發(fā)布時(shí)間：2018-11-21 08:43

【摘要】：在低溶解氧(DO)狀態(tài)下,以城市生活污水為研究對(duì)象,將神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)的方法應(yīng)用到污水處理過(guò)程中,建立了基于神經(jīng)網(wǎng)絡(luò)的氨氧化過(guò)程預(yù)測(cè)控制模型,預(yù)測(cè)并控制污水處理氨氧化過(guò)程.該模型分為兩部分,一是根據(jù)在線pH值變化預(yù)測(cè)氨氧化結(jié)束時(shí)間,其相關(guān)系數(shù)R值為0.9985;二是根據(jù)在線pH值實(shí)時(shí)預(yù)測(cè)氨氮濃度,R值為0.9083.試驗(yàn)結(jié)果表明該模型預(yù)測(cè)精度高、可控性好,具有較好的適應(yīng)性和穩(wěn)定性,對(duì)實(shí)現(xiàn)并穩(wěn)定短程硝化以及促進(jìn)主流工藝(厭氧氨氧化)有重要的指導(dǎo)和借鑒意義.
[Abstract]:Under the condition of low dissolved oxygen (DO), the neural network prediction method was applied to the process of wastewater treatment, and the predictive control model of ammonia oxidation process based on neural network was established. Predict and control the ammonia oxidation process of wastewater treatment. The model is divided into two parts: one is to predict the end time of ammonia oxidation according to the change of on-line pH value, the correlation coefficient R is 0.9985; the other is real-time prediction of ammonia nitrogen concentration based on on-line pH value, R value is 0.9083. The experimental results show that the model has high prediction accuracy, good controllability, good adaptability and stability, and has important guidance and reference significance for realizing and stabilizing short-cut nitrification and promoting mainstream process (anaerobic ammonia oxidation).
【作者單位】： 北京工業(yè)大學(xué)北京市水質(zhì)科學(xué)與水環(huán)境恢復(fù)工程重點(diǎn)實(shí)驗(yàn)室;中國(guó)人民大學(xué)環(huán)境學(xué)院;
【基金】：國(guó)家自然科學(xué)基金(51508561) 北京市委組織部青年拔尖團(tuán)隊(duì) 北京市優(yōu)秀人才培養(yǎng)資助計(jì)劃
【分類號(hào)】：X703

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 顧升波;王淑瑩;彭永臻;;短程深度脫氮中試工藝的低溫啟動(dòng)和維持[J];環(huán)境科學(xué);2013年08期

2 李澤兵;李軍;李妍;馬家軒;王曉毅;;短程硝化反硝化技術(shù)研究進(jìn)展[J];給水排水;2011年09期

3 田建平;曹東衛(wèi);李海楠;;LM-BP神經(jīng)網(wǎng)絡(luò)在于橋水庫(kù)水質(zhì)預(yù)測(cè)中的應(yīng)用[J];水利信息化;2010年04期

4 王淑瑩;黃惠s，

本文編號(hào)：2346465