Viscosity Estimation of Water-PVP Solutions from Droplets Using Artificial Neural Networks and Image Processing.

The viscosity of a liquid is the property that measures the liquid internal resistance to flow. Viscosity monitoring is essential for quality control in many industrial areas, such as the chemical, pharmaceutical, and energy-related industries. Capillary viscometers are the most used instrument for measuring viscosity. Still, they are expensive and complex, which represents a challenge in industries where accurate and real-time viscosