国产日韩欧美一区二区三区三州_亚洲少妇熟女av_久久久久亚洲av国产精品_波多野结衣网站一区二区_亚洲欧美色片在线91_国产亚洲精品精品国产优播av_日本一区二区三区波多野结衣 _久久国产av不卡

?

水電站下游魚類產(chǎn)卵場(chǎng)水溫的人工神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型

2018-03-09 05:29柳海濤孫雙科鄭鐵剛李廣寧
關(guān)鍵詞:松花江水文站吉林

柳海濤,孫雙科,鄭鐵剛,李廣寧

?

水電站下游魚類產(chǎn)卵場(chǎng)水溫的人工神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型

柳海濤,孫雙科,鄭鐵剛,李廣寧

(流域水循環(huán)模擬與調(diào)控國(guó)家重點(diǎn)實(shí)驗(yàn)室,中國(guó)水利水電科學(xué)研究院,北京 100038)

豐滿電站下游松花江水文站河段分布有一系列魚類產(chǎn)卵場(chǎng),電站擬通過分層取水調(diào)控下泄水溫,改善下游魚類生存環(huán)境。該文基于大量實(shí)測(cè)數(shù)據(jù)分析,建立了松花江站水溫的人工神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型,通過輸入上游吉林水文站的水溫與流量,以及地區(qū)氣象條件,可計(jì)算出下游松花江站2日后的水溫變化。根據(jù)中長(zhǎng)期天氣預(yù)報(bào)數(shù)據(jù)與電站泄流計(jì)劃,采用該模型通過2日遞推的方法,可預(yù)測(cè)出下游魚類產(chǎn)卵場(chǎng)的水溫變化過程。運(yùn)用2006-2013年實(shí)測(cè)數(shù)據(jù)對(duì)網(wǎng)絡(luò)模型進(jìn)行訓(xùn)練,然后對(duì)2014年松花江站水溫變化過程進(jìn)行計(jì)算,計(jì)算值與實(shí)測(cè)值的變化過程甚為吻合,相關(guān)系數(shù)為0.992,水溫平均誤差為0.51 ℃。在水溫生態(tài)調(diào)度運(yùn)行期間,根據(jù)產(chǎn)卵場(chǎng)水溫變化的預(yù)報(bào)數(shù)據(jù),可適當(dāng)調(diào)控電站下泄水溫,保持適宜的魚類產(chǎn)卵條件。

模型;水文;魚;水電站;生態(tài)調(diào)度;產(chǎn)卵場(chǎng);水溫;人工神經(jīng)網(wǎng)絡(luò)

0 引 言

大型水庫在確保興利目標(biāo)的同時(shí),通過合理的生態(tài)環(huán)境調(diào)度,可以補(bǔ)償水庫對(duì)河流生態(tài)系統(tǒng)的不利影響[1-2]。從工程實(shí)踐的角度來看,水庫生態(tài)環(huán)境調(diào)度的主要內(nèi)容包括:維持確保下游河道生態(tài)蓄水量[3-4],創(chuàng)造生態(tài)洪水脈沖[5-6];改善下游河道水體水質(zhì),提高下游河道自凈能力[7];通過選擇性取水,緩解低溫水下泄的不利影響[8];通過機(jī)械與建筑設(shè)施,改變水庫下泄水流的溶解氧濃度,改善下游生態(tài)環(huán)境等[9]。其中,在水溫生態(tài)調(diào)度方面,主要研究?jī)?nèi)容包括:通過對(duì)下游河段進(jìn)行生態(tài)調(diào)查與分析,結(jié)合興利目標(biāo)確定合理的人造洪峰過程[10];然后通過數(shù)學(xué)模型與物理模型試驗(yàn)的手段[11-12],確定庫區(qū)水溫分層及泄水調(diào)度方式與下泄水溫的對(duì)應(yīng)關(guān)系;通過原型觀測(cè)資料分析,確定水庫下泄水溫與下游長(zhǎng)距離河道水溫之間的關(guān)系[13]。通過上述研究,制定合理的水溫生態(tài)調(diào)度方案,減輕水庫低溫水下泄對(duì)下游河道生態(tài)系統(tǒng)的不利影響。

吉林豐滿電站水庫為水溫分層型水庫,目前正在實(shí)施全面治理(重建)工程,其中上游120 m處舊壩被部分保留,形成前置擋墻,以提取表層溫水,同時(shí)左岸原有三期電廠提取深層冷水,通過改變引水比例,實(shí)現(xiàn)調(diào)控下泄水溫的功能。工程完建后,在下游魚類繁殖季節(jié),擬通過合理的運(yùn)行調(diào)度,形成一定歷時(shí)的漲水-升溫過程,促進(jìn)當(dāng)?shù)佤~類生長(zhǎng)。為此,本文通過研究水庫運(yùn)行調(diào)度對(duì)下游魚類產(chǎn)卵場(chǎng)水溫的影響,建立相關(guān)的水溫實(shí)時(shí)預(yù)報(bào)模型,用以調(diào)控下泄水溫,保持適宜的魚類生長(zhǎng)環(huán)境。

