本文選題：臺(tái)區(qū)負(fù)荷預(yù)測(cè)　切入點(diǎn)：聚類(lèi)　出處：《華北電力大學(xué)(北京)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：隨著電力用電信息采集系統(tǒng)的優(yōu)化升級(jí),電力用戶(hù)側(cè)除了能夠采集到傳統(tǒng)的電能使用信息外,也能獲得用戶(hù)電能質(zhì)量、96點(diǎn)負(fù)荷曲線、用電行為偏好等多維度反映用戶(hù)特性的數(shù)據(jù)。配電臺(tái)區(qū)負(fù)荷預(yù)測(cè)作為傳統(tǒng)負(fù)荷預(yù)測(cè)領(lǐng)域的細(xì)分,是進(jìn)行精細(xì)化的臺(tái)區(qū)用電管理、運(yùn)行調(diào)度與網(wǎng)架結(jié)構(gòu)優(yōu)化的新興手段。受制于臺(tái)區(qū)負(fù)荷的隨機(jī)性與多樣性,傳統(tǒng)預(yù)測(cè)方法在臺(tái)區(qū)級(jí)的預(yù)測(cè)中表現(xiàn)欠佳,而借助大數(shù)據(jù)平臺(tái)這一新興工具,對(duì)用戶(hù)側(cè)采集到的冗雜且大量的用戶(hù)特性信息加以利用,能夠有效提高臺(tái)區(qū)負(fù)荷預(yù)測(cè)的預(yù)測(cè)精度與預(yù)測(cè)適應(yīng)性。本文分析臺(tái)區(qū)負(fù)荷數(shù)據(jù)的特點(diǎn)及其關(guān)聯(lián)因素,借助大數(shù)據(jù)平臺(tái)中K-means、BIRCH、WARD等聚類(lèi)算法,對(duì)上海市2977個(gè)臺(tái)區(qū)進(jìn)行聚類(lèi)劃分,結(jié)合實(shí)際臺(tái)區(qū)行業(yè)構(gòu)成分析了各聚類(lèi)簇的特征及劃分的有效性。聚類(lèi)結(jié)果表明臺(tái)區(qū)內(nèi)用戶(hù)對(duì)氣象、經(jīng)濟(jì)等因素均有不同的敏感度,以聚類(lèi)結(jié)果為依據(jù),對(duì)不同用電特性的用戶(hù)類(lèi)分別建立預(yù)測(cè)模型是提高臺(tái)區(qū)負(fù)荷預(yù)測(cè)精度的有效手段。提出基于機(jī)器學(xué)習(xí)中的嶺回歸算法與自適應(yīng)思想的自適應(yīng)嶺回歸預(yù)測(cè)模型,依據(jù)自適應(yīng)程度將模型劃分為3種模式,對(duì)臺(tái)區(qū)負(fù)荷建模并實(shí)際訓(xùn)練、預(yù)測(cè)。3種模式在訓(xùn)練用時(shí)、預(yù)測(cè)精度、敏感度方面表現(xiàn)出不同特性,適用不同預(yù)測(cè)環(huán)境。在聚類(lèi)與回歸模型構(gòu)建的基礎(chǔ)上,進(jìn)一步提出基于聚類(lèi)及自適應(yīng)嶺回歸技術(shù)的臺(tái)區(qū)負(fù)荷預(yù)測(cè)方法,設(shè)計(jì)3種聚類(lèi)特征選取方式與5種聚類(lèi)算法作為預(yù)測(cè)自適應(yīng)優(yōu)化的可調(diào)模塊,增強(qiáng)了預(yù)測(cè)的動(dòng)態(tài)優(yōu)化與誤差調(diào)控能力。使用該方法對(duì)上海市某包含487戶(hù)用戶(hù)的臺(tái)區(qū)進(jìn)行預(yù)測(cè),算例結(jié)果顯示在不同預(yù)測(cè)環(huán)境中通過(guò)優(yōu)選聚類(lèi)特征、聚類(lèi)算法及模型參數(shù)自適應(yīng)調(diào)節(jié),該預(yù)測(cè)方法能達(dá)到較高的預(yù)測(cè)精度與環(huán)境適應(yīng)能力。
[Abstract]:With the optimization and upgrading of electric power information acquisition system, the power user side can not only collect the traditional power usage information, but also obtain the 96 point load curve of power quality. As a subdivision of traditional load forecasting field, distribution station area load forecasting is a fine management of station area power consumption. Due to the randomness and diversity of the load in Taiwan area, the traditional forecasting method is not good in the prediction of Taiwan district level, but with the help of big data platform, which is a new tool, It can effectively improve the forecasting accuracy and adaptability of station load forecasting by using the miscellaneous and large amount of user characteristic information collected by user side. This paper analyzes the characteristics of station load data and its related factors. With the help of K-means-BIRCHWARD and other clustering algorithms in big data platform, the cluster classification of 2977 stations in Shanghai is carried out, and the characteristics of each cluster and the validity of the division are analyzed in combination with the industry composition of the actual station. The clustering results show that the users in the station are sensitive to meteorology. Economic and other factors have different sensitivity, based on clustering results, It is an effective method to improve the accuracy of load forecasting for different users with different power consumption characteristics. An adaptive ridge regression prediction model based on machine learning and adaptive thinking is proposed. According to the adaptive degree, the model is divided into three models, and the load in the station area is modeled and trained in practice. The prediction of the three models shows different characteristics in training time, prediction accuracy and sensitivity. Based on the construction of clustering and regression models, a load forecasting method based on clustering and adaptive ridge regression is proposed. Three kinds of clustering feature selection methods and five clustering algorithms are designed as adjustable modules for adaptive predictive optimization, which enhances the dynamic optimization and error control ability of prediction. The proposed method is used to predict a certain station area in Shanghai, which includes 487 users. The simulation results show that the prediction method can achieve high prediction accuracy and environmental adaptability by optimizing clustering features, clustering algorithm and adaptive adjustment of model parameters in different prediction environments.
【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(lèi)號(hào)】：TM715

