基于RBF神经网络模型对污泥减量优化的研究

Study on Optimization of Sludge Reduction Based on RBF Neural Network Model

研究了温度、pH值、反应时间、高铁酸钾投加量等因素对污泥溶胞效果和分解效果的影响,并通过建立RBF神经网络模型对实验进行优化。研究结果表明,温度为60℃、反应时间为2~4 h、pH值为12、高铁酸钾投加量5.5mg/(gSS)的条件下,污泥减量处理效果显著且较为经济。此外,RBF神经网络模型计算的污泥溶胞率、污泥分解率值与实验得出的结果,两者相对误差均小于5%,验证了该模型的良好拟合性。

The effects of temperature,pH value,reaction time and potassium ferrate dosage on sludge lysis and decomposition are studied,and the RBF neural network model is established to optimize the experiment.The results show that the sludge reduction treatment effect is remarkable and more economical under the conditions of temperature of 60℃,reaction time of 2~4 h,pH value of 12 and potassium ferrate dosage of 5.5 mg/(gSS).In addition,the relative errors of the sludge lysis rate and sludge decomposition rate calculates by the RBF neural network model and the experimental results are less than 5%,which verifies the good fitting of the model.