目前對(duì)于河流水文條件的預(yù)測(cè)方法主要有3類,第1類是采用統(tǒng)計(jì)分析的方法,建立預(yù)測(cè)因子與目標(biāo)因子的相關(guān)關(guān)系[14-17],該類方法計(jì)算簡(jiǎn)便,對(duì)實(shí)測(cè)資料要求較低,可用于河流水文條件的實(shí)時(shí)預(yù)報(bào),不足之處在于其無法考慮包含多個(gè)預(yù)測(cè)因子的復(fù)雜非線性問題。第2類是采用數(shù)學(xué)模型進(jìn)行預(yù)報(bào)的方法,通過建立研究對(duì)象的數(shù)學(xué)模型[18-20],對(duì)目標(biāo)因子進(jìn)行預(yù)測(cè)計(jì)算。該類方法計(jì)算精度高,物理意義明確,能夠精確描述目標(biāo)因子的變化規(guī)律,預(yù)測(cè)時(shí)間序列可以精確到數(shù)小時(shí)甚至分鐘。不足之處在于,對(duì)實(shí)測(cè)水文氣象數(shù)據(jù)要求較高,同時(shí)運(yùn)算復(fù)雜,需要事先通過大量計(jì)算,形成相應(yīng)的數(shù)據(jù)庫,然后由調(diào)度系統(tǒng)根據(jù)實(shí)際條件插值求解目標(biāo)因子,如此又會(huì)產(chǎn)生新的誤差[21]。第3類是采用人工神經(jīng)網(wǎng)絡(luò)進(jìn)行預(yù)測(cè)的方法[22-26],該方法通過網(wǎng)絡(luò)模型對(duì)實(shí)測(cè)數(shù)據(jù)進(jìn)行學(xué)習(xí),可以模擬多個(gè)預(yù)測(cè)因子之間的相互作用對(duì)目標(biāo)因子的影響,模型運(yùn)算速度較快,可以被水庫調(diào)度系統(tǒng)實(shí)時(shí)調(diào)用。不足之處在于,隨著問題復(fù)雜性增加,網(wǎng)絡(luò)學(xué)習(xí)過程難度也會(huì)加大,此時(shí)需要對(duì)網(wǎng)絡(luò)結(jié)構(gòu)與學(xué)習(xí)算法進(jìn)行分析與優(yōu)化[27-28]。

鑒于神經(jīng)網(wǎng)絡(luò)在快速模擬復(fù)雜非線性問題方面的優(yōu)勢(shì),本文擬采用人工神經(jīng)網(wǎng)絡(luò),對(duì)松花江站水溫進(jìn)行預(yù)報(bào)。通過相關(guān)性分析確定影響松花江站水溫的外部水文氣象因子;分析建立反映水溫變化機(jī)理的基本表達(dá)形式,由此確定網(wǎng)絡(luò)模型的預(yù)測(cè)因子與目標(biāo)因子;通過對(duì)實(shí)測(cè)數(shù)據(jù)資料進(jìn)行分析與學(xué)習(xí),確定網(wǎng)絡(luò)預(yù)報(bào)模型的基本結(jié)構(gòu)與學(xué)習(xí)矩陣;最后,針對(duì)網(wǎng)絡(luò)預(yù)報(bào)模型的精度與適應(yīng)性進(jìn)行驗(yàn)證,為進(jìn)一步實(shí)際應(yīng)用打下基礎(chǔ)。

1 研究區(qū)域概況

豐滿電站庫首、下游水文站及魚類產(chǎn)卵場(chǎng)的位置見見圖1。下泄水流經(jīng)過摻混先到達(dá)20 km處吉林水文站,然后流經(jīng)160 km的河段,到達(dá)下游松花江水文站,該站附近分布有龍王廟、榆樹十八盤、飲馬河口等主要魚類產(chǎn)卵場(chǎng)。

圖1 研究區(qū)域概況圖

通過分析豐滿電站、吉林水文站與松花江水文站的水溫實(shí)測(cè)資料,發(fā)現(xiàn)電站下泄水溫與吉林站水溫之間具有較強(qiáng)的相關(guān)性,并已建立經(jīng)驗(yàn)關(guān)系來預(yù)測(cè)未來吉林站水溫變化,而對(duì)于吉林站至松花江站之間河段,兩者水溫相關(guān)性較差。究其原因在于,該段河道水流歷時(shí)較長(zhǎng),水體與周邊環(huán)境進(jìn)行熱交換,至產(chǎn)卵場(chǎng)時(shí)水溫發(fā)生顯著改變。因此,需要根據(jù)流域水文氣象資料,對(duì)下游松花江站水溫進(jìn)行實(shí)時(shí)預(yù)測(cè)。

2 水文與氣象資料分析

2.1 松花江站水溫的外部影響因子

根據(jù)水文部門提供的資料,針對(duì)下游魚類繁殖季節(jié),整理得到吉林站與松花江站在2006-2014年份5?8月的水溫-流量逐日數(shù)據(jù)。對(duì)于該區(qū)域氣象資料,則采用國(guó)家氣象局?jǐn)?shù)據(jù)中心提供的長(zhǎng)春國(guó)家基準(zhǔn)氣候站的逐日氣象資料,參考變量包括平均氣壓、氣溫、相對(duì)濕度,以及平均風(fēng)速、最大風(fēng)速、最大風(fēng)速風(fēng)向、日照時(shí)數(shù)等。限于篇幅,具體數(shù)據(jù)這里不再列出。

