基于深度学习的郑州市空气预测模型  

Zhengzhou Air Quality Forecasting Model Based on Deep Learning

作　　者：温情 吴建军[1] 李智慧[1] WEN Qing;WU Jianjun;LI Zhihui(College of Information Science and Engineering,Henan University of Technology,Zhengzhou Henan 450001,China)

机构地区：[1]河南工业大学信息科学与工程学院,河南郑州450001

出　　处：《信息与电脑》2021年第14期86-88,共3页Information & Computer

摘　　要：传统的空气质量预测方法存在效率低、效果差等问题,难以实现精确预测。长短期记忆网络(LongShortTerm Memory,LSTM)通过加入隐藏神经单元和门控单元,实现对时间序列之间相互信息的学习,解决了数据长期依赖问题。本文构建LSTM神经网络模型提取和学习郑州市空气质量数据中的时间特征和空间特征,实现对PM2.5的长期预测。The traditional air quality prediction method has some problems,such as low efficiency and poor effect,which is difficult to achieve accurate prediction.The long short-term memory(LSTM)network solves the problem of long-term data dependence by adding hidden neural units and gating units to extract effective mutual information between time series.Using LSTM can further realize the prediction algorithm with less human intervention,rapid convergence and better results.In this paper,the LSTM neural network model is constructed to extract and learn the temporal and spatial features in the air quality data of Zhengzhou city to achieve the long-term prediction of PM25.

关 键 词：空气质量 深度学习 LSTM 预测模型 时序 预测精度 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]