RF-TOPSIS-MCS-Based Comprehensive Evaluation Method for Reservoir Water Quality and Its Application
In the management of reservoirs,accurate assessment of water quality is crucial.However,uncertainties in data collection and limitations in the allocation of weights to water quality factors can both impact the evaluation results.To address these issues,a novel method which integrates Random Forest(RF)classification prediction,Technique for Order Preference by Similarity to Ideal Solution(TOPSIS),and Monte Carlo Simulation(MCS)is developed for comprehensive water quality evaluation using the Dahuofang Reservoir as a study case.Initially,a dataset with a normal distribution is generated using MCS based on the actual monthly average water quality data.Subsequently,the RF is utilized for weight assignment in the TOPSIS method and a membership function is employed for the comprehensive evaluation of water quality.The results of case study are consistent with the water quality grades provided in the monthly water environmental quality reports of Liaoning Province,indicating that the proposed approach can provide robust technical support for the water quality assessment of large reservoirs.
random forest weightingTOPSISMonte Carlo simulationreservoir water quality assessment
国家科技期刊平台
NSTL
万方数据
