A Mixed Prediction Method of rf-hw Water Quality Index Based on Meteorological Factors

Abstract

The invention discloses a mixed prediction method of RF-HW water quality index based on meteorological factors, comprising the following steps: collecting meteorological factor data and water quality index data, and after filtering the collected data for abnormal values, constructing time series data including meteorological factors and A sample database of water quality index time series data; construct a water quality index prediction model including Holt-Winter model, random forest model and random forest correction model; when applied, input meteorological factor time series data and water quality index time series data into In the water quality index forecasting model, the mixed forecast value of the water quality index time series data in the future is calculated. Using the combination of random forest algorithm and Holt-Winter model to overcome the fixed defects of a single model can improve the prediction stability and accuracy of the model.

Description

technical field [0001] The invention belongs to the technical field of water environment monitoring, and in particular relates to an RF-HW water quality index mixed prediction method based on meteorological factors. Background technique [0002] The safety of drinking water is a necessary condition for people's life. As the birthplace and supply source of drinking water, drinking water source is the key to ensuring the safety of drinking water. By predicting the changes of raw water quality in water sources such as rivers, lakes, and reservoirs, effective treatment measures can be taken in advance for the fluctuation of water quality to ensure the quality of water supply. [0003] Water quality prediction models can be divided into mechanism models and non-mechanism models. The mechanism model is based on the physical, chemical and biological reaction mechanisms of various components in the water cycle in the water body, and describes the changes of the water system through...

AI Technical Summary

Problems solved by technology

[0007] In view of the above, the purpose of the present invention is to provide a kind of RF-HW water quality index mixed prediction method based on meteorological factors, this method adopts random forest algorithm (Random forest, RF) and Holt-Winters model (Holt-Winters, HW ) to overcome the fixed defects of a single model, which can improve the prediction stability and accuracy of the model

Images