Water quality turbidity prediction model based on improved Mogrifier LSTM algorithm
Water environment and resource protection represent crucial tasks in the present era.To enhance water quality model prediction accuracy and develop more comprehensive water resource management strategies,a water quality turbidity data prediction model based on the improved Mogrifer LSTM algorithm was proposed to achieve accurate prediction of water quality data.Firstly,the model employed CNN to extract features from complex water quality data,effectively addressing the nonlinear and unstable characteristics of such data.Additionally,the traditional Mogrifier mechanism was optimized by introducing an expansion coefficient and hyperparameters was optimized using PSO algorithm.Through modified Mogrifier mechanism,context information from different moments in LSTM models was fused together to enhance interaction among water quality data.Comparison with many traditional models showed that CNN-improved Mogrifier LSTM model yielded better prediction results.
water quality predictionturbidityconvolutional neural networkMogrifier LSTM
万方数据
维普