表1為各種外部影響因子與下游松花江水文站水溫的相關(guān)性分析結(jié)果。其中,吉林站水溫、長(zhǎng)春站平均氣溫、相對(duì)濕度、日照時(shí)數(shù)等4個(gè)變量與松花江站水溫之間呈正比關(guān)系,而吉林站流量、長(zhǎng)春站風(fēng)速與松花江站水溫呈反比關(guān)系。一般地,相關(guān)系數(shù)絕對(duì)值0.1~0.3被認(rèn)為弱相關(guān)[29],從表中計(jì)算結(jié)果來看,吉林站水溫,吉林站流量,長(zhǎng)春站平均氣溫、相對(duì)濕度、平均風(fēng)速的相關(guān)系數(shù)均大于0.3。長(zhǎng)春站日照時(shí)數(shù)的相關(guān)系數(shù)平均值為0.219,但在部分年份達(dá)到或超過了0.3,故予以保留。長(zhǎng)春站氣壓與最大風(fēng)速的相關(guān)系數(shù)明顯小于0.3,故略去不計(jì)。由此確定了6個(gè)影響松花江站水溫的外部因子,分別為吉林站水溫,吉林站流量,長(zhǎng)春站平均氣溫、相對(duì)濕度、平均風(fēng)速和日照時(shí)數(shù)。

表1 外部影響因子與松花江站水溫之間的相關(guān)系數(shù)

2.2 松花江站水溫預(yù)報(bào)的預(yù)見期分析

吉林站與松花江站相距約160 km,根據(jù)河道生態(tài)流量估算,下泄水流從吉林站到達(dá)松花江站需要1-3日。圖2為2010年5-8月兩站流量逐日變化過程線。由圖2可知,吉林站與松花江站流量過程線不重合,若將松花江站流量采用推后2日數(shù)據(jù),則兩站的變化規(guī)律基本吻合。進(jìn)一步對(duì)比分析其余年份水文資料,同樣表明松花江站的水文因子相對(duì)吉林站均存在2日左右的預(yù)見期。因此在水溫預(yù)報(bào)模型中各變量均具有時(shí)間屬性。

圖2 2010年5?8月吉林站與松花江站同日流量過程

2.3 水溫預(yù)報(bào)模型的基本物理關(guān)系

在松花江水文站水溫預(yù)報(bào)過程中,需要考慮上游吉林站水文條件、水流沿程與外界的熱交換,以及松花江站本地初始水溫等影響因素。由此可建立松花江站水溫變化的基本物理關(guān)系

式(1)為一個(gè)2日遞推公式,反映了計(jì)算時(shí)段內(nèi)水文與氣象條件對(duì)于松花江站水溫變化的影響。將式(1)右端7個(gè)參數(shù)作為神經(jīng)網(wǎng)絡(luò)模型的預(yù)測(cè)因子,將松花江站2日水溫差作為目標(biāo)因子,構(gòu)建實(shí)測(cè)數(shù)據(jù)樣本,對(duì)神經(jīng)網(wǎng)絡(luò)進(jìn)行學(xué)習(xí)與驗(yàn)證,從而建立起松花江站水溫預(yù)報(bào)模型。

3 水溫預(yù)報(bào)的人工神經(jīng)網(wǎng)絡(luò)

3.1 神經(jīng)網(wǎng)絡(luò)基本結(jié)構(gòu)

目前用于模擬多維非線性系統(tǒng)的神經(jīng)網(wǎng)絡(luò),主要有BP網(wǎng)絡(luò)和RBF(radial basis function)網(wǎng)絡(luò),其中BP網(wǎng)絡(luò)在學(xué)習(xí)過程中采用全局逼近的方法,因此具有較好的泛化能力,即當(dāng)預(yù)測(cè)數(shù)據(jù)超出學(xué)習(xí)范圍時(shí),該網(wǎng)絡(luò)模型仍能得到合理的結(jié)果。然而,該網(wǎng)絡(luò)面對(duì)包含較多數(shù)據(jù)誤差的非線性問題時(shí),學(xué)習(xí)能力不足,甚至無法收斂。相比而言,RBF網(wǎng)絡(luò)在學(xué)習(xí)過程中采用局部逼近的方法,對(duì)于各種復(fù)雜非線性問題,具有較好的學(xué)習(xí)能力,但其泛化能力有所不足,因此,需對(duì)學(xué)習(xí)后的網(wǎng)絡(luò)模型進(jìn)行檢驗(yàn),或者通過改進(jìn)神經(jīng)元激發(fā)函數(shù),以提高其泛化能力[30]。

本項(xiàng)研究中,實(shí)測(cè)水文與氣象資料均包含較多數(shù)據(jù)誤差,故采用RBF神經(jīng)網(wǎng)絡(luò)進(jìn)行研究。通用RBF網(wǎng)絡(luò)的基本結(jié)構(gòu)如圖3所示,其預(yù)測(cè)因子數(shù)和目標(biāo)因子數(shù)可根據(jù)實(shí)際問題的因子分析進(jìn)行調(diào)整,隱含層中的神經(jīng)元個(gè)數(shù)可通過樣本數(shù)據(jù)的學(xué)習(xí)加以確定。

注: 為預(yù)測(cè)因子;為神經(jīng)元激發(fā)函數(shù);為目標(biāo)因子;與為加權(quán)系數(shù)。

由式(1)可知,松花江站水溫預(yù)報(bào)模型的預(yù)測(cè)因子數(shù)為7個(gè),目標(biāo)因子數(shù)為1個(gè),則輸入樣本數(shù)據(jù)的神經(jīng)元激發(fā)函數(shù)可表示為

這樣,網(wǎng)絡(luò)的目標(biāo)因子可以用下式表示

式中為樣本編號(hào),為加權(quán)系數(shù)向量。由于樣本的目標(biāo)因子已知,當(dāng)隱含層的神經(jīng)元激發(fā)函數(shù)矩陣確定后,由式(3)可以得到線性輸出層的權(quán)向量

