發(fā)布時間：2018-11-02 10:15

【摘要】：隨著智能電網(wǎng)不斷發(fā)展,電力行業(yè)信息化程度不斷提高,智能化元件設備不斷應用到電力系統(tǒng),電網(wǎng)自動化平臺積累并處理了海量電力大數(shù)據(jù)。電力大數(shù)據(jù)具備大數(shù)據(jù)的基本特征,具有體量大、多樣性、價值密度低及快速性特點,用傳統(tǒng)數(shù)據(jù)方法難以及時、有效的處理。此外,電力大數(shù)據(jù)是電網(wǎng)運行過程中生產(chǎn)設備和監(jiān)控設備產(chǎn)生的數(shù)據(jù),具有明顯時序特征。電力大數(shù)據(jù)的快速增長,給電力生產(chǎn)的計量、優(yōu)化及調度工作帶來的巨大的挑戰(zhàn)。傳統(tǒng)的數(shù)據(jù)處理方法限于數(shù)據(jù)處理能力,主要基于對整體信息進行采樣,根據(jù)對采樣數(shù)據(jù)確定電網(wǎng)典型運行方式,將電網(wǎng)典型運行方式分析結果進行外推,獲得電網(wǎng)長期運行行為特性。而大數(shù)據(jù)的處理方法,可以直接以全體數(shù)據(jù)作為研究對象,利用數(shù)據(jù)推動的分析方法直接從數(shù)據(jù)中獲得有效信息,進行電網(wǎng)運行狀態(tài)表征及電網(wǎng)運行優(yōu)化。本文提出了適用于電力大數(shù)據(jù)的數(shù)據(jù)預處理方法,利用大數(shù)據(jù)的冗余特點和電氣量之間的物理關系,對缺陷數(shù)據(jù)進行修補,提高了電力大數(shù)據(jù)的利用效率。對電力大數(shù)據(jù)進行篩選,根據(jù)研究目標選取電網(wǎng)運行關鍵信息,構建電網(wǎng)運行數(shù)據(jù)集,為基于運行數(shù)據(jù)對電網(wǎng)進行分析提供了數(shù)據(jù)基礎。提出了區(qū)分關鍵屬性的部分優(yōu)先聚類方法,對電網(wǎng)運行方式進行聚類。在此基礎上求取電網(wǎng)典型運行方式,為電網(wǎng)優(yōu)化研究提供有效的工具。通過提取電力大數(shù)據(jù)集每個時刻關鍵屬性數(shù)據(jù),生成對應時刻的電網(wǎng)運行方式特征參數(shù)數(shù)據(jù)集,并用聚類及聚類融合方法對該數(shù)據(jù)集進行聚類,可以得到電網(wǎng)典型運行方式及各典型運行方式出現(xiàn)概率,本文建立了基于電網(wǎng)運行數(shù)據(jù)集的電網(wǎng)有功網(wǎng)損評估模型,提出了電網(wǎng)有功網(wǎng)損評估方法。在此基礎上建立了計及勵磁系統(tǒng)調差系數(shù)的潮流計算模型,對發(fā)電機勵磁系統(tǒng)調差系數(shù)進行優(yōu)化,分析了不同調差系數(shù)方案下發(fā)電機對電網(wǎng)無功電壓調節(jié)的影響,提出了基于電網(wǎng)數(shù)據(jù)集的發(fā)電機勵磁系統(tǒng)調差系數(shù)優(yōu)化整定方法,以提高電網(wǎng)電壓水平,降低電網(wǎng)有功網(wǎng)損,可以充分考慮到電網(wǎng)發(fā)電、負荷等不確定性,使得優(yōu)化結果更適于電網(wǎng)實際運行情況。
[Abstract]:With the development of smart grid and the improvement of power industry informatization, intelligent components and equipments have been applied to the power system, and the power system automation platform has accumulated and dealt with massive power big data. Power big data has the basic characteristics of big data, it has the characteristics of large volume, diversity, low value density and rapidity, so it is difficult to deal with it in time and effectively with traditional data methods. In addition, power big data is the data generated by production equipment and monitoring equipment in the operation of power grid, which has obvious timing characteristics. The rapid growth of power big data brings great challenges to the measurement, optimization and dispatch of power production. The traditional data processing method is limited to the data processing ability, mainly based on sampling the whole information, according to the sampling data to determine the typical operation mode of the grid, and extrapolating the analysis results of the typical operation mode of the power network. The long-term operation behavior characteristics of power grid are obtained. Big data's processing method can directly take the whole data as the research object, use the data-driven analysis method to obtain the effective information directly from the data, and carry on the power network operation state representation and the grid operation optimization. In this paper, a data preprocessing method suitable for power big data is proposed. The defect data can be repaired by taking advantage of the physical relationship between the redundant characteristics and electrical quantities of big data, and the utilization efficiency of power big data can be improved. In this paper, big data is screened, the key information of power grid operation is selected according to the research objective, and the data set of power grid operation is constructed, which provides a data basis for the analysis of power grid based on operation data. A partial priority clustering method for distinguishing key attributes is proposed to cluster the operation mode of power grid. On the basis of this, the typical operation mode of power grid is obtained, which provides an effective tool for the research of power network optimization. By extracting the key attribute data at each time of power big data set, the characteristic parameter data set of power grid operation mode is generated at the corresponding time, and the data set is clustered by clustering and clustering fusion method. The typical operation mode and the occurrence probability of each typical operation mode can be obtained. In this paper, an active power network loss evaluation model based on grid operation data set is established, and an evaluation method for active power network loss is proposed. On the basis of this, a power flow calculation model considering the adjustment coefficient of excitation system is established, and the adjustment coefficient of generator excitation system is optimized, and the influence of generator on the regulation of reactive power and voltage of power network under different schemes of adjustment coefficient is analyzed. Based on the data set of power grid, a method for optimizing the adjustment coefficient of generator excitation system is proposed to improve the voltage level of power network and reduce the loss of active power network. The uncertainty of power generation and load can be fully taken into account. The optimization results are more suitable for the actual operation of the power network.
【學位授予單位】：華北電力大學(北京)
【學位級別】：博士
【學位授予年份】：2017
【分類號】：TM732

