论文部分内容阅读
为评价城市供水管网的安全性,保障其正常运行,笔者基于多元分类最小二乘支持向量机(LS-SVM)的方法,在对城市供水管网安全运行影响因素总结与分析的基础上,构建供水管网安全性评价的指标因素集与评价模型,通过对有限的经验数据的学习,建立供水管网安全性与其影响因素之间的非线性关系。运用该模型进行实例仿真模拟,通过与实际安全等级及BP神经网络模型预测安全等级之间的对比表明:基于LS-SVM的供水管网安全性评价方法具有较高的精度,正确分类率可以达到83.33%。
In order to evaluate the safety of urban water supply network and ensure its normal operation, based on the multi-classification LS-SVM method, based on the summary and analysis of the influencing factors of urban water supply network safety operation, Construct index set and evaluation model of water supply network safety evaluation, and through the study of limited empirical data, establish the nonlinear relationship between the safety of water supply network and its influential factors. The model is used to simulate the case. The comparison between the actual safety level and the BP neural network predicts the safety level shows that the LS-SVM based water supply network safety evaluation method has higher accuracy and the correct classification rate can reach 83.33%.