本文關(guān)鍵詞：基于功率預(yù)測的風(fēng)電并網(wǎng)優(yōu)化調(diào)度研究　出處：《北京交通大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 小世界網(wǎng)絡(luò) 粒子群優(yōu)化算法 BP神經(jīng)網(wǎng)絡(luò) 風(fēng)電功率預(yù)測 動態(tài)經(jīng)濟調(diào)度

【摘要】：風(fēng)能是一種可再生清潔能源。大力發(fā)展風(fēng)力發(fā)電對改善能源結(jié)構(gòu)、應(yīng)對氣候變化和能源安全問題具有十分重要的意義。然而風(fēng)能具有波動性、間歇性和不可控性等特點,大規(guī)模風(fēng)電接入電網(wǎng)對電力系統(tǒng)可靠運行和經(jīng)濟調(diào)度帶來了巨大挑戰(zhàn)。因此,研究風(fēng)電功率預(yù)測和含風(fēng)電場的電力系統(tǒng)優(yōu)化調(diào)度為提高風(fēng)電利用水平和智能電網(wǎng)建設(shè)具有重要的經(jīng)濟意義和實際價值。本文基于含風(fēng)電的電力系統(tǒng)經(jīng)濟調(diào)度問題,展開了如下研究： 1.將具有較大聚類系數(shù)和較小平均路徑的NW型小世界網(wǎng)絡(luò)作為粒子群優(yōu)化算法的拓撲結(jié)構(gòu),提出了小世界鄰域粒子群優(yōu)化(SW-PSO)算法。仿真分析表明,該算法在求解高維尋優(yōu)問題上具有搜索速度快、尋優(yōu)精度高的優(yōu)點,適宜于優(yōu)化神經(jīng)網(wǎng)絡(luò)預(yù)測模型和求解大規(guī)模非線性數(shù)學(xué)規(guī)劃問題。 2.研究了基于SW-PSO算法的BP神經(jīng)網(wǎng)絡(luò)預(yù)測模型。采用風(fēng)電場實際數(shù)據(jù)建立風(fēng)機風(fēng)速-功率曲線,利用NWP信息實現(xiàn)了日前風(fēng)電功率預(yù)測。由于NWP數(shù)據(jù)自身存在誤差影響最終風(fēng)電功率的預(yù)測精度,將風(fēng)電功率歷史數(shù)據(jù)與NWP數(shù)據(jù)相結(jié)合實現(xiàn)了風(fēng)電功率超短期滾動預(yù)測。 3.在風(fēng)電功率預(yù)測的基礎(chǔ)上,將風(fēng)電預(yù)測誤差和穿透功率計入系統(tǒng)旋轉(zhuǎn)備用當(dāng)中,建立了含風(fēng)電場電力系統(tǒng)動態(tài)經(jīng)濟調(diào)度數(shù)學(xué)模型。該模型是一個含有多約束條件的大規(guī)模非線性數(shù)學(xué)規(guī)劃問題,利用SW-PSO算法對其進行求解,并采用調(diào)整策略修正粒子保證了其在可行域中飛行尋優(yōu)。通過算例驗證了SW-PSO算法在求解風(fēng)電并網(wǎng)優(yōu)化調(diào)度問題方面的有效性和可行性。 4.將機會約束規(guī)劃引入到含風(fēng)電場的電力系統(tǒng)動態(tài)經(jīng)濟調(diào)度模型當(dāng)中,以概率形式描述相關(guān)約束條件,建立了風(fēng)電并網(wǎng)電力系統(tǒng)隨機優(yōu)化調(diào)度模型。算例表明應(yīng)用蒙特卡洛隨機模擬技術(shù)和SW-PSO算法對求解電力系統(tǒng)隨機優(yōu)化調(diào)度問題是行之有效的,能夠為決策者提供既滿足電網(wǎng)安全穩(wěn)定運行又符合新能源消納需求的合理規(guī)劃方案。
[Abstract]:Wind energy is a renewable and clean energy. The development of wind power to improve the energy structure, has very important significance to deal with the problem of climate change and energy security. However, the wind is volatile, intermittent and uncontrollable, large scale wind power integration has brought great challenges for reliable operation and economic dispatch of power system. Therefore, study on wind power forecasting and wind power system scheduling to improve wind power has important economic significance and practical value of utilization and development of smart power grids. The power system with wind power based on the economic dispatch problem, are done as follows:
1. will have a relatively large clustering coefficient and small average path of the NW type small world network as the topology of particle swarm optimization algorithm, proposed the small world neighborhood particle swarm optimization (SW-PSO) algorithm. Simulation results show that the algorithm has fast search speed in solving high dimension optimization problems, find the advantages of high precision, suitable for the model and solving large-scale nonlinear programming problem to predict the optimization of neural network.
2. of the BP neural network forecasting model based on SW-PSO algorithm. The actual wind data set fan speed power curve, the day before the wind power prediction using NWP NWP data information. Due to existence of wind power prediction accuracy of the final error, the wind power history data combined with NWP data the ultra short term wind power forecast.
3. based on wind power prediction, the prediction error and wind power penetration into the spinning reserve, established in wind power system dynamic economic dispatch mathematical model. The model is a multi constraint conditions of large-scale nonlinear programming problems, using SW-PSO algorithm to solve it, and the modified particle adjustment strategy to ensure its flight search in the feasible region. The examples demonstrate that the SW-PSO algorithm in solving the optimization problem of wind power integration in terms of effectiveness and feasibility.
4. the chance constrained programming is introduced into the power system dynamic economic dispatch model with wind farms, with probability to describe the relevant constraints, a stochastic optimal scheduling of wind power grid power system model. Examples show that the application of Monte Carlo simulation technique and SW-PSO algorithm is effective for solving stochastic optimal scheduling problem of power system, can provide both to meet the safe and stable operation of power grid and reasonable planning of the new energy consumptive demand for decision makers.

