Study on water quality evaluation of Xijiang river based on cloud computing and GA-BP neural network

Number of views： 11

分享到:

摘要：

为提高水质评价精度,针对BP神经网络的预测结果易受初始连接权值和阈值的影响以及易陷入局部极值的问题,提出一种遗传算法优化BP神经网络的水质评价模型.针对水质评价数据特性,引入Multi-Agent和分布式思想,利用云计算的MapReduce框架对GA-BP模型进行并行化改进,提高其处理海量高维水质评价数据的能力.为证明所提算法的效果,选择西江2011-2015年的水质监测数据为研究对象,研究结果表明,与PSO-BP,GA-BP,DE-BP和BP相比,所提算法MR-GA-BP不但可以提高水质评价的精度,而且能够降低计算资源的消耗,缩短训练时间,具有很好的并行性能.

In order to improve the accuracy of water quality evaluation,a genetic algorithm was proposed to optimize the water quality evaluation model of BP neural network.According to the characteristics of water quality evaluation data,Multi-Agent and distributed ideas are introduced,and the GA-BP model is improved in parallel with the MapReduce framework of cloud computing,so as to improve its ability to process massive high-dimensional water quality evaluation data.In order to prove the effect of the algorithm,the water quality monitoring data of Xijiang river from 2011 to 2015 were selected as the research object.The research results show that,compared with PSO-BP,GA-BP,DE-BP and BP,the MR-GA-BP algorithm can not only improve the accuracy of water quality evaluation,but also reduce the consumption of computing resources,shorten the training time and have good parallel performance.FEWER

作者：

纪广月

Ji Guangyue(Department of public education,Guangdong University of Business and Technology,Zhaoqing 526020,China)

机构地区：

广东工商职业技术大学公共教学部

出处：

《betway官方app 学报：自然科学版》 CAS 北大核心 2020年第3期34-40,共7页

基金：

广东省教育厅高校特色创新类项目(自然科学)(2017GKTSCX109)。

关键词：

遗传算法云计算 BP神经网络水质评价粒子群算法

genetic algorithm cloud computing BP neural network water quality evaluation particle swarm optimization algorithm