中文摘要：

建立臭氧降解微囊藻毒素MC-LR的人工神经网络模型，研究臭氧投加量、MC-LR初始质量浓度、pH等对降解速率的影响，并以反向传播算法的神经网络模型对多因素条件下的降解效果进行仿真预测。研究结果表明：降解速率不受初始MC-LR质量浓度的影响；臭氧投加量的增加能有效提高MC-LR的降解速率；pH降低能大幅度改善降解效果，尤其在酸性条件下，pH的变化对降解速率的影响程度更大；在具备酸性条件和臭氧量较高时，在短时间内即可达到很高的去除率，否则降解效果不明显；该模型能预测复杂多因素试验条件下的有机物降解效果，为试验及实际降解MC-LR提供理论指导，克服了初等函数模型的局限性。

英文摘要：

An artificial neural network （ANN） model of microcystin-LR （MC-LR） degradation by ozonatlon was smama. The affect on degradation of Ozone dose, MC-LR initial mass concentration and pH was investigated. And the removal effect with various factors was simulated and predicted by the model. The results show that the degradation rate is invariable with different MC-LR initial mass concentrations. The addition of ozone dose can increase the MC-LR degradation rate effectively, the decline of pH can improve the degradation effect obviously, especially in acidity condition. A big removal efficiency can be gotten in a short time with acidity condition and large ozone dose, The ANN model can be used to predict the degradation effect of MC-LR with complex various factors, provide theoretical foundation for MC-LR degradation and overcome the limitation of common model.