【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TM614

【參考文獻】

相關(guān)期刊論文 前10條

1 江岳文;陳沖;溫步瀛;;含風(fēng)電場的電力系統(tǒng)機組組合問題隨機模擬粒子群算法[J];電工技術(shù)學(xué)報;2009年06期

2 雷亞洲;與風(fēng)電并網(wǎng)相關(guān)的研究課題[J];電力系統(tǒng)自動化;2003年08期

3 陳海焱;陳金富;段獻忠;;含風(fēng)電場電力系統(tǒng)經(jīng)濟調(diào)度的模糊建模及優(yōu)化算法[J];電力系統(tǒng)自動化;2006年02期

4 王彩霞;魯宗相;;風(fēng)電功率預(yù)測信息在日前機組組合中的應(yīng)用[J];電力系統(tǒng)自動化;2011年07期

5 茆美琴;周松林;蘇建徽;;基于脊波神經(jīng)網(wǎng)絡(luò)的短期風(fēng)電功率預(yù)測[J];電力系統(tǒng)自動化;2011年07期

6 吳棟梁;王揚;郭創(chuàng)新;劉毅;高振興;;電力市場環(huán)境下考慮風(fēng)電預(yù)測誤差的經(jīng)濟調(diào)度模型[J];電力系統(tǒng)自動化;2012年06期

7 張雪雯;李艷君;;基于自調(diào)節(jié)粒子群算法的電力系統(tǒng)經(jīng)濟負荷分配[J];電網(wǎng)技術(shù);2006年18期

8 杜穎;盧繼平;李青;鄧穎玲;;基于最小二乘支持向量機的風(fēng)電場短期風(fēng)速預(yù)測[J];電網(wǎng)技術(shù);2008年15期

9 劉純;范高鋒;王偉勝;戴慧珠;;風(fēng)電場輸出功率的組合預(yù)測模型[J];電網(wǎng)技術(shù);2009年13期

10 袁鐵江;晁勤;吐爾遜·伊不拉音;李義巖;;面向電力市場的含風(fēng)電電力系統(tǒng)的環(huán)境經(jīng)濟調(diào)度優(yōu)化[J];電網(wǎng)技術(shù);2009年20期

相關(guān)博士學(xué)位論文 前4條

1 楊志凌;風(fēng)電場功率短期預(yù)測方法優(yōu)化的研究[D];華北電力大學(xué)（北京）;2011年

2 高陽;風(fēng)電場風(fēng)電功率預(yù)測方法研究[D];沈陽農(nóng)業(yè)大學(xué);2011年

3 王揚;風(fēng)電短期預(yù)測及其并網(wǎng)調(diào)度方法研究[D];浙江大學(xué);2011年

4 周瑋;含風(fēng)電場的電力系統(tǒng)動態(tài)經(jīng)濟調(diào)度問題研究[D];大連理工大學(xué);2010年

，

本文編號：1422056