论文部分内容阅读
水质为水库运行管理的重要因素,水质变量包括溶解氧、总磷、叶绿素a及透明度等,是水库富营养化判定的重要指标。在线性回归法分析的基础上,优化环境因子参数,分别采用多元线性回归模型、径向基函数模型与自适应模糊神经网络推理模型对辽宁省大伙房水库的水质进行预测,并通过平均绝对误差、均方根误差及相关系数判定水质模型的预测效果。结果表明,自适应模糊神经网络推理模型预测效果明显优于多元线性回归模型和径向基神经网络模型,因此自适应模糊神经网络推理模型更适合于大伙房水库的水质预测。
Water quality is an important factor in reservoir operation and management. Water quality variables include dissolved oxygen, total phosphorus, chlorophyll a, and transparency, which are important indicators of reservoir eutrophication. Based on the linear regression analysis, the parameters of environmental factors were optimized, and the water quality of Dahuofang Reservoir in Liaoning Province was predicted by multivariate linear regression model, radial basis function model and adaptive fuzzy neural network reasoning model. The average absolute error , Root mean square error and correlation coefficient to determine the water quality model prediction. The results show that the adaptive fuzzy neural network inference model is obviously superior to the multiple linear regression model and RBF neural network model. Therefore, the adaptive fuzzy neural network inference model is more suitable for the water quality prediction of Dahuofang Reservoir.