諸飛 俞阿龍



摘 要: 針對工廠污水水質(zhì)評價(jià)準(zhǔn)確性低、實(shí)時(shí)性差等問題,提出一種將改進(jìn)遺傳算法(GA)和BP神經(jīng)網(wǎng)絡(luò)相結(jié)合對工廠污水的水質(zhì)進(jìn)行評價(jià)的方法。GA?BP神經(jīng)網(wǎng)絡(luò)不僅具有BP神經(jīng)網(wǎng)絡(luò)的非線性映射能力,還具有遺傳算法的全局搜索能力。采用自適應(yīng)算法對GA的交叉率和變異率進(jìn)行改進(jìn),用GA優(yōu)化BP的權(quán)值和閾值,將最優(yōu)權(quán)值和閾值送給BP神經(jīng)網(wǎng)絡(luò)進(jìn)行訓(xùn)練、預(yù)測,并與傳統(tǒng)BP進(jìn)行比較。實(shí)驗(yàn)結(jié)果表明,改進(jìn)的GA?BP神經(jīng)網(wǎng)絡(luò)無論是收斂性、準(zhǔn)確性還是實(shí)時(shí)性都優(yōu)于傳統(tǒng)BP網(wǎng)絡(luò)。該方法用于污水水質(zhì)評價(jià)具有應(yīng)用推廣價(jià)值。
關(guān)鍵詞: 工廠污水; 水質(zhì)分類; 改進(jìn)GA; BP神經(jīng)網(wǎng)絡(luò); 污水監(jiān)測; 自適應(yīng)算法
中圖分類號: TN711?34; TP183 文獻(xiàn)標(biāo)識碼: A 文章編號: 1004?373X(2018)11?0133?06
Research on factory sewage monitoring system based on improved GA?BP neural network
ZHU Fei1, YU Along2
(1. School of Physics and Electronic?Electrical Engineering, Ningxia University, Yinchuan 750021, China;
2. School of Physics and Electronic Electrical Engineering, Huaiyin Normal University, Huaian 223300, China)
Abstract: A water quality evaluation method of factory sewage is proposed to solve the problems of low accuracy and poor timeliness of factory sewage quality assessment, which is based on the combination of improved GA and BP neural network. The GA?BP neural network has the nonlinear mapping ability of BP neural network, and global search ability of genetic algorithm. The adaptive algorithm is used to improve the crossover rate and mutation rate of GA. The GA is used to optimize the weight and threshold of BP neural network. The optimal weight and threshold are sent to BP neural network for training, prediction and comparison with traditional BP neural network. The experimental results show that the convergence, accuracy and timeliness of the improved GA?BP neural network are better than those of traditional BP neural network. The method has a certain application and promotion value for water quality evaluation of sewage.
Keywords: factory sewage; water quality classification; improved GA; BP neural network; sewage monitoring; adaptive algorithm
目前,我國工廠污水處理技術(shù)較發(fā)達(dá)國家落后,污水水質(zhì)評價(jià)準(zhǔn)確性差,有很多主觀的判斷成分,導(dǎo)致未達(dá)標(biāo)的污水隨意排放[1]。水體污染日趨廣泛和嚴(yán)重,加劇了水資源的短缺狀況。從環(huán)境保護(hù)的角度來說,對工廠排放污水的水質(zhì)進(jìn)行準(zhǔn)確有效的評價(jià)顯得尤為重要。
針對工廠污水排放具有不確定性,處理過程多變、非線性、時(shí)變、強(qiáng)隨機(jī)等特點(diǎn)[2],科學(xué)有效地對工廠污水水質(zhì)進(jìn)行準(zhǔn)確有效的評價(jià)是本文研究的內(nèi)容。對水質(zhì)評價(jià)的方法很多,文獻(xiàn)[3]使用單因子污染指數(shù)法對丹江口水庫水質(zhì)進(jìn)行評價(jià);文獻(xiàn)[4]利用主成分分析法對黃河口及鄰近水域水質(zhì)進(jìn)行評價(jià);文獻(xiàn)[5]采用人工神經(jīng)網(wǎng)絡(luò)法對大沽河濕地海水水質(zhì)進(jìn)行評價(jià)。……