式中dD分別是第組樣本的目標(biāo)值和對(duì)應(yīng)的網(wǎng)絡(luò)計(jì)算值。

3.2 神經(jīng)網(wǎng)絡(luò)的學(xué)習(xí)與模型預(yù)報(bào)方法

由于樣本點(diǎn)數(shù)據(jù)分布復(fù)雜且數(shù)值噪聲較大,采用傳統(tǒng)的聚類方法無法區(qū)分噪聲、邊界點(diǎn)和核心對(duì)象,同時(shí)依靠經(jīng)驗(yàn)人為指定神經(jīng)元的個(gè)數(shù)也有一定困難。對(duì)此,本文采用的辦法是先從任一神經(jīng)元開始訓(xùn)練,對(duì)比其網(wǎng)絡(luò)計(jì)算值D與樣本目標(biāo)值d之間的絕對(duì)誤差,然后將產(chǎn)生最大誤差值的樣本輸入向量x作為新的神經(jīng)元中心t,重新學(xué)習(xí)并判斷式(5)中的誤差函數(shù)值是否小于目標(biāo)誤差,如不滿足,則繼續(xù)增加新的神經(jīng)元,直到滿足誤差要求或達(dá)到最大神經(jīng)元數(shù)為止。計(jì)算過程中,神經(jīng)元的鄰閾值可取實(shí)測(cè)數(shù)據(jù)的最大分布范圍。為便于移植,采用FORTRAN語言編碼。

4 神經(jīng)網(wǎng)絡(luò)模型水溫預(yù)報(bào)結(jié)果分析

4.1 神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型計(jì)算驗(yàn)證

將2006-2013年5-8月的吉林水文站當(dāng)日水溫、流量,長(zhǎng)春站次日平均氣象條件,松花江站同日水溫差等7個(gè)預(yù)報(bào)參數(shù)作為樣本預(yù)測(cè)向量,將松花江站2日水溫差值作為樣本目標(biāo)向量,對(duì)神經(jīng)網(wǎng)絡(luò)進(jìn)行訓(xùn)練,學(xué)習(xí)過程的水溫平均收斂誤差為0.1 ℃。然后,運(yùn)用訓(xùn)練后的網(wǎng)絡(luò)模型,采用2日遞推的方法,對(duì)2014年5-8月松花江站水溫變化過程進(jìn)行預(yù)報(bào)。具體方法如下:從5月1日與2日開始,根據(jù)吉林站的當(dāng)日水溫流量、長(zhǎng)春站次日氣象數(shù)據(jù)、松花江站當(dāng)日水溫差,分別預(yù)報(bào)得到松花江水文站5月3日與4日的水溫差,并得到相應(yīng)的水溫絕對(duì)值,然后以此為初始水溫,重復(fù)上述計(jì)算,進(jìn)一步計(jì)算得到松花江站5月5日與6日的水溫差及絕對(duì)值,以此遞推…,可得到5-8月整個(gè)時(shí)段內(nèi)松花江站水溫的變化過程,具體計(jì)算結(jié)果參見圖4。

圖4 2014年5-8月松花江站水溫變化過程的實(shí)測(cè)值與預(yù)測(cè)值對(duì)比

由于神經(jīng)網(wǎng)絡(luò)模型采用2日遞推的方法,推求產(chǎn)卵場(chǎng)的水溫變化過程,如果神經(jīng)網(wǎng)絡(luò)模型不能正確計(jì)算水流沿程熱交換影響,則隨著時(shí)間的推移,水溫預(yù)報(bào)誤差逐漸累積,將會(huì)產(chǎn)生巨大的偏差。從圖4中計(jì)算結(jié)果可知,在推求長(zhǎng)時(shí)間序列水溫變化過程中,本文模型顯示出了良好的收斂性。水溫實(shí)測(cè)值與預(yù)測(cè)值的相關(guān)關(guān)系繪制于圖5。計(jì)算表明,兩者相關(guān)系數(shù)達(dá)到0.992,水溫均方誤差為0.51 ℃,相對(duì)誤差小于8%,表明本文模型在計(jì)算精度上可以滿足水庫實(shí)時(shí)調(diào)度的要求。

圖5 2014年5-8月松花江站水溫實(shí)測(cè)值與預(yù)測(cè)值對(duì)應(yīng)關(guān)系

4.2 神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型應(yīng)用分析

以2011年5-8月的水文氣象條件為例,吉林站與松花江站的現(xiàn)狀水溫逐日平均變化過程見圖6a。分析表明,5-8月松花江站水溫較之吉林站平均升高約4.6 ℃。根據(jù)豐滿電站重建工程分層取水的水溫研究成果[31],通過分層取水調(diào)控下泄水溫,5-8月份吉林站水溫較之現(xiàn)狀可整體升高3.2 ℃。上述條件下,采用神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型,計(jì)算得到松花江站的水溫變化過程見圖6b。

圖6 2011年5?8月松花江站水溫的時(shí)間變化過程

分析表明,5-8月松花江站水溫比吉林站水溫平均升高約3.3 ℃左右,較之現(xiàn)狀下降1.3 ℃。究其原因在于,當(dāng)上游吉林站下泄水溫升高后,水體與大氣溫差減小,兩者熱交換量降低,使得溫升值有所下降。表明該模型能夠較好地反映上游電站水溫調(diào)控對(duì)于下游魚類產(chǎn)卵場(chǎng)水溫的影響。

