基于PSO-BP的藻类膜系统处理己内酰胺模拟污水水质预测

Water quality prediction of simulated caprolactam wastewatertreated by PSO-BP algae membrane system

通过对藻类膜系统处理的465组己内酰胺模拟废水进行测定,探讨进水水质、生物指标、操作指标与出水水质的相关关系及影响规律;以系统各参数为输入信号、出水水质为输出信号,利用粒子群(PSO)算法优化反向传播(BP)神经网络,建立了PSO-BP藻类膜系统水质预测模型,并与其他预测模型对比分析。结果表明:藻类膜系统对高浓度含氮有机废水处理效果显著,当水力停留时间设为5 d时,系统出水平均总氮含量、总磷含量和化学需氧量分别为6.68 mg/L、0.11 mg/L和23.86 mg/L;PSO-BP水质预测模型预测结果判定系数为0.93~0.95,为藻类膜系统在污水处理过程中水质的高效调控提供了科学参考。

The correlation and influence of influent water quality,biological index,operation index and effluent water quality were discussed by measuring 465 groups of simulated caprolactam wastewater treated by algae membrane system.Taking the system parameters as the input signal and the effluent quality as the output signal,the particle swarm optimization(PSO)algorithm was used to optimize the back propagation(BP)neural network,and the water quality prediction model for PSO-BP algae membrane system was established and compared with other prediction models.The results showed that the algae membrane system had a significant treatment effect on the high-concentration nitrogen-containing organic wastewater;the average total nitrogen content,total phosphorus content and chemical oxygen demand in the effluent were 6.68 mg/L,0.11 mg/L and 23.86 mg/L,respectively,when the hydraulic retention time was set as 5 d;and the determination coefficient of the prediction results of PSO-BP water quality prediction model was 0.93-0.95,which provided a scientific reference for the efficient regulation of water quality in the wastewater treatment process of algae membrane system.