized as follows: firstly, a new method to analyze the fluctuation of wind power is proposed, that is, the longitudinal moment probability analysis method, which is based on the measured historical data. According to the statistics of wind power output at the same time every day for 365 days or longer days, the probability distribution results of 96 different times are obtained, and the probability characteristics of wind power output expressed by piecewise function are summed up by function fitting. On this basis, the prediction of wind power can be evaluated. This method not only proves that the probability distribution characteristic of longitudinal moment is the inherent attribute of wind power generation, but also provides the basis for the realization of subsequent power classification method. Secondly, in order to maximize the output of wind power to meet the demand of dispatching, the energy storage system is introduced, and the optimal calculation method of energy storage capacity of wind farm considering battery life and over-discharge phenomenon is proposed. In this method, the life damage caused by discharge depth and overdischarge phenomenon is reduced to the operating cost, and the energy which is not satisfied with the expected output is converted into the penalty cost, and the inherent cost of the energy storage equipment is considered at the same time. The optimal energy storage capacity is solved by taking the minimum comprehensive economic cost as the optimization objective, the power constraint, the capacity constraint, the battery life constraint as the constraint conditions and the genetic algorithm as the solution method. After the energy storage system is configured with this capacity, the fluctuation of wind power can be minimized to the greatest extent from the aspects of economy and reliability, and the dispatching demand can be satisfied. Thirdly, in order to reduce the capacity of energy storage system, reduce the cost of energy storage and improve the utilization rate of wind power, a wind power classification method based on longitudinal time probability analysis and interval estimation theory is proposed. The new wind farm energy storage capacity is optimized based on the classification method. Wind power classification is to divide wind power into first output, second output and third output, among which the first two are of high reliability and can be directly used in wind power dispatching, and the third is used to optimize energy storage capacity. The cost of energy storage can be greatly reduced by using the small storage capacity of the three-stage output, and the sum of the first-order output and the three-stage output after the storage can be taken as the output of the wind field, which can effectively improve the stability and utilization ratio of the output of the wind field. Finally, on the basis of the above research, the wind field output of the small energy storage system is used as the historical data to predict the wind power, compared with the prediction using the original wind power data without the energy storage. The prediction accuracy of the former is improved significantly. This method of energy storage after classification has practical significance for the reliable dispatching of wind power.
【學(xué)位授予單位】：山東大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TM614
【參考文獻】
相關(guān)期刊論文 前10條
1 袁越;李強;李群;張新松;;風(fēng)電功率特性分析及其不確定性解決方案[J];電力科學(xué)與技術(shù)學(xué)報;2011年01期
2 高德賓;李群;金元;于駿;張健男;;東北電網(wǎng)風(fēng)電運行特性分析與研究[J];電力技術(shù);2010年02期
3 雷亞洲;與風(fēng)電并網(wǎng)相關(guān)的研究課題[J];電力系統(tǒng)自動化;2003年08期
4 陳海焱;陳金富;段獻忠;;含風(fēng)電場電力系統(tǒng)經(jīng)濟調(diào)度的模糊建模及優(yōu)化算法[J];電力系統(tǒng)自動化;2006年02期
5 劉昌金;胡長生;李霄;陳敏;徐德鴻;;基于超導(dǎo)儲能系統(tǒng)的風(fēng)電場功率控制系統(tǒng)設(shè)計[J];電力系統(tǒng)自動化;2008年16期
6 王芝茗;蘇安龍;魯順;;基于電力平衡的遼寧電網(wǎng)接納風(fēng)電能力分析[J];電力系統(tǒng)自動化;2010年03期
7 畢大強;葛寶明;王文亮;柴建云;;基于釩電池儲能系統(tǒng)的風(fēng)電場并網(wǎng)功率控制[J];電力系統(tǒng)自動化;2010年13期
8 肖創(chuàng)英;汪寧渤;陟晶;丁坤;;甘肅酒泉風(fēng)電出力特性分析[J];電力系統(tǒng)自動化;2010年17期
9 丁明;徐寧舟;畢銳;;用于平抑可再生能源功率波動的儲能電站建模及評價[J];電力系統(tǒng)自動化;2011年02期
10 李智;韓學(xué)山;楊明;鐘世民;;基于分位點回歸的風(fēng)電功率波動區(qū)間分析[J];電力系統(tǒng)自動化;2011年03期
，本文編號：2313082
本文鏈接：http://sikaile.net/kejilunwen/dianlilw/2313082.html
最近更新
教材專著
