摘要
作为城市景观的重要组成部分,城市河道的水位和水质直接影响着区域生态发展。为找出城市河道水位和水质的估算模型,本文采用改进的鲸鱼优化算法对长短期记忆神经网络模型进行优化,得出了IGA-WOALSTM模型,并以南排河为例,验证了该模型精度。结果表明,城市河道水位和水质在时间上均呈现明显的规律性,IGA-WOA-LSTM模型模拟结果在变化趋势和精度方面均最优,一致性指数均在0.9以上。
As an important component of urban landscapes,water level and water quality in urban river channels directly influence regional ecological development.In order to establish estimation models for water level and water quality in urban river channels,this paper utilizes an improved whale optimization algorithm to optimize the Long Short-Term Memory(LSTM)neural network model,resulting in the IGA-WOA-LSTM model.The model’s accuracy is validated using Nanpai River as a case study.The results demonstrate that both water level and water quality in urban river channels exhibit clear temporal patterns.The IGA-WOA-LSTM model yields optimal results in terms of trend simulation and accuracy,with consistency indices exceeding 0.9.
作者
龙幸幸
LONG Xingxing(Xingtai Xiangyu Water Conservancy Survey and Design Co.,Ltd.,Xingtai 054000,China)
出处
《水资源开发与管理》
2023年第9期53-59,共7页
Water Resources Development and Management
关键词
城市河道
水位
水质
鲸鱼算法
LSTM模型
urban river channel
water level
water quality
whale algorithm
LSTM model
