【摘要】：城市配電網(wǎng)之所以能夠成為電力系統(tǒng)中舉足輕重的構(gòu)成部分,不僅因?yàn)槟軌驅(qū)Y源的合理配置起到不可或缺的作用,同時對于保障當(dāng)?shù)毓╇娰|(zhì)量和促進(jìn)當(dāng)?shù)亟?jīng)濟(jì)發(fā)展也起到了極大的促進(jìn)作用,因此合理、科學(xué)、完善的配電網(wǎng)規(guī)劃就顯得尤為重要。隨著經(jīng)濟(jì)的快速發(fā)展,原有的配電網(wǎng)建設(shè)遠(yuǎn)遠(yuǎn)跟不上這種發(fā)展速度,在一定的程度上甚至制約了電力系統(tǒng)的發(fā)展,進(jìn)而影響經(jīng)濟(jì)的繁榮發(fā)展,為了滿足ZW市經(jīng)濟(jì)發(fā)展所要的電力負(fù)荷需求、以及資源整合的目的,該區(qū)對配電網(wǎng)進(jìn)行合理科學(xué)的規(guī)劃和改造,同時對電網(wǎng)規(guī)劃的方案進(jìn)行科學(xué)有效的決策是十分有必要的,一方面能夠驗(yàn)證規(guī)劃項(xiàng)目的有效性,另一方面能夠通過決策研究發(fā)現(xiàn)配電網(wǎng)規(guī)劃項(xiàng)目存在的問題,為今后其他的配電網(wǎng)規(guī)劃項(xiàng)目提供重要的參考依據(jù)。電力負(fù)荷預(yù)測是供電規(guī)劃的基礎(chǔ)和核心,電力負(fù)荷預(yù)測是城市電網(wǎng)規(guī)劃中的基礎(chǔ)工作,對規(guī)劃的質(zhì)量起關(guān)鍵作用。由于電力負(fù)荷受到各種不確定的隨機(jī)因素的影響,在電力負(fù)荷的數(shù)據(jù)序列中存在非平穩(wěn)性、非線性和隨時間變化的特點(diǎn),本文一改傳統(tǒng)的電力負(fù)荷預(yù)測方法—空間負(fù)荷預(yù)測法,將智能算法應(yīng)用到實(shí)際的電網(wǎng)規(guī)劃的負(fù)荷預(yù)測中,通過集合經(jīng)驗(yàn)?zāi)B(tài)分解(Ensemble Empirical Mode Decomposition,EEMD)將電力負(fù)荷的數(shù)據(jù)序列分解為一系列相互獨(dú)立的固有模態(tài)函數(shù)和一個殘余函數(shù),并采用遺傳算法(Genetic Algorithm,GA),將分解之后的一系列函數(shù)賦予一定的權(quán)重,最后根據(jù)每個函數(shù)的序列特征分別采用最小二乘法支持向量機(jī)(Least Square Support Vector Machine,LSSVM)和非參數(shù)廣義自回歸條件異方差(non-parametric generalized auto-regressive conditional heteroskedasticity,NPGARCH)進(jìn)行預(yù)測,最后將預(yù)測的結(jié)果累加,得到目標(biāo)電力負(fù)荷預(yù)測結(jié)果,實(shí)例證明,該方法能夠有效提高電力負(fù)荷預(yù)測的精度,從而為配電網(wǎng)規(guī)劃的科學(xué)性和安全性提供可靠的基礎(chǔ)。其次,對ZW市的配電網(wǎng)概況進(jìn)行分析,從配網(wǎng)規(guī)劃的目標(biāo)、原則以及重點(diǎn)出發(fā),研究了目前該市配電網(wǎng)中各電壓等級中存在的問題,結(jié)合這些問題,分析研究了在配電網(wǎng)規(guī)劃中涉及到的利益主體的利益需求,以此構(gòu)建了多目標(biāo)群決策的評價指標(biāo),提出了基于熵權(quán)理論的配電網(wǎng)規(guī)劃方案的多目標(biāo)群決策模型。將該模型應(yīng)用到ZW市配電網(wǎng)規(guī)劃中,結(jié)果表明該該市配電網(wǎng)規(guī)劃方案滿足各方主體的總體利益,因此該規(guī)劃方案是比較合理科學(xué)的。
[Abstract]:The reason why urban distribution network can become an important component of power system is not only because it can play an indispensable role in the rational allocation of resources. At the same time, it plays a great role in ensuring the quality of local power supply and promoting the development of local economy. Therefore, reasonable, scientific and perfect distribution network planning is particularly important. With the rapid development of economy, the original distribution network construction can not keep up with the speed of development. To a certain extent, it even restricts the development of the power system, and then affects the prosperity of the economy. In order to meet the demand of electric power load for the economic development of ZW city and the purpose of resource integration, it is necessary to plan and transform the distribution network reasonably and scientifically, and to make scientific and effective decision on the power network planning scheme at the same time. On the one hand, it can verify the effectiveness of the planning project, on the other hand, it can find out the problems existing in the distribution network planning project through decision-making research, which provides an important reference for other distribution network planning projects in the future. Power load forecasting is the basis and core of power supply planning, and power load forecasting is the basic work of urban power network planning, which plays a key role in the quality of power supply planning. Because the power load is affected by various uncertain random factors, and there are non-stationary, nonlinear and time-varying characteristics in the data series of power load, this paper changes the traditional power load forecasting method-space load forecasting method. The intelligent algorithm is applied to load forecasting in practical power network planning. The data sequence of power load is decomposed into a series of independent inherent mode functions and a residual function by means of the set empirical mode decomposition (Ensemble Empirical Mode decomposition). Genetic algorithm (GA) is used to give a certain weight to a series of functions after decomposition. Finally, according to the sequence characteristics of each function, the least square support vector machine (Least Square Support Vector machine) and the nonparametric generalized autoregressive conditional heteroscedasticity (NPGARCH) are used to predict each function. The result of target power load forecasting is obtained. An example shows that this method can effectively improve the accuracy of power load forecasting, thus providing a reliable basis for the scientific and security of distribution network planning. Secondly, the general situation of distribution network in ZW city is analyzed. From the goal, principle and emphasis of distribution network planning, the problems existing in each voltage grade of distribution network in this city are studied, and these problems are combined. This paper analyzes and studies the interest needs of the stakeholders involved in the distribution network planning, constructs the evaluation index of multi-objective group decision making, and puts forward a multi-objective group decision-making model of distribution network planning scheme based on entropy weight theory. The model is applied to the distribution network planning of ZW city. The results show that the distribution network planning scheme of ZW city meets the overall interests of all parties, so the plan is reasonable and scientific.
【學(xué)位授予單位】：華北電力大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TM715

