本文選題：感潮河段 + 走航式ADCP��； 參考：《海洋通報(bào)》2016年05期

【摘要】：針對(duì)現(xiàn)有基于徑向基函數(shù)的感潮河段局域時(shí)空流場(chǎng)模型構(gòu)建中節(jié)點(diǎn)選取缺乏依據(jù)的現(xiàn)狀,利用走航式ADCP資料研究了聚類算法、貪婪算法和梯度算法等3種節(jié)點(diǎn)位置選取方法,結(jié)合實(shí)驗(yàn)將3種不同節(jié)點(diǎn)位置選取方法下獲得的余流場(chǎng)和潮流場(chǎng)與調(diào)和分析方法的結(jié)果進(jìn)行對(duì)比分析,并采用交叉驗(yàn)證法評(píng)估節(jié)點(diǎn)選取方法構(gòu)建模型。結(jié)果表明,3種節(jié)點(diǎn)選取方法各有特點(diǎn),在實(shí)際應(yīng)用中應(yīng)根據(jù)不同的流場(chǎng)特點(diǎn)及工程需要選擇合適的方法構(gòu)建流場(chǎng)。
[Abstract]:In this paper , three kinds of node selection methods , such as clustering algorithm , greedy algorithm and gradient algorithm , are studied in this paper , which is based on radial basis function , and three kinds of node location selection methods , such as clustering algorithm , greedy algorithm and gradient algorithm , are studied . The results show that the selection methods of three different nodes are different and the flow field should be constructed according to different characteristics of flow field and engineering needs .

【作者單位】： 武漢大學(xué)測(cè)繪學(xué)院;武漢大學(xué)動(dòng)機(jī)學(xué)院;
【基金】：國(guó)家自然科學(xué)基金(41576107;41376109;41176068) 國(guó)家重大專項(xiàng)(2016YFB0501703)
【分類號(hào)】：P731.2
，

本文編號(hào)：1891531