污水处理中出水水质COD在线预测仿真
|
摘要
针对当前方法在线预测污水处理中出水水质COD浓度值时存在预测结果与实际值偏差较大、的问题,提出了基于最小二乘支持向量机的污水处理出水水质COD预测模型。将污水处理中出水水质COD光谱信号做多层分解,在各个分解尺度下设置适当阈值,将分解信号中小于该设置阈值的信号小波系数做置零处理,并采用小波逆变换实现污水处理中出水水质COD光谱信号重构,完成小波去噪;对去噪后的数据进行降维处理;采用最小二乘支持向量机法建立水样吸光度和水质COD浓度预测模型,利用该模型分析和预测污水处理中出水水质COD。仿真结果表明,所建模型实现了污水处理中出水水质COD在线预测,且具有预测准确性较高、稳定性较强、收敛速度较快的优点。
In order to solve the problem that there is a large deviation between the prediction results and the actual COD concentration in the on-line prediction of effluent quality COD concentration in sewage treatment,a COD prediction model based on least square support vector machine( LS support vector machine) is proposed. The COD spectral signal of effluent quality in sewage treatment is decomposed in multiple layers,the appropriate threshold is set at each decomposition scale,and the wavelet coefficient of the decomposed signal is reduced to zero,and the wavelet inverse transform is used to reconstruct the effluent quality COD spectral signal in sewage treatment to complete wavelet denoising; dimension reduction of the data after denoising is carried out; and the least square support direction is adopted. The prediction model of water sample absorbance and water quality COD concentration was established by measuring machine method. The model was used to analyze and predict the effluent quality COD. in sewage treatment. The simulation results show that the model realizes the on-line prediction of effluent quality COD in sewage treatment,and has the advantages of high prediction accuracy,strong stability and fast convergence speed.
In order to solve the problem that there is a large deviation between the prediction results and the actual COD concentration in the on-line prediction of effluent quality COD concentration in sewage treatment,a COD prediction model based on least square support vector machine( LS support vector machine) is proposed. The COD spectral signal of effluent quality in sewage treatment is decomposed in multiple layers,the appropriate threshold is set at each decomposition scale,and the wavelet coefficient of the decomposed signal is reduced to zero,and the wavelet inverse transform is used to reconstruct the effluent quality COD spectral signal in sewage treatment to complete wavelet denoising; dimension reduction of the data after denoising is carried out; and the least square support direction is adopted. The prediction model of water sample absorbance and water quality COD concentration was established by measuring machine method. The model was used to analyze and predict the effluent quality COD. in sewage treatment. The simulation results show that the model realizes the on-line prediction of effluent quality COD in sewage treatment,and has the advantages of high prediction accuracy,strong stability and fast convergence speed.
引文
[1]吕蒙，胡映天，高亚，等．基于水样类型识别的光谱COD测量方法[J]．光谱学与光谱分析，2017,37(12):3797-3802.
[2]刘梦，伯鑫，孟凡琳，等．2015年中国城镇污水处理厂达标排放评估[J]．环境工程，2017,35(10):77-81.
[3]黄健，于孝坤，张华，等．p H对污水生物处理中COD荧光法快速表征的影响[J]．环境污染与防治，2017,39(12):1285-1288.
[4]曹生现，韩宇，李房玉．基于图像技术的化学需氧量检测研究[J]．工业水处理，2017,37(5):90-94.
[5]仲洋，夏凤毅，廉继尧．紫外-近红外光谱法测定废水COD含量[J]．环境工程学报，2017,11(2):1300-1304.
[6]胡博，叶峻宏，贾衍琰，等．超声-分光光度法快速测定水样中的化学需氧量[J]．应用化工，2017,46(1):190-193.
[7]袁红春，吕苏娜．水产养殖水质异常优化预测仿真研究[J]．计算机仿真，2017,34(12):447-450.
[8]徐玉妃，杨昆，罗毅，等．滇池草海水质等级预测模型研究[J]．水生态学杂志，2018,39(1):1-8.
[9]张双，周集体．高盐有机化工废水中COD与TOC的相关性[J]．化工环保，2018,38(1):122-126.
[10]孙伯寅，董国庆，张荣．支持向量机在水源水化学耗氧量预测中的应用[J]．环境与健康杂志，2016,33(6):544-547.
[2]刘梦，伯鑫，孟凡琳，等．2015年中国城镇污水处理厂达标排放评估[J]．环境工程，2017,35(10):77-81.
[3]黄健，于孝坤，张华，等．p H对污水生物处理中COD荧光法快速表征的影响[J]．环境污染与防治，2017,39(12):1285-1288.
[4]曹生现，韩宇，李房玉．基于图像技术的化学需氧量检测研究[J]．工业水处理，2017,37(5):90-94.
[5]仲洋，夏凤毅，廉继尧．紫外-近红外光谱法测定废水COD含量[J]．环境工程学报，2017,11(2):1300-1304.
[6]胡博，叶峻宏，贾衍琰，等．超声-分光光度法快速测定水样中的化学需氧量[J]．应用化工，2017,46(1):190-193.
[7]袁红春，吕苏娜．水产养殖水质异常优化预测仿真研究[J]．计算机仿真，2017,34(12):447-450.
[8]徐玉妃，杨昆，罗毅，等．滇池草海水质等级预测模型研究[J]．水生态学杂志，2018,39(1):1-8.
[9]张双，周集体．高盐有机化工废水中COD与TOC的相关性[J]．化工环保，2018,38(1):122-126.
[10]孙伯寅，董国庆，张荣．支持向量机在水源水化学耗氧量预测中的应用[J]．环境与健康杂志，2016,33(6):544-547.