Prediction and Analysis of Heavy Metal Leaching Trend in Solid Waste Building Materials
[Purposes]Heavy metal leaching prediction model based on Long Short Term Memory(LSTM)neural network was proposed to address the poor predictive control effect of heavy metal leaching in solid waste building materials.[Methods]The model was used to predict and analyze the leaching amounts of heavy metal elements such as Cr,Zn,and Pb in granular and block shaped unburned bricks.To further improve the applicability and training convergence speed of the model,the parameters of the Adam algo-rithm were optimized.Adopting an improved prediction model to simulate and verify the leaching amounts of Cr,Zn,and Pb in block shaped unburned bricks.[Findings]The results indicate that the pre-diction of heavy metal leaching from block shaped unburned bricks,the decision coefficients R2 of the model for predicting Cr,Zn,and Pb are all greater than 0.97,which indicates a relatively accurate predic-tion result.[Conclusions]This study has significance for the control of heavy metal release in solid waste building materials.
NETL
NSTL
万方数据
