论文部分内容阅读
为了更有效地对水环境质量进行综合评价,论文提出了一种改进的T-S(Takagi-Sugeno)模糊神经网络水质评价模型,该模型首先通过减法聚类确定模糊C均值聚类(FCM)的初始聚类中心和聚类数目,改善传统FCM算法对聚类中心初值选取的随机性及样本的敏感性,降低陷入局部最优解的可能性。将减法聚类改进的FCM算法应用到T-S模糊神经网络的特征提取中,对T-S模糊神经网络模型进行结构辨识,提高评价模型的准确性和收敛速度。通过与传统的T-S模糊神经网络比较,水质评价结果准确率更高。
In order to evaluate the quality of water environment more effectively, an improved TS (Takagi-Sugeno) fuzzy neural network water quality evaluation model is proposed in this paper. Firstly, the initial FCM (Fuzzy C-Means Clustering) Clustering centers and the number of clusters to improve the randomness and sample sensitivity of the traditional FCM algorithm for the selection of initial clustering centers and reduce the possibility of falling into the local optimal solution. The modified FCM algorithm based on subtractive clustering was applied to the feature extraction of T-S fuzzy neural network, and the structure of T-S fuzzy neural network model was identified to improve the accuracy and convergence speed of the evaluation model. Compared with the traditional T-S fuzzy neural network, water quality evaluation results with higher accuracy.