5 討 論

目前對(duì)于神經(jīng)網(wǎng)絡(luò)在河流水溫實(shí)時(shí)預(yù)報(bào)中的應(yīng)用,主要應(yīng)當(dāng)考慮以下4個(gè)方面的問題:首先是影響因子的選擇問題。一些學(xué)者在研究過程中,僅采用了部分影響因子作為預(yù)測(cè)因子,來構(gòu)建神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型,造成預(yù)報(bào)誤差較大[24]。這種情況下,針對(duì)神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)與學(xué)習(xí)算法進(jìn)行反復(fù)優(yōu)化與調(diào)整,對(duì)模型預(yù)報(bào)精度改善仍然有限[28]。本文研究中,首先針對(duì)水文氣象因子進(jìn)行相關(guān)性分析,篩選出目標(biāo)水溫的主要影響因子,然后再建立神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型,可較完整地反應(yīng)預(yù)測(cè)因子之間的相互作用對(duì)目標(biāo)因子的影響,避免產(chǎn)生較大的系統(tǒng)誤差。

第二是神經(jīng)網(wǎng)絡(luò)模擬對(duì)象的表達(dá)形式問題。以往研究中,均是在確定影響因子之后,直接建立類似黑箱的預(yù)報(bào)模型[32],該類模型對(duì)于物理現(xiàn)象的內(nèi)部機(jī)理不具有解釋性,在實(shí)際應(yīng)用中可能會(huì)出現(xiàn)較大偏差,甚至不合理的解答。在本文研究中,模擬對(duì)象是一個(gè)160 km河段在2日內(nèi)的熱交換過程,為此,先通過分析建立了能夠反映其物理過程的基本表達(dá)形式,然后再將其轉(zhuǎn)化為神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型。該模型計(jì)算驗(yàn)證表明,在2014年長(zhǎng)系列水溫過程預(yù)報(bào)中,盡管下泄流量、下泄水溫、氣象條件不斷變化,模型預(yù)測(cè)水溫保持了良好的精度與收斂性。

第三是神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)因子的時(shí)間屬性問題。在實(shí)際問題研究中,預(yù)測(cè)因子與目標(biāo)因子通常并不符合時(shí)間對(duì)應(yīng)關(guān)系,例如降雨-徑流的預(yù)報(bào)問題,從降雨到形成徑流需要一定的滯后時(shí)間,此為模型固有的預(yù)報(bào)期[26]。本文研究中,通過對(duì)吉林站與松花江站之間流量過程線進(jìn)行分析,確認(rèn)兩站之間存在2日預(yù)報(bào)期,同時(shí)由于實(shí)測(cè)資料均為逐日平均數(shù)據(jù),模型預(yù)測(cè)因子中的上游吉林站下泄水溫取當(dāng)日值,氣象條件則為第2日內(nèi)平均值,下游松花江站水溫則取第3日值,這樣構(gòu)建了2日預(yù)報(bào)模型。

第四是神經(jīng)網(wǎng)絡(luò)模型的空間擴(kuò)展性問題。以往采用神經(jīng)網(wǎng)絡(luò)方法,一般僅針對(duì)某一條河流的局地水溫進(jìn)行預(yù)報(bào)。當(dāng)需要預(yù)測(cè)水溫的站點(diǎn)較多時(shí),一般采用2種處理方法[33],一種方法是針對(duì)每個(gè)站點(diǎn)分別建立預(yù)報(bào)模型,然后組合在一起,該方法適用于預(yù)測(cè)站點(diǎn)相對(duì)獨(dú)立,目標(biāo)因子關(guān)聯(lián)度較小的情況。另一種方法是在現(xiàn)有網(wǎng)絡(luò)的基礎(chǔ)上,通過增加目標(biāo)因子的數(shù)量,實(shí)現(xiàn)多個(gè)站點(diǎn)的水溫預(yù)測(cè)。該方法適用于站點(diǎn)相互關(guān)聯(lián),且預(yù)報(bào)因子的數(shù)據(jù)條件基本相同的情況。本文研究中,采用了一種通用神經(jīng)網(wǎng)絡(luò),其目標(biāo)因子可不限于松花江站水溫,若要預(yù)測(cè)河道其他站點(diǎn)的水溫,只需要增加新站點(diǎn)目標(biāo)因子及其實(shí)測(cè)水溫資料,然后進(jìn)一步學(xué)習(xí)形成新的預(yù)報(bào)模型。

另外,由于實(shí)測(cè)水文氣象資料中包含一定測(cè)量誤差,同時(shí)神經(jīng)網(wǎng)絡(luò)的結(jié)構(gòu)與學(xué)習(xí)方法也造成計(jì)算誤差,使得本文模型具有0.5 ℃左右的平均誤差。在實(shí)際工程應(yīng)用中,應(yīng)當(dāng)通過連續(xù)多日預(yù)報(bào),根據(jù)一段時(shí)間內(nèi)產(chǎn)卵場(chǎng)的水溫平均值,調(diào)控下泄水溫。