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王德文;孫志偉;;電力用戶(hù)側(cè)大數(shù)據(jù)分析與并行負(fù)荷預(yù)測(cè)[J];中國(guó)電機(jī)工程學(xué)報(bào);2015年03期

2 秦礪寒;李順昕;韓江磊;牛東曉;朱正甲;;基于FOA優(yōu)化的BP神經(jīng)網(wǎng)絡(luò)在夏季空調(diào)降溫負(fù)荷預(yù)測(cè)中的應(yīng)用[J];華東電力;2014年12期

3 王德金;;基于大數(shù)據(jù)技術(shù)的短期負(fù)荷分析與預(yù)測(cè)[J];華東電力;2014年10期

4 李瑾;劉金朋;王建軍;;采用支持向量機(jī)和模擬退火算法的中長(zhǎng)期負(fù)荷預(yù)測(cè)方法[J];中國(guó)電機(jī)工程學(xué)報(bào);2011年16期

5 趙淵;張夏菲;謝開(kāi)貴;;非參數(shù)自回歸方法在短期電力負(fù)荷預(yù)測(cè)中的應(yīng)用[J];高電壓技術(shù);2011年02期

6 何永秀;王躍錦;楊麗芳;何海英;羅濤;;基于最小二乘支持向量機(jī)的居民用電預(yù)測(cè)研究[J];電力需求側(cè)管理;2010年03期

7 李妮;江岳春;黃珊;毛李帆;;基于累積式自回歸動(dòng)平均傳遞函數(shù)模型的短期負(fù)荷預(yù)測(cè)[J];電網(wǎng)技術(shù);2009年08期

8 賈慧敏;何光宇;方朝雄;李可文;姚宇臻;黃妹妹;;用于負(fù)荷預(yù)測(cè)的層次聚類(lèi)和雙向夾逼結(jié)合的多層次聚類(lèi)法[J];電網(wǎng)技術(shù);2007年23期

9 金義雄;段建民;徐進(jìn);衛(wèi)功存;蒯圣宇;李宏仲;王承民;;考慮氣象因素的相似聚類(lèi)短期負(fù)荷組合預(yù)測(cè)方法[J];電網(wǎng)技術(shù);2007年19期

10 顧丹珍;艾芊;陳陳;沈善德;;自適應(yīng)神經(jīng)網(wǎng)絡(luò)在負(fù)荷動(dòng)態(tài)建模中的應(yīng)用[J];中國(guó)電機(jī)工程學(xué)報(bào);2007年16期

相關(guān)碩士學(xué)位論文 前2條

1 李小燕;考慮氣象因素的電力系統(tǒng)短期負(fù)荷預(yù)測(cè)研究[D];華南理工大學(xué);2013年

2 馮曉蒲;基于實(shí)際負(fù)荷曲線的電力用戶(hù)分類(lèi)技術(shù)研究[D];華北電力大學(xué);2011年

，

本文編號(hào)：1635149