Studying Rheology of Tannery Waste Liquor Using Artificial Neural Networks

Understanding the rheology of tannery waste is essential for efficient waste management and the design of required devices such as pumps and agitators. Experimental studies have shown that tannery waste exhibits shear-thinning, non-Newtonian behavior, where the increase in solid concentration and decrease in temperature lead to higher shear stress and viscosity. To verify these findings, an artificial neural network (ANN) model was developed to predict the rheological behavior of tannery waste. The model takes solid concentration, temperature and shear rate as inputs and provides shear stress and viscosity as outputs. The ANN demonstrated exceptional accuracy, achieving mean squared error (MSE) values of 0.0003 and 0.0001 and determination coefficients (R²) of 0.9999 and 0.9945 for shear stress and viscosity, respectively. These results confirm that the ANN can accurately replicate the experimental observations and outperform traditional curve-fitting techniques, offering a robust tool for analyzing the complex rheology of tannery waste. This approach underscores the potential of integrating machine learning into waste management systems, promoting sustainable industrial practices.