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 邱是;;基于EMD分解與ARMA模型的電力負(fù)荷預(yù)測[J];四川水力發(fā)電;2016年01期

2 那迪;孫莉;張全;滿飛宇;陳云鶴;;不確定語言多屬性群決策中屬性權(quán)重確定方法[J];科技視界;2016年02期

3 孫義;黃海峰;丁建華;;多屬性群決策權(quán)重調(diào)整自適應(yīng)算法[J];計(jì)算機(jī)工程與應(yīng)用;2014年02期

4 劉達(dá);;基于誤差校正的中長期負(fù)荷預(yù)測模型[J];電網(wǎng)技術(shù);2012年08期

5 丁世飛;齊丙娟;譚紅艷;;支持向量機(jī)理論與算法研究綜述[J];電子科技大學(xué)學(xué)報(bào);2011年01期

6 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)測分析及實(shí)現(xiàn)[J];機(jī)械設(shè)計(jì)與制造;2010年02期

7 范高鋒;王偉勝;劉純;戴慧珠;;基于人工神經(jīng)網(wǎng)絡(luò)的風(fēng)電功率預(yù)測[J];中國電機(jī)工程學(xué)報(bào);2008年34期

8 王鴻斌;張立毅;胡志軍;;人工神經(jīng)網(wǎng)絡(luò)理論及其應(yīng)用[J];山西電子技術(shù);2006年02期

9 徐軍華,劉天琪;基于小波分解和人工神經(jīng)網(wǎng)絡(luò)的短期負(fù)荷預(yù)測[J];電網(wǎng)技術(shù);2004年08期

10 朱大奇;人工神經(jīng)網(wǎng)絡(luò)研究現(xiàn)狀及其展望[J];江南大學(xué)學(xué)報(bào);2004年01期

相關(guān)博士學(xué)位論文 前1條

1 王國勛;基于多目標(biāo)決策的數(shù)據(jù)挖掘模型選擇研究[D];電子科技大學(xué);2013年

相關(guān)碩士學(xué)位論文 前4條

1 劉劍沖;含分布式電源的城市配電網(wǎng)規(guī)劃方案綜合評價研究[D];華北電力大學(xué);2015年

2 祁之強(qiáng);智能電網(wǎng)環(huán)境下配電網(wǎng)規(guī)劃與運(yùn)營風(fēng)險(xiǎn)型決策模型研究[D];華北電力大學(xué);2015年

3 易于;配電網(wǎng)規(guī)劃方案的綜合評價與決策[D];上海電力學(xué)院;2012年

4 邴建平;基于多目標(biāo)群決策的區(qū)域水資源配置方案評價研究[D];河海大學(xué);2007年

，

本文編號：2183166