兼顧供電量組分特性的最優(yōu)GM(1,N )季度電量預(yù)測(cè)方法
李京平1,陳丹伶2,曾繁華1,王鑫2,方嵩1
(1 廣東電網(wǎng)有限責(zé)任公司中山供電局,廣東 中山 528400;2 廣州市奔流電力科技有限公司,廣東 廣州 510640)
摘 要:提出考慮供電量組分多層級(jí)劃分及外部因素影響,利用關(guān)聯(lián)度尋優(yōu)方法構(gòu)造最優(yōu)GM(1,N )電量預(yù)測(cè)模型。根據(jù)供電地區(qū)的行業(yè)用電分類,對(duì)總供電量的組分進(jìn)行分層級(jí)劃分和重要性分析;計(jì)算各重要組分及外部影響因素與供電量的關(guān)聯(lián)度,并依據(jù)關(guān)聯(lián)度大小對(duì)各影響因素進(jìn)行排序,再通過(guò)建立不同N下的GM(1,N ) 模型,根據(jù)預(yù)測(cè)精度確定最優(yōu)GM(1,N ) 模型。采用該模型對(duì)廣東電網(wǎng)中山供電局的供電量數(shù)據(jù)進(jìn)行預(yù)測(cè)分析,證明了該模型的預(yù)測(cè)結(jié)果具有較高的準(zhǔn)確性。
關(guān)鍵詞:季度電量預(yù)測(cè);GM(1,N ) 模型;行業(yè)用電分類;外部影響因素
中圖分類號(hào):TM715 文獻(xiàn)標(biāo)識(shí)碼:A 文章編號(hào):1007-3175(2018)01-0027-05
Optimal GM (1, N) Quarter Electric Quantity Forecasting Method Considering Characteristics of Power Supply Components
LI Jing-ping1, CHEN Dan-ling2, ZENG Fan-hua1, WANG Xin2, FANG Song1
(1 Zhongshan Power Supply Bureau, Guangdong Power Grid Co., Ltd, Zhongshan 528400, China;
2 Guangzhou Power Electrical Engineering Technology Co., Ltd, Guangzhou 510640, China)
Abstract: This paper proposed to use the correlation optimization method to construct the optimal GM (1, N) electric quantity prediction model considering the power supply components multilevel division and external influencing factors. According to the industry power utilization classification of power supply area, this paper carried out the power supply components multilevel division, analyzed the importance of the power supply components and calculated the correlation between each important component, together with external influencing factors and the power supply components. Each influencing factor was sorted based on the correlation and the GM (1, N) model of different N was established to determine the optimal one according to the prediction accuracy. The actual power supply data of Zhongshan power supply bureau of Guangdong power grid verifies the high accuracy of this model’s forecasting algorithm.
Key words: quarter power supply forecasting; GM (1, N) model; industry power utilization classification; external influencing factor
參考文獻(xiàn)
[1] PESSANHA J F M, LEON N.Long-term forecasting of household and residential electric customers in Brazil[J].IEEE Latin America Transactions,2012,10(2):1537-1543.
[2] 程潛善,方華亮. 一種應(yīng)用大數(shù)據(jù)技術(shù)的中長(zhǎng)期負(fù)荷預(yù)測(cè)方法[J]. 武漢大學(xué)學(xué)報(bào)(工學(xué)版),2017,50(2):239-244.
[3] KANDIL M S, El-DEBEIKY S M, HASANIEN N E. Long-term load forecasting for fast developing utility using a knowledge-based expert system[J]. IEEE Power Engineering Review, 2002,22(4):78.
[4] 張強(qiáng),王毅,李鼎睿,朱文俊. 基于X -12- A R I M A季節(jié)分解與年度電量校正的月度電量預(yù)測(cè)[ J ] . 電力建設(shè),2017,38(1):76-83.
[5] 吳鈺, 王杰. 綜合最優(yōu)灰色支持向量機(jī)模型在季節(jié)型電力負(fù)荷預(yù)測(cè)中的應(yīng)用[J]. 華東電力,2012(1):18-21.
[6] 宋曉華,祖丕娥,伊靜,劉達(dá). 基于改進(jìn)GM(1,1)和SVM的長(zhǎng)期電量?jī)?yōu)化組合預(yù)測(cè)模型[J]. 中南大學(xué)學(xué)報(bào)(自然科學(xué)版),2012,43(5):1803-1807.
[7] 劉宇,郭林,陽(yáng)鋒,江登笠,任鈴,李君. 基于改進(jìn)灰色理論的中長(zhǎng)期負(fù)荷預(yù)測(cè)方法研究[J]. 電網(wǎng)與清潔能源,2016,32(8):51-56.
[8] 任工昌, 劉麗, 苗新強(qiáng). 改進(jìn)灰色模型在電力負(fù)荷中的預(yù)測(cè)分析及實(shí)現(xiàn)[J]. 機(jī)械設(shè)計(jì)與制造,2010(2):232-234.
[9] 李曉波.GM(1,N ) 改進(jìn)模型在年度售電量預(yù)測(cè)中的應(yīng)用[J]. 中國(guó)新技術(shù)新產(chǎn)品,2016(3):4-5.
[10] WU Yichun, CHENG Zhenying, LI Miao.Med-long term system structure forecasting of power consumption based on grey derived model[C]//Proceedings of 2013 IEEE International Conference on Grey Systems and Intelligent Services (GSIS),2013.
[11] 劉洪久,沙巨山,季明月,胡彥蓉. 蘇州電力需求的影響因素及電量預(yù)測(cè)研究[J]. 常熟理工學(xué)院學(xué)報(bào),2014,28(5):7-11.
[12] 葛斐,榮秀婷,石雪梅,楊欣,李周. 基于經(jīng)濟(jì)、氣象因素的安徽省年最大負(fù)荷預(yù)測(cè)方法研究[J].中國(guó)電力,2015,48(3):84-87.
[13] 鄭軍偉. 基于灰色系統(tǒng)理論的數(shù)據(jù)關(guān)聯(lián)度建模及其應(yīng)用[D]. 杭州:杭州電子科技大學(xué),2011.
[14] 趙莉琴,劉敬嚴(yán). 基于灰色GM (1,N ) 模型的河北物流貨運(yùn)量預(yù)測(cè)[J]. 石家莊鐵道大學(xué)學(xué)報(bào)(社會(huì)科學(xué)版),2015,9(2):10-15.