本文選題：風(fēng)電預(yù)測(cè) + 時(shí)間序列��； 參考：《昆明理工大學(xué)》2017年碩士論文

【摘要】：風(fēng)力發(fā)電作為一種清潔能源,在全世界范圍內(nèi)發(fā)展迅速。雖然風(fēng)電具有一定的優(yōu)勢(shì),主要表現(xiàn)在降低污染物排放和發(fā)電成本等方面,但是隨之而來(lái)的是大規(guī)模風(fēng)電并網(wǎng)會(huì)給電力系統(tǒng)產(chǎn)生一定的影響。風(fēng)電的最顯著特征是具有較大的隨機(jī)性,大規(guī)模的風(fēng)電并網(wǎng)將對(duì)電網(wǎng)的安全運(yùn)行帶來(lái)挑戰(zhàn),并對(duì)系統(tǒng)調(diào)度計(jì)劃的安排產(chǎn)生一定不良的影響�？墒窃谧畛踹M(jìn)行電網(wǎng)規(guī)劃的時(shí)候,是沒(méi)有將風(fēng)電并網(wǎng)考慮在內(nèi)的。因此,本文主要對(duì)兩方面的內(nèi)容進(jìn)行研究:提高風(fēng)電的預(yù)測(cè)精度和含清潔可再生能源的復(fù)合型能源的電力系統(tǒng)的優(yōu)化調(diào)度。首先使用單一的BP神經(jīng)網(wǎng)絡(luò)進(jìn)行預(yù)測(cè),預(yù)測(cè)精度不高。為了提高預(yù)測(cè)精度,使用遺傳算法對(duì)BP神經(jīng)網(wǎng)絡(luò)有關(guān)參數(shù)進(jìn)行優(yōu)化,預(yù)測(cè)精度較單一BP神經(jīng)網(wǎng)絡(luò)有所提高;隨后又建立時(shí)間序列模型,優(yōu)化網(wǎng)絡(luò)的輸入變量。以我國(guó)某風(fēng)電場(chǎng)的某臺(tái)機(jī)組為例,分別使用三種方法對(duì)風(fēng)機(jī)出力進(jìn)行直接和間接預(yù)測(cè)。仿真結(jié)果表明,這兩種方法提高了單一 BP神經(jīng)網(wǎng)絡(luò)的預(yù)測(cè)精度,適合風(fēng)電場(chǎng)的短期預(yù)測(cè)。在提高風(fēng)力發(fā)電預(yù)測(cè)精度的基礎(chǔ)上,對(duì)風(fēng)-火-水復(fù)合型能源電力系統(tǒng)建立多目標(biāo)調(diào)度模型并進(jìn)行求解。對(duì)于多目標(biāo)、多約束的模型,利用傳統(tǒng)的線(xiàn)性或動(dòng)態(tài)規(guī)劃求解方法失效。本文采用粒子群智能算法尋優(yōu)。針對(duì)基本的粒子群算法易于陷入局部最優(yōu),本章在增加擾動(dòng)的LDW的基礎(chǔ)上,提出動(dòng)態(tài)自適應(yīng)的改變慣性權(quán)重系數(shù)和較小概率的引入隨機(jī)個(gè)體的改進(jìn)粒子群算法。仿真分析結(jié)果驗(yàn)證了算法的優(yōu)越性能。在穩(wěn)定的24節(jié)點(diǎn)系統(tǒng)中配以少量風(fēng)電機(jī)組和水電機(jī)組來(lái)建立優(yōu)化調(diào)度模型,以期為大的新能源并網(wǎng)的電力系統(tǒng)提供理論性和應(yīng)用型的依據(jù)。在傳統(tǒng)的經(jīng)濟(jì)調(diào)度問(wèn)題的基礎(chǔ)上,加入了火電機(jī)組不頻繁改變出力作為目標(biāo)函數(shù),保證火電機(jī)組運(yùn)行的經(jīng)濟(jì)性。以水電站的發(fā)電用水量不同,將調(diào)度日分為三種情形,按水電站的汛期、平水期和枯水期三個(gè)時(shí)期的發(fā)電特點(diǎn)進(jìn)行調(diào)度,并利用本文基于動(dòng)態(tài)自適應(yīng)的改變慣性權(quán)重系數(shù)和較小概率的引入隨機(jī)粒子的改進(jìn)PSO算法進(jìn)行求解。仿真結(jié)果表明:水電站在風(fēng)-火-水復(fù)合型能源電力系統(tǒng)中發(fā)揮著調(diào)節(jié)作用,水電站的發(fā)電可用水量的多少?zèng)Q定著水電站在復(fù)合型能源電力系統(tǒng)中的調(diào)節(jié)能力的強(qiáng)弱。
[Abstract]:As a clean energy, wind power generation is developing rapidly all over the world. Although wind power has some advantages, mainly in reducing pollutant emissions and generation costs, but with the large-scale wind power grid will have a certain impact on the power system. The most prominent feature of wind power is that it has a large randomness. The large-scale wind power grid connection will bring challenges to the safe operation of the power grid and have a certain negative impact on the scheduling of the system. However, in the initial planning of the grid, wind power was not taken into account. Therefore, this paper mainly focuses on two aspects: improving the prediction accuracy of wind power and optimal dispatching of power system with clean and renewable energy. First, a single BP neural network is used for prediction, and the prediction accuracy is not high. In order to improve the prediction accuracy, genetic algorithm is used to optimize the parameters of BP neural network, and the prediction accuracy is improved compared with the single BP neural network, and then the time series model is established to optimize the input variables of the network. Taking one unit of a wind farm in China as an example, three methods are used to predict the fan output directly and indirectly. The simulation results show that these two methods can improve the prediction accuracy of single BP neural network and are suitable for short-term wind farm prediction. On the basis of improving the prediction accuracy of wind power generation, the multi-objective dispatching model of wind-fire water complex energy power system is established and solved. For multi-objective and multi-constrained models, traditional linear or dynamic programming methods are used to solve the problems. In this paper, particle swarm optimization (PSO) algorithm is used. The basic particle swarm optimization (PSO) algorithm is easy to fall into local optimum. In this chapter, an improved particle swarm optimization algorithm is proposed based on increasing the disturbance of LDW, which can change the inertia weight coefficient and the probability of random individuals. The simulation results verify the superior performance of the algorithm. In a stable 24-bus system, a few wind turbines and hydropower units are used to establish the optimal dispatching model, which can provide theoretical and applied basis for the power system with large new energy connected to the grid. On the basis of the traditional economic scheduling problem, the thermal power unit is added as the objective function of infrequently changing the output force to ensure the economic operation of the thermal power unit. According to the different water consumption of hydropower stations, the dispatching days are divided into three cases, and the dispatching is carried out according to the characteristics of the hydropower stations in the flood season, the average water period and the low water period. An improved PSO algorithm based on dynamic adaptive variable inertial weight coefficient and small probability is used to solve the problem. The simulation results show that the hydropower station plays a regulating role in the wind-fire water complex energy power system, and the amount of water available to generate electricity in the hydropower station determines the regulating capacity of the hydropower station in the composite energy power system.
【學(xué)位授予單位】：昆明理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TM73

