Water quality prediction method based on attention neural network

Abstract

The invention discloses a water quality prediction method based on an attention neural network. The method comprises the following steps: step 1, collecting data of a certain water quality index; 2, carrying out z-score standardization processing on the data of the water quality indexes; 3, constructing a training set Training by using the processed data of the water quality indexes; 4, constructing a water quality prediction model based on the attention neural network; 5, training the constructed water quality prediction model based on the attention neural network by using the constructed training set Training; and 6, predicting the data of the water quality index at the future moment by using the trained attention neural network-based water quality prediction model. According to the water quality prediction method provided by the invention, an attention mechanism is introduced into the bidirectional LSTM neural network, different importance degrees of the data of the water quality indexes at each moment to the prediction result are considered, and different weights are given to the water quality indexes at each moment, so that relative errors generated during water quality prediction are reduced, and the prediction accuracy is improved.

Description

technical field [0001] The invention belongs to the field of water quality prediction, in particular to a water quality prediction method based on an attention neural network. Background technique [0002] In recent years, with the rapid development of my country's economy, natural or human factors have caused natural water bodies to continue to deteriorate. At the same time, the sharp increase in domestic and industrial water consumption has led to the gradual depletion of land water. Therefore, the water problem has become the most serious environmental problem in our country. . There are various pollutants in the water body, and their concentration directly affects the quality of the water quality. If the water quality problems can be found in advance, the reasons can be analyzed, and corresponding preventive measures can be formulated to improve the water quality environment of the water body. Insufficient resources, so research on predicting the changing trend of water ...

AI Technical Summary

Problems solved by technology

Although LSTM has achieved good accuracy in water quality prediction, the LSTM neural network also has some limitations. For example, the degree of influence of water quality indicators on the prediction results at different times is the same, and it does not consider that the water quality indicators at the latest time may affect the prediction results. bigger problem

Images