【摘要】：隨著營(yíng)配一體化協(xié)同工作機(jī)制的建立和營(yíng)配數(shù)據(jù)的打通,用電客戶的用電數(shù)據(jù)能夠與客戶檔案、繳費(fèi)等數(shù)據(jù)關(guān)聯(lián)起來。客戶的用電數(shù)據(jù)中隱含著客戶的用電行為特征,對(duì)這些用電數(shù)據(jù)進(jìn)行深度挖掘并研究客戶類型,可以幫助電網(wǎng)了解客戶的個(gè)性化、差異化服務(wù)需求,從而使電網(wǎng)公司進(jìn)一步拓展服務(wù)的深度和廣度,為未來的電力需求側(cè)響應(yīng)政策的制定提供數(shù)據(jù)支撐。本論文首先是基于大數(shù)據(jù)技術(shù)通過深入分析梳理,確定了用電行為分析的外部數(shù)據(jù)和內(nèi)部數(shù)據(jù)的數(shù)據(jù)來源,分析了海量數(shù)據(jù)存儲(chǔ)技術(shù)和海量數(shù)據(jù)預(yù)處理技術(shù),其次,研究了用電行為分析的關(guān)鍵技術(shù),主要包括聚類算法、最優(yōu)聚類評(píng)價(jià)算法、日期匹配算法、曲線相似性度量算法等用電行為分析算法。再次,構(gòu)建用電分析模型構(gòu)建方案,詳細(xì)描述了用電分析模型的建模思路,研究了主網(wǎng)的用電特征和用電行為,然后結(jié)合主網(wǎng)的用電模式研究主網(wǎng)該模式下海量客戶的用電行為,從而利用模式匹配技術(shù)匹配主網(wǎng)用電模式和該模式下海量客戶的用電模式,建立歷史日削峰填谷匹配關(guān)系,并通過實(shí)證研究對(duì)用電分析模型進(jìn)行驗(yàn)證。最后,對(duì)用電行為分析管理軟件進(jìn)行系統(tǒng)架構(gòu)設(shè)計(jì)、系統(tǒng)功能設(shè)計(jì)、系統(tǒng)數(shù)據(jù)庫設(shè)計(jì),并闡述系統(tǒng)的實(shí)現(xiàn)成果。用電行為分析管理軟件可以比較準(zhǔn)確的對(duì)用電客戶的用電行為進(jìn)行分析,有助于電力公司進(jìn)一步指導(dǎo)用戶個(gè)性化智能用電,對(duì)提高電網(wǎng)需求側(cè)能效管理水平具有建設(shè)性意義。
[Abstract]:With the establishment of the cooperative working mechanism of operation and distribution integration and the opening of the distribution data, the electricity consumption data of the electric customers can be associated with the data of the customers' files, payment and so on. The customer's electricity consumption data is implicit in the customer's electricity behavior characteristic. It can help the power grid to understand the customer's individualized and differentiated service demand by digging these data deeply and studying the customer's type. So the grid company can further expand the depth and breadth of the service, and provide data support for the future DSM policy formulation. Firstly, based on big data technology, the external and internal data sources of electrical behavior analysis are determined, and the mass data storage technology and mass data preprocessing technology are analyzed. The key technologies of electrical behavior analysis are studied, including clustering algorithm, optimal clustering evaluation algorithm, date matching algorithm, curve similarity measurement algorithm and so on. Thirdly, the construction scheme of power consumption analysis model is built, the modeling idea of power consumption analysis model is described in detail, and the characteristics and behavior of power consumption of main network are studied. Then combining the power consumption mode of the main network, the paper studies the power consumption behavior of the massive customers under the main network mode, and then uses the pattern matching technology to match the power consumption mode of the main network and the massive customers under this mode, and establishes the matching relationship between the historical peak cutting and filling valley. And through the empirical study to verify the electricity analysis model. Finally, the system architecture design, system function design, system database design, and the achievement of the system are described. The analysis and management software of power consumption behavior can accurately analyze the power consumption behavior of customers, which is helpful for power companies to guide users to use electricity for personal intelligence, and to improve the management level of energy efficiency on the demand side of power grid.
【學(xué)位授予單位】：華北電力大學(xué)(北京)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP311.13;TM714;TM732

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 張蕾;崔志坤;李井泉;白濤;;改進(jìn)K-means模型在電力系統(tǒng)用戶行為分析中的應(yīng)用[J];無線電工程;2017年03期

2 王歡;李紅輝;張駿溫;;改進(jìn)K-means聚類的云任務(wù)調(diào)度算法[J];計(jì)算機(jī)與現(xiàn)代化;2017年02期

3 陳鵬偉;陶順;肖湘寧;李璐;張劍;;短時(shí)間尺度用電行為相關(guān)性分析網(wǎng)絡(luò)模型[J];電力系統(tǒng)自動(dòng)化;2017年03期

4 劉云霞;;基于動(dòng)態(tài)時(shí)間規(guī)整的面板數(shù)據(jù)聚類方法研究及應(yīng)用[J];統(tǒng)計(jì)研究;2016年11期

5 張素潔;趙懷慈;;最優(yōu)聚類個(gè)數(shù)和初始聚類中心點(diǎn)選取算法研究[J];計(jì)算機(jī)應(yīng)用研究;2017年06期

6 莊緒強(qiáng);;基于云計(jì)算技術(shù)的用戶用電智能分析技術(shù)研究[J];自動(dòng)化與儀器儀表;2016年02期

7 戴月明;張朋;吳定會(huì);;基于密度檢測(cè)的EM算法[J];計(jì)算機(jī)應(yīng)用研究;2016年09期

8 伍育紅;;聚類算法綜述[J];計(jì)算機(jī)科學(xué);2015年S1期

9 王繼業(yè);季知祥;史夢(mèng)潔;黃復(fù)鵬;朱朝陽;張東霞;;智能配用電大數(shù)據(jù)需求分析與應(yīng)用研究[J];中國(guó)電機(jī)工程學(xué)報(bào);2015年08期

10 余冬梅;;基于障礙約束的空間聚類算法綜述[J];計(jì)算機(jī)系統(tǒng)應(yīng)用;2015年01期

相關(guān)碩士學(xué)位論文 前3條

1 陳旭;基于神經(jīng)網(wǎng)絡(luò)的電力系統(tǒng)短期負(fù)荷預(yù)測(cè)研究[D];華中科技大學(xué);2015年

2 潘家騰;基于Hive數(shù)據(jù)倉庫的用戶行為模型研究[D];北京郵電大學(xué);2015年

3 冀明;聚類分析在電力客戶關(guān)系管理中的應(yīng)用研究[D];華北電力大學(xué);2012年

，

本文編號(hào)：2278025