面向水环境监测的结合ARIMA和LSTM模型的水质预测研究

打开文本图片集

中图分类号：X832

文章编号：1674-6139（2026）04-0120-06

文献标志码：B

WaterQualityPrediction Based on ARIMAand LSTM Models for Water Environment Monitoring

Hu Bentao，Zhao Yue，Jiang Peng，Yan Shengyu，Liu Weisong （HeilongjiangEcological Environment Monitoring Center，Harbin15Oo9O，China）

Abstract：Waterenvironmentmonitoring，asanimportantomponentofenvironmentalprotectionandwaterresourcemanagement， playsacrucialoleinessingateralityIndertopredictwaterqualitywaterualitpredictioodelombinngutoive movingaveragemodelandongshort-termmemorynetworkmodelwasdesigedndinstalledonawaterqualitymonitoringplatform.The PearsoncorelatiocoefcientwasalsousedtotesthecorelationetweewaterqualityfactorsinthesudyTheesultssowedthathe maximum accuracy of the designed prediction model was 98. 65% ， 98.21% ， 98.72% ，and 98.91% for dissolved oxygen，pH，conductivity，andaterpeaturetsspecielsposeeoftsiedlafoetnatoftompaoo el.Researchanddesignpredictivemodelsandplatformshavegodperformance，whichcanpromotetheprotectionandgoveranceofwater environment.

Keywords：monitoring；ARIMA；LSTM；waterquality；forecast

前言

水环境监测作为评估水质状况和预警污染事件的关键手段，在保护水资源、维护生态平衡和保障人类健康上发挥着重要的作用[1-2]。（剩余6308字）