人工神经网络方法在内蒙古砷中毒病区的应用

Application of artificial neural network on arseniasis area of Inner Mongolia

目的判明砷中毒病区水中以砷为主的各种复杂化学成份之间的非线性关系，以预测砷中毒发病率。方法采用人工神经网络ＢＰ模型进行研究。结果首次建立了内蒙古砷中毒病区水砷与砷中毒关系的人工神经网络ＢＰ模型。结论检验结果表明，该模型具有较高的模拟和仿真功能，用模型预测砷中毒发病率有较高的精度，绝对误差为０．５７，因此。

Objective To find out nonlinear relationship between arseniasis and main matter of arsenic and the other complex chemical compounds in water in arseniasis area and forecast the incidence rate of arseniasis.Methods Model of artificial neural network was used.Results The model of artificial neural network of the relationship between arsenic in water and arseniasis in Inner Mongolia was established firstly.Conclusions The examination indicates that the model has good analogue function,the accurate degree to forecast the incidence rate is high and absolute error is 0.57.The model has great applied and popularized value.