6 結(jié) 論

通過對(duì)吉林水文站與松花江水文站的水文資料,以及長(zhǎng)春站的氣象資料進(jìn)行分析處理,運(yùn)用RBF(radial basis function)神經(jīng)網(wǎng)絡(luò),建立了松花江站水溫的預(yù)報(bào)模型。該預(yù)報(bào)模型中除了考慮吉林水文站的水溫以外,增加了流量、氣溫、濕度等氣象條件,提高了松花江站的水溫預(yù)測(cè)精度。運(yùn)用該神經(jīng)網(wǎng)絡(luò)對(duì)2006-2013年的水文、氣象實(shí)測(cè)數(shù)據(jù)進(jìn)行學(xué)習(xí),然后針對(duì)2014年實(shí)測(cè)條件下的松花江站水溫過程進(jìn)行預(yù)測(cè),計(jì)算值與實(shí)測(cè)值的規(guī)律較為吻合,相關(guān)系數(shù)達(dá)0.992,水溫預(yù)測(cè)平均誤差為0.51 ℃。由于河道水流從吉林水文站至松花江水文站,需要?dú)v時(shí)2日左右,在神經(jīng)網(wǎng)絡(luò)的預(yù)測(cè)因子中,吉林站的水溫與流量采用了當(dāng)日值,長(zhǎng)春站氣象條件采用了第2值,而松花江站水溫則為第3日預(yù)測(cè)值,因此該模型具有2日預(yù)報(bào)期。在實(shí)際運(yùn)行期間,可根據(jù)天氣預(yù)報(bào)中長(zhǎng)期數(shù)據(jù)與電站泄水計(jì)劃,采用2日遞推的方法預(yù)測(cè)未來時(shí)段下游魚類產(chǎn)卵場(chǎng)的水溫變化過程,實(shí)現(xiàn)下泄水溫與流量的動(dòng)態(tài)調(diào)控,可保持適宜的魚類產(chǎn)卵條件。

[1] 駱文廣,楊國(guó)錄,宋云浩,等. 再議水庫生態(tài)環(huán)境調(diào)度[J]. 水科學(xué)進(jìn)展,2016,27(2):317-326.

Luo Wenguang, Yang Guolu, Song Yunhao, et al. Reconsidering ecologically-sound environmental operations at reservoirs[J]. Advances in Water Science, 2016, 27(2): 317-326. (in Chinese with English abstract)

[2] 徐楊,常福宣,陳進(jìn),等. 水庫生態(tài)調(diào)度研究綜述[J]. 長(zhǎng)江科學(xué)院院報(bào),2008,25(6):33-37.

Xu Yang, Chang Fuxuan, Chen Jin, et al. Review of research on ecological operation of reservoir[J]. Journal of Yangtze River Scientific Research Institute, 2008, 25(6): 33-37. (in Chinese with English abstract)

[3] Hughes D A, Hannart P A. desktop model used to provide an initial estimate of the ecological instream flow requirements of rivers in South Africa[J]. Journal of Hydrology, 2003, 270(3): 167-181.

[4] Harman C, Stewardson M. Optimizing dam release rules to meet environmental flow targets[J]. River Research and Applications, 2005, 21(2/3): 113-129.

[5] Junk W J. Amazonian Floodplains: Their Ecology, Present and Potential Use [J]. Revue Dhydrobiologie Tropicale, 1982, 15(4):285-301.

[6] Kingsford R T, Curtin A L, Porter J. Water flow on cooper creek in arid Australia determine ‘boom’ and ‘bust’ periods for waterbirds[J]. Biological Conservation, 1999, 8(8): 231-248.

[7] Naiman R J, Turner M G. A future perspective on north American’s fresh water ecosystems[J]. Ecological Applications, 2000, 10: 958-966.

[8] Lovich J, Melis T S. The state of the Colorado river ecosystem in Grand Canyon: Lessons from 10 year of adaptive ecosystem management [J]. International Journal of River Basin Management, 2007, 5(3): 207-221.

[9] Higgins J M, Brock W G. Overview of reservoir release improvements at 20 TVA dams[J]. Journal of Energy Engineering, 1999, 125(1): 1-17.

[10] 尹正杰,楊春花,許繼軍. 考慮不同生態(tài)流量約束的梯級(jí)水庫生態(tài)調(diào)度初步研究[J]. 水力發(fā)電學(xué)報(bào),2013,32(3):66-70,81.

Yin Zhengjie, Yang Chunhua, Xu Jijun. Ecological regulation of cascade dams in the lower Jinsha river under different ecological flow constraints[J]. Journal of Hydroelectric Engineering, 2013, 32(3): 66-70, 81. (in Chinese with English abstract)

[11] 柳海濤,孫雙科,王曉松,等. 大型深水庫分層取水水溫模型試驗(yàn)研究[J]. 水力發(fā)電學(xué)報(bào),2012,31(1):129-134.

Liu Haitao, Sun Shuangke, Wang Xiaosong, et al. Water temperature model test for a multi-level power intake of large deep reservoir [J]. Journal of Hydroelectric Engineering, 2012, 31(1): 129-134. (in Chinese with English abstract)

[12] 鄭鐵剛,孫雙科,柳海濤,等. 大型分層型水庫下泄水溫對(duì)取水高程敏感性分析研究[J]. 水利學(xué)報(bào),2015,46(6):714-722,731.

Zheng Tiegang, Sun Shuangke, Liu Haitao, et al. Sensibility analysis of intake elevation on water temperature discharged from a large thermal stratified reservoir[J]. Journal of Hydraulic Engineering, 2015, 46(6): 714-722, 731. (in Chinese with English abstract)

[13] 蒲靈,李克峰,莊春義,等. 天然河流水溫變化規(guī)律的原型觀測(cè)研究[J]. 四川大學(xué)學(xué)報(bào):自然科學(xué)版,2006,43(3):615-617.