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 姚燕芬;;芻議清潔能源優(yōu)先的風(fēng)-水-火電力系統(tǒng)聯(lián)合優(yōu)化調(diào)度[J];通訊世界;2014年17期

2 南曉強(qiáng);李群湛;趙元哲;邱大強(qiáng);;計(jì)及風(fēng)電預(yù)測(cè)可信度的經(jīng)濟(jì)調(diào)度及輔助決策方法[J];電力系統(tǒng)自動(dòng)化;2013年19期

3 楊曉萍;王文堅(jiān);薛斌;黃強(qiáng);;風(fēng)、火、水電短期聯(lián)合優(yōu)化調(diào)度研究[J];水力發(fā)電學(xué)報(bào);2013年04期

4 師洪濤;楊靜玲;丁茂生;王金梅;;基于小波—BP神經(jīng)網(wǎng)絡(luò)的短期風(fēng)電功率預(yù)測(cè)方法[J];電力系統(tǒng)自動(dòng)化;2011年16期

5 徐曼;喬穎;魯宗相;;短期風(fēng)電功率預(yù)測(cè)誤差綜合評(píng)價(jià)方法[J];電力系統(tǒng)自動(dòng)化;2011年12期

6 孫榮富;張濤;梁吉;;電網(wǎng)接納風(fēng)電能力的評(píng)估及應(yīng)用[J];電力系統(tǒng)自動(dòng)化;2011年04期

7 洪翠;林維明;溫步瀛;;風(fēng)電場(chǎng)風(fēng)速及風(fēng)電功率預(yù)測(cè)方法研究綜述[J];電網(wǎng)與清潔能源;2011年01期

8 李俊芳;張步涵;謝光龍;李妍;毛承雄;;基于灰色模型的風(fēng)速-風(fēng)電功率預(yù)測(cè)研究[J];電力系統(tǒng)保護(hù)與控制;2010年19期

9 張曉花;趙晉泉;陳星鶯;;節(jié)能減排多目標(biāo)機(jī)組組合問(wèn)題的模糊建模及優(yōu)化[J];中國(guó)電機(jī)工程學(xué)報(bào);2010年22期

10 王曉蘭;王明偉;;基于小波分解和最小二乘支持向量機(jī)的短期風(fēng)速預(yù)測(cè)[J];電網(wǎng)技術(shù);2010年01期

相關(guān)博士學(xué)位論文 前1條

1 高芳;智能粒子群優(yōu)化算法研究[D];哈爾濱工業(yè)大學(xué);2008年

相關(guān)碩士學(xué)位論文 前7條

1 楊佳俊;含風(fēng)電場(chǎng)的電力系統(tǒng)動(dòng)態(tài)經(jīng)濟(jì)調(diào)度研究[D];山東大學(xué);2014年

2 都晨;基于模糊聚類(lèi)的GA-BP風(fēng)電場(chǎng)短期風(fēng)速及功率預(yù)測(cè)的研究[D];南京理工大學(xué);2013年

3 丁蕾;吐魯番地區(qū)農(nóng)村飲水安全工程水質(zhì)現(xiàn)狀調(diào)查與分析[D];新疆醫(yī)科大學(xué);2012年

4 王晨暉;基于多目標(biāo)的水庫(kù)優(yōu)化調(diào)度研究[D];華中科技大學(xué);2012年

5 張燕;含風(fēng)電場(chǎng)的電力系統(tǒng)動(dòng)態(tài)經(jīng)濟(jì)調(diào)度[D];華北電力大學(xué);2011年

6 龐峰;模擬退火算法的原理及算法在優(yōu)化問(wèn)題上的應(yīng)用[D];吉林大學(xué);2006年

7 趙文博;基于動(dòng)態(tài)規(guī)劃法地級(jí)電力企業(yè)購(gòu)網(wǎng)電量的分配[D];華北電力大學(xué)（北京）;2005年

，

本文編號(hào)：2075483