利用回歸和人工神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)城市雨水徑流總磷模型的有效性

摘要： 利用人工神經(jīng)網(wǎng)絡(luò)(ANN),探討在不無(wú)監(jiān)測(cè)系統(tǒng)的集水區(qū)城市降水質(zhì)量預(yù)測(cè)的適用性.預(yù)測(cè)使用常規(guī)的氣候和地理數(shù)據(jù)集,通過(guò)構(gòu)建背景傳播的神經(jīng)網(wǎng)絡(luò)和回歸聯(lián)合模型,克服利用逐步回歸的方法對(duì)數(shù)據(jù)進(jìn)行分析時(shí)違背獨(dú)立數(shù)據(jù)假設(shè)的問(wèn)題.研究通過(guò)交叉驗(yàn)證用于確定停止降水時(shí)間為輸入變量參數(shù),利用地區(qū)平均濃度(EMC)作為獨(dú)立的變量,構(gòu)建的模型比用負(fù)荷量構(gòu)建的模型更精確.數(shù)據(jù)域和輸入變量的選擇對(duì)回歸模型的準(zhǔn)確性也有較大影響.但計(jì)算效率、動(dòng)量和隱節(jié)點(diǎn)數(shù)目的選擇等因素,對(duì)人工神經(jīng)網(wǎng)絡(luò)模型準(zhǔn)確性的影響較小.同時(shí),回歸和人工神經(jīng)網(wǎng)絡(luò)模型的降水質(zhì)量預(yù)測(cè)結(jié)果十分相似,但在不無(wú)監(jiān)測(cè)系統(tǒng)的集水區(qū)域城市降水質(zhì)量的預(yù)測(cè)方面,回歸模型更有實(shí)效性. Abstract： This paper investigates the applicability of using artificial neural networks (ANNs) to predict urban stormwater quality at unmonitored catchments.Back-propagation neural networks and regression models were constructed using a set of general climatic and geographic data.Violation of the assumption of data independence lead to the inclusion of insignificant variables when the data was analysed using stepwise regression.To overcome this problem, cross validation was used to determine when to cease input variable entry.Models constructed using event mean concentration (EMC) as the dependent variable were more accurate than those using load.The data domain and selection of input variables had a significant effect upon the accuracy of the regression models.Whereas the choice of learning rates, momentum and number of hidden nodes had an insignificant effect upon the accuracy of the ANN models.Regression and ANN models yielded similar predictions.However, the efficiency of the regression models made them a more pragmatic approach for predicting urban stormwater quality at unmonitored sites. 作 者： 梅·D    西瓦庫(kù)瑪·M    MAY D    SIVAKUMAR M   作者單位： 澳大利亞伍龍貢大學(xué)環(huán)境工程,新南威爾士州,伍龍貢,2500  期 刊： 沈陽(yáng)化工學(xué)院學(xué)報(bào)    Journal： JOURNAL OF SHENYANG INSTITUTE OF CHEMICAL TECHNOLOGY  年，卷(期)： 2010, 24(1)  分類號(hào)： X820.4  關(guān)鍵詞： 人工神經(jīng)網(wǎng)絡(luò)    城市雨水水質(zhì)    總磷    模型    Keywords： artificial neural networks    urban stormwater quality    phosphorus    model   

【利用回歸和人工神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)城市雨水徑流總磷模型的有效性】相關(guān)文章：

物理《靜電的利用和防止》教案08-26

高一化學(xué)上學(xué)期氮和磷教案08-26

《城市人類垃圾的分類及循環(huán)利用》教案（精選10篇）05-08

《人類對(duì)細(xì)菌和真菌的利用》的教學(xué)反思12-13

開(kāi)發(fā)和利用實(shí)驗(yàn)教學(xué)資源04-28

營(yíng)養(yǎng)物質(zhì)的吸收和利用的生物教案08-26

含磷高的食物有哪些03-20

基于遺傳算法的地下水位動(dòng)態(tài)預(yù)測(cè)雙線性模型09-08

碳源和硝態(tài)氮濃度對(duì)反硝化聚磷的影響及ORP的變化規(guī)律12-28

雨水和谷雨有什么區(qū)別之處02-19