Pu Ling, Li Kefeng, Zhuang Chunyi, et al. Field measurement of distribution of water temperature’s change along a river[J]. Journal of Sichuan University(Natural Science Edition), 2006, 43(3): 615-617. (in Chinese with English abstract)

[14] Yang D, Peterson A. River water temperature in relation to local air temperature in the Mackenzie and Yukon basins[J]. Arctic, 2017, 70(1): 47-58.

[15] Jackson F L, Fryer R J, Hannah D M, et al. A spatio-temporal statistical model of maximum daily river temperatures to inform the management of Scotland’s Atlantic salmon rivers under climate change[J]. Science of the Total Environment, 2018, 612: 1543-1558.

[16] Garner G, Malcolm I A, Sadler J P, et al. The role of riparian vegetation density, channel orientation and water velocity in determining river temperature dynamics[J]. Journal of Hydrology, 2017, 553: 471-485.

[17] Laanaya F, St-Hilaire A, Gloaguen E. Water temperature modelling: Comparison between the generalized additive model, logistic, residuals regression and linear regression models[J]. International Association of Scientific Hydrology Bulletin, 2017, 62(7): 1078-1093.

[18] 鞠鵬飛,陳青生,陳衛(wèi). 基于支流水溫對(duì)漢江干流河道水溫影響的數(shù)值模擬[J]. 水電能源科學(xué),2013(5):84-87.

Ju Pengfei, Chen Qingsheng, Chen Wei. Numerical simulation of influence of water temperature of tributary on Hanjiang main river[J]. Water Resources & Power, 2013(5): 84-87. (in Chinese with English abstract)

[19] 脫友才,周晨陽,梁瑞峰,等. 水電開發(fā)對(duì)大渡河瀑布溝以下河段的水溫影響[J]. 水科學(xué)進(jìn)展,2016,27(2):299-306.

Tuo Youcai, Zhou Chenyang, Liang Ruifeng, et al. Influence of hydroelectric development on water temperature downstream the Pubugou hydropower station of the Dadu River[J]. Advances in Water Science, 2016, 27(2): 299-306. (in Chinese with English abstract)

[20] 陳衛(wèi),熊君. 丹江口水庫下游河道水溫恢復(fù)過程二維數(shù)值模擬[J]. 西北水電,2016(6):73-76.

Chen Wei, Xiong Jun. 2D numerical simulation of recovery process of water temperature at downstream river course of danjiangkou reservoir[J]. Northwest Hydropower, 2016(6): 73-76. (in Chinese with English abstract)

[21] 李玉榮,許銀山,閔要武,等. 三峽水庫實(shí)時(shí)調(diào)度水文氣象預(yù)報(bào)應(yīng)用風(fēng)險(xiǎn)及控制[J]. 人民長(zhǎng)江,2015,46(23):15-19.

Li Yurong, Xu Yinshan, Min Yaowu, et al. Research on risk and control of hydro-meteorological forecasting application in real-time operation of Three Gorges Reservoir[J]. Yangtze River, 2015, 46(23): 15-19. (in Chinese with English abstract)

[22] Wambua R M. Prediction of Missing Hydro-Meteorological Data Series Using Artificial Neural Networks (ANN) for Upper Tana River Basin, Kenya[J]. International Journal of Climatology, 2016, 31(2):159-161.

[23] Hadzima-Nyarko M, Rabi A, ?perac M. Implementation of artificial neural networks in modeling the water-air temperature relationship of the river drava[J]. Water Resources Management, 2014, 28(5): 1379-1394.

[24] Deweber J T, Wagner T. A regional neural network ensemble for predicting mean daily river water temperature[J]. Journal of Hydrology, 2015, 517(2): 187-200.

[25] Chithra N R, Thampi S G, Surapaneni S, et al. Prediction of the likely impact of climate change on monthly mean maximum and minimum temperature in the Chaliyar river basin, India, using ANN-based models[J]. Theoretical & Applied Climatology, 2015, 121(3/4): 581-590.

[26] Astray G, Soto B, Lopez D, et al. Application of transit data analysis and artificial neural network in the prediction of discharge of Lor River, NW Spain[J]. Water Science & Technology, 2016, 73(7): 1756-1767.

[27] Piccolroaz S, Calamita E, Majone B, et al. Prediction of river water temperature: A comparison between a new family of hybrid models and statistical approaches[J]. Hydrological Processes, 2016, 30(21): 3901-3917.

[28] Piotrowski A P, Napiorkowski M J, Napiorkowski J J, et al. Comparing various artificial neural network types for water temperature prediction in rivers[J]. Journal of Hydrology, 2015, 529(1): 302-315.

[29] 黃鸞玉,黎小正,謝宗升,等. 來賓紅水河珍稀魚類保護(hù)區(qū)水質(zhì)指標(biāo)相關(guān)性研究[J]. 環(huán)境保護(hù)科學(xué), 2017, 43(2):108-111.

Huang Luanyu, Li Xiaozheng, Xie Zongsheng, et al. Study of the correlations between water quality indicators of rare fish protected area in Laibin Section of Hong River [J] Environmental Protection Science, 2017, 43(2):108-111.

