基于SSA-Bi-LSTM的港口环境空气质量指数预测  

Prediction of Port Ambient Air Quality Index Based on SSA-Bi-LSTM

作　　者：田序伟 杨凯 殷彤 郑冰 曾仕豪 TIAN Xuwei;YANG Kai;YIN Tong;ZHENG Bing;ZENG Shihao(Zhejiang Institute of Communications Co.,Ltd.,Hangzhou Zhejiang 410000,China)

机构地区：[1]浙江数智交院科技股份有限公司,浙江杭州410000

出　　处：《交通节能与环保》2023年第5期32-36,41,共6页Transport Energy Conservation & Environmental Protection

基　　金：浙江省交通运输厅科技计划项目(2021031)。

摘　　要：为贯彻绿色港口发展理念,提升港口大气污染监测治理能力,本研究提出了一种基于双向长短期记忆神经网络模型(Bidirectional LSTM,Bi-LSTM)的港口空气质量指数预测算法,并引入了麻雀搜索算法(Sparrow SearchAlgorithm,SSA)进行参数优化,以提高预测精度和模型的稳定性。选取浙江省嘉兴市乍浦港空气质量监测数据和气象数据为实验样本进行模型训练,实验结果显示,SSA-Bi-LSTM模型相比LSTM、CNN-LSTM、Bi-LSTM模型误差更小,通过该算法的应用,能够更准确地预测港口环境空气质量指数,为港口大气监测与治理提供科学依据和决策支持。To promote the development of green ports and enhance the monitoring and control capabilities of port air pollution,this study proposes a port air quality index prediction algorithm based on the Bidirectional Long-Short Term Memory(Bi-LSTM)neural network model.The Sparrow Search Algorithm(SSA)is introduced for parameter optimization to improve prediction accuracy and model stability.The experimental training dataset consists of air quality monitoring data and meteorological data from Zhapu Port Area in Jiaxing City,Zhejiang province,collected in March 2022.The experimental results show that the SSA-BI-LSTM model has less error than that of LSTM,CNN-LSTM and Bi-LSTM models.By applying this algorithm,more accurate predictions of the port's environmental air quality index can be obtained,providing scientific basis and decision-making support for port air monitoring and management.

关 键 词：交通碳排放 空气质量指数预测 双向长短期记忆神经网络 麻雀搜索算法 港口大气监测 

分 类 号：X831[环境科学与工程—环境工程]