【相似文獻】

相關期刊論文 前10條

1 ;電力行業(yè)電網(wǎng)運行與控制標準化技術委員會成立[J];電力建設;2000年08期

2 劉曉賢;;對銀南電網(wǎng)開展“六復核”工作的探討[J];寧夏電力;2004年03期

3 陳彥鋒;;創(chuàng)新電網(wǎng)安全經(jīng)濟運行的對策[J];中國高新技術企業(yè);2007年15期

4 林森;;2008年3月華東電網(wǎng)電網(wǎng)運行總體情況[J];華東電力;2008年04期

5 ;電網(wǎng)運行培訓部[J];山東電力高等�？茖W校學報;2010年02期

6 吳敏;;冰凍威脅電網(wǎng)安全 廣西電網(wǎng)調度中心積極防控[J];廣西電業(yè);2011年01期

7 姜瑞峰;;加強電網(wǎng)運行維護安全工作的措施[J];農(nóng)村電氣化;2013年06期

8 蒙定中;大電網(wǎng)與大機組的協(xié)調(一)[J];電力系統(tǒng)自動化;1986年05期

9 ;提高認識、狠抓落實，努力開創(chuàng)我局電網(wǎng)運行工作的新局面──張惠清總工程師在青海電網(wǎng)運行工作會議上的總結講話[J];青海電力;1996年03期

10 諸兵;;淺談電網(wǎng)運行技術的優(yōu)化[J];科技與企業(yè);2013年24期

相關會議論文 前10條

1 ;全國電網(wǎng)運行與控制標準化技術委員會成立[A];《電站信息》2009年第1期[C];2009年

2 閔勇;魯宗相;喬穎;徐飛;;風電從“并網(wǎng)”到“融入”大電網(wǎng)運行的轉變[A];2010年電力系統(tǒng)自動化學術研討會論文集[C];2010年

3 韓自奮;;甘肅電網(wǎng)氣象信息服務系統(tǒng)應用功能探討[A];水電站梯級調度及自動控制技術研討會論文集[C];2008年

4 吳琛;李玲芳;張丹;張杰;程e，

本文編號：2305735