[30] 柳海濤,孫雙科,劉之平,等. 泄洪霧化預(yù)測(cè)的人工神經(jīng)網(wǎng)絡(luò)方法探討[J]. 水利學(xué)報(bào),2005,36(10):1241-1245. Liu Haitao, Sun Shuangke, Liu Zhiping, et al. Atomization prediction based on artificial neural networks for flood releasing of high dams[J]. Journal of Hydraulic Engineering, 2005, 36(10): 1241-1245. (in Chinese with English abstract)

[31] 鄭鐵剛,孫雙科,柳海濤,等. 大型分層型水庫下泄水溫對(duì)取水高程敏感性分析研究[J]. 水利學(xué)報(bào),2015,46(6):714-722.

Zheng Tiegang, Sun Shuangke, Liu Haitao, et al. Sensibility analysis of intake elevation on water temperature discharged from a large thermal stratified reservoir [J]. Journal of Hydraulic Engineering, 2015, 46(6): 714-722. (in Chinese with English abstract)

[32] Liu W C, Chen W B. Prediction of water temperature in a subtropical subalpine lake using an artificial neural network and three-dimensional circulation models[J]. Computers & Geosciences, 2012, 45: 13-25.

[33] Risley J C, Roehl E A, Conrads P A. Estimating Water Temperatures in Small Streams in Western Oregon Using Neural Network Models[R]. Center for Integrated Data Analytics Wisconsin Science Center, 2003.

Prediction of water temperature regulation for spawning sites at downstream of hydropower station by artificial neural network method

Liu Haitao, Sun Shuangke, Zheng Tiegang, Li Guangning

(100038)

In this study, the water temperature regulation were carried out through the selective intake facilities in Fengman Hydropower station to improve the downstream living environment. The power plant released flow first reaches Jilin hydrologic station at 20 km downstream, and then through the 160 km long reach, arrive at the Songhuajiang hydrologic station, where there are a series of spawning sites of black carp, grass carp, silver carp, etc. The field data analysis showed that, there was a strong correlation between the water temperature of the power plant and Jilin Station, so the empirical relationship has been established based on the statistical analysis of the measured data in earlier research. However, there was obvious difference and poor correlation between the water temperature of Jilin Station and Songhuajiang station. The main reason was that the heat exchange between the channel water and the surrounding environment led to a significant change in water temperature. Firstly, by analyzing the correlation coefficients between all the hydrological and meteorological factors with the water temperature of Songhuajiang station, the six external influence factors were identified, including the flow and water temperature of Jilin Station, and the air temperature, relative humidity, wind speed and sunshine duration of Changchun meteorological station. Then, based on the field data, the water temperature prediction model of Songhuajiang station was established by using a RBF (radial basis function) neural network, which can automatically select the sample vectors with maximum error as a new neuron until to finally reach the required precision. It took about 2 days to flow from Jilin to Songhuajiang station, so the model predictors had temporal and spatial attributes. The flow and water temperature of Jilin station should be the values of the first day, the climate conditions of Changchun station were of the next day, and the water temperature of Songhuajiang station was of the third day. Therefore the neural network model actually reflected a heat exchange process within two days. According to the medium or long term weather forecast data and power station discharge plan, the neural network model can be used to predict the time course of the water temperature at the spawning sites by using the above two day recursive method. The model was trained by the field data in 2006 - 2013, and to predict the temperature time course in 2014, the time variation of the calculated and measured water temperatures were in good agreement, the average deviation was 0.51 ℃, and the correlation coefficient was 0.992. In May 2010 to August, for example, the average temperature increased from Jilin to Songhuajiang station was 4.6 ℃. When the released water temperature upstream rose 3.2 ℃ by regulation, because of the decrease of the heat exchange between the channel water and the surrounding environment, the temperature increased between the two stations dropped 3.3 ℃. It was proved that this model can better reflect the influence of heat exchange along the river on the water temperature of downstream spawning field. During the water temperature regulation, the water temperature at spawning sites will be predicted, and the releasing discharge of power plant is adjusted properly, to provide suitable spawning conditions.

models; hydrology; fish; hydropower station; ecological operation; spawning sites; water temperature; artificial neural network

2017-09-02

2018-01-15

國(guó)家重點(diǎn)研發(fā)規(guī)劃(2016YFC0401708);國(guó)家自然基金項(xiàng)目(51679262)

柳海濤,博士,教授級(jí)高級(jí)工程師,主要研究方向?yàn)樯鷳B(tài)水力學(xué)與水工水力學(xué)。Email:htliou@163.com.

10.11975/j.issn.1002-6819.2018.04.022

S931.1

A

1002-6819(2018)-04-0185-07

柳海濤,孫雙科,鄭鐵剛,李廣寧. 水電站下游魚類產(chǎn)卵場(chǎng)水溫的人工神經(jīng)網(wǎng)絡(luò)預(yù)報(bào)模型[J]. 農(nóng)業(yè)工程學(xué)報(bào),2018,34(4):185-191.doi:10.11975/j.issn.1002-6819.2018.04.022 http://www.tcsae.org

Liu Haitao, Sun Shuangke, Zheng Tiegang, Li Guangning. Prediction of water temperature regulation for spawning sites at downstream of hydropower station by artificial neural network method[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(4): 185-191. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2018.04.022 http://www.tcsae.org

猜你喜歡
松花江水文站吉林
松花江
金沙江壩下水文站中泓浮標(biāo)系數(shù)穩(wěn)定性分析
A Spring Coat for Sarah
SL流量計(jì)在特殊河段的應(yīng)用——以河源水文站為例
吉林卷
西雙版納州水文站網(wǎng)合理布設(shè)及測(cè)驗(yàn)方法探析
長(zhǎng)江干流沙市水文站汛期流量變化趨勢(shì)分析
吉林卷
松花江緑石硯
松花江上