River water quality predicting method

Abstract

The invention provides a river water quality predicting method. The method comprises following steps: step one, based on time sequence test data of water quality parameters of a target water area, using an ARIMA prediction model, which is established in advanced based on historical time sequence data samples of the water quality parameters, to primarily predict the primary prediction data of the water quality parameters at designated time and calculate the predicted residual data; and step two, based on the time sequence test data, primary prediction data, and the predicted residual data, using the historical time sequence data samples and the ARIMA prediction model to predict prediction data of the water quality parameters at designated time through a wavelet neural network model constructed by genetic algorithm. Compared with a conventional water quality prediction method, the versatility, prediction precision, convergence speed, and efficiency are all enhanced.

Description

technical field [0001] The invention relates to the technical field of environmental prediction, and more particularly, to a river water quality prediction method. Background technique [0002] Prediction of river water quality is a prerequisite for realizing flexible management of river water systems and preventing and controlling water pollution. Factors affecting water quality in river basins include water quality parameters such as PH value, dissolved oxygen (DO), electrical conductivity (EC), turbidity (TU), ammonia nitrogen (NH3-N), chemical oxygen demand (COD) and permanganate , the prediction of changes in dissolved oxygen, ammonia nitrogen, total phosphorus, and total nitrogen is more conducive to the realization of pollution control and water source management in different river basins. [0003] At present, the commonly used single prediction methods include water quality simulation prediction, neural network model prediction, time series prediction method, gray p...

AI Technical Summary

Problems solved by technology

[0004] However, in the process of water quality prediction, the above-mentioned single prediction methods cannot take into account various influencing factors of water quality changes, and the processing method is single, so the prediction results are often not accurate enough, and the prediction performance is not good.

Because the combined water quality prediction method based on ARIMA and ANN does not consider the relationship between the original data and the residual data, the neural network parameters are easy to fall into the local optimal value, the prediction effect is poor, and the accuracy is not high.

Images