基于主成分分析的GA-BP模型在城市需水预测中的应用

发布时间：2018-07-06 08:43

本文选题：主成分分析 + BP神经网络　；参考：《南水北调与水利科技》2017年06期

【摘要】：针对城市需水预测模型中需水量影响因子多、影响因子之间普遍存在多重共线问题,以及BP神经网络收敛速度慢、易陷入局部最优等缺点,提出一种由主成分分析、遗传算法及BP神经网络三者相结合的改进预测模型。以泰州市为实例,建立以主成分分析筛选需水量主要影响因子,遗传算法优化BP网络连接权值和阈值的需水预测模型,预测结果与BP神经网络预测模型预测结果做对比。结果表明:改进预测模型对泰州市2003-2014年需水量预测的平均相对误差为0.564%,最大相对误差为1.681%,精度优于BP神经网络预测模型;改进预测模型预测值与实际泰州市需水量吻合良好且训练速度更快、预测精度更高,可作为需水预测的一种有效方法。
[Abstract]:In view of the many factors affecting the water demand in the urban water demand forecasting model, the common multiple collinear problem exists among the influencing factors, and the slow convergence speed of the BP neural network and easy to fall into the local optimality, an improved prediction model is proposed by combining the principal component analysis, the genetic algorithm and the BP neural network three. A case of Taizhou is established. The main influence factors of water demand are selected by principal component analysis, and the genetic algorithm is used to optimize the water demand prediction model of BP network connection weights and thresholds. The prediction results are compared with the prediction results of the BP neural network prediction model. The results show that the average relative error of the improved prediction model for the 2003-2014 year water demand forecast in Taizhou is 0.564% and the maximum relative value is relative. The error is 1.681%, and the accuracy is better than the BP neural network prediction model. The improved prediction model is in good agreement with the actual water demand in Taizhou, and the training speed is faster and the prediction accuracy is higher. It can be used as an effective method for water demand prediction.
【作者单位】：河海大学水利水电学院;泰州市水资源管理处;
【基金】：国家重点计划研发课题(2016YFC0400909;2016YFC0402605) 江苏省高校优势学科建设工程资助项目(水利工程)~~
【分类号】：TP18;TV213.4

【相似文献】