【摘要】：南海海島地處深海,遠(yuǎn)離大陸,島上用電長(zhǎng)期依賴(lài)柴油發(fā)電,海島軍民用電及日常用水保障困難。南海海域遼闊,西沙、南沙、東沙諸島蘊(yùn)含有豐富的風(fēng)能資源。隨著風(fēng)力發(fā)電技術(shù)日益成熟和小型風(fēng)機(jī)的快速發(fā)展,在南海海島應(yīng)用風(fēng)力發(fā)電技術(shù)已具有現(xiàn)實(shí)可行性。本文根據(jù)東沙島、永興島、珊瑚島和太平島地面氣象觀測(cè)數(shù)據(jù),對(duì)南海海島進(jìn)行了風(fēng)能資源評(píng)估和風(fēng)力發(fā)電機(jī)組選型的研究。論文首先收集、整理南海東沙島、永興島、珊瑚島和太平島4個(gè)地面氣象觀測(cè)站1996年至2012年間的氣象數(shù)據(jù)；然后運(yùn)用卡方(χ2)檢驗(yàn)和均方根(RMSE)檢驗(yàn)對(duì)威布爾分布(Weibull)和瑞利分布(Reyleigh)進(jìn)行檢驗(yàn),發(fā)現(xiàn)Weibull分布能更準(zhǔn)確地刻畫(huà)南海海島實(shí)際數(shù)據(jù)的統(tǒng)計(jì)特性；利用兩參數(shù)Weibull分布評(píng)估了各站點(diǎn)的風(fēng)速和風(fēng)能密度等風(fēng)能資源情況；最后,根據(jù)1996年至2012年間氣象數(shù)據(jù),利用TMA 10kW、Jacobs 10kW、BWC Excel-S10kW和富蘭德30kW(FL-30)風(fēng)電機(jī)組的風(fēng)速功率輸出特性曲線,計(jì)算各站點(diǎn)的發(fā)電量,并基于容量系數(shù)法估算了容量系數(shù)值。評(píng)估發(fā)現(xiàn)FL-30的CF值最大,發(fā)電量最多,可選擇FL-30為適用的典型風(fēng)機(jī)；還結(jié)合Weibull分布和富蘭德30kW風(fēng)機(jī)的風(fēng)速功率輸出特性曲線估算出的數(shù)據(jù),分析了所選風(fēng)機(jī)在東沙島、永興島等4個(gè)氣象站點(diǎn)風(fēng)電出力的季節(jié)性規(guī)律。分析1996年至2012年間的風(fēng)速數(shù)據(jù),提取了每日最大風(fēng)速,分析了南海四島的最大風(fēng)情況,發(fā)現(xiàn)這4個(gè)站點(diǎn)的大風(fēng)頻次各不相同。利用PP圖(Probability Plot)、QQ圖(Quantile-Quantile Plot)重現(xiàn)期水平函數(shù)圖和密度函數(shù)圖診斷和檢驗(yàn)基于廣義Pareto分布的閾值模型的合理性,發(fā)現(xiàn)利用閾值模型能準(zhǔn)確地?cái)M合所調(diào)研島嶼最大風(fēng)速特性；然后利用剩余函數(shù)圖選取了閾值u,再根據(jù)廣義P areto分布模型、利用極大似然估計(jì)法計(jì)算各站點(diǎn)五十年一遇最大風(fēng)速。數(shù)值仿真結(jié)果表明永興島、珊瑚島五十年一遇極大風(fēng)值速分別為42.25m/s和的42.04m/s,適合選用Ⅱ型風(fēng)機(jī)；東沙島五十年一遇極大風(fēng)值速35.20m/s,適合選用Ⅲ型風(fēng)機(jī)；太平島五十年一遇極大風(fēng)45.38m/s,推薦選用Ⅰ型風(fēng)電機(jī)組。
[Abstract]:The South China Sea Island is located in the deep sea, far away from the mainland. The island relies on diesel for a long time to generate electricity, and it is difficult for the island's military and people to use electricity and daily water. The South China Sea is vast, Xisha, Nansha and Dongsha islands are rich in wind energy resources. With the increasing maturity of wind power generation technology and the rapid development of small wind turbines, it is feasible to apply wind power generation technology to the South China Sea islands. Based on the meteorological observation data of Dongsha Island, Yongxing Island, Coral Island and Taiping Island, the wind energy resource evaluation and wind turbine selection of the South China Sea island are studied in this paper. Firstly, the meteorological data of four surface meteorological observatories, Dongsha Island, Yongxing Island, Coral Island and Taiping Island, in the South China Sea from 1996 to 2012 are collected and collated. Then the Chi-square (蠂 2) test and root mean square (RMSE) test were used to test the Weibull distribution (Weibull) and Rayleigh distribution (Reyleigh). It was found that the Weibull distribution could more accurately describe the statistical characteristics of the actual data of the South China Sea island. The wind energy resources such as wind speed and wind energy density are evaluated by using two-parameter Weibull distribution. Finally, based on the meteorological data from 1996 to 2012, the wind power output curves of TMA 10kW, Jacobs 10kW, BWC Excel-S10kW and Fuland 30kW (FL-30) wind turbines are used to calculate the output of each station. Based on the capacity coefficient method, the value of the capacity coefficient is estimated. It is found that FL-30 has the largest CF value and the most power generation, and FL-30 can be selected as the applicable typical fan. The seasonal rule of wind power output in Dongsha Island and Yongxing Island of the selected wind turbine in Dongsha Island and Yongxing Island is analyzed based on the data estimated from the Weibull distribution and the wind speed power output curve of the Fuland 30kW fan. The wind speed data from 1996 to 2012 are analyzed, the daily maximum wind speed is extracted, and the maximum wind speed of the four islands in the South China Sea is analyzed. It is found that the gale frequency of these four stations is different. The PP graph (Probability Plot), QQ graph (Quantile-Quantile Plot) level function graph and density function graph are used to diagnose and verify the rationality of the threshold model based on the generalized Pareto distribution. It is found that the threshold model can accurately fit the maximum wind velocity characteristics of the investigated islands. Then the threshold u is selected by the residual function graph and the maximum wind speed is calculated by maximum likelihood estimation method according to the generalized P-areto distribution model. The results of numerical simulation show that the maximum wind velocity of Yongxing Island and Coral Island is 42.04m / s of 42.25m/s and 42.04m / s, respectively, which is suitable for the selection of type 鈪，

本文編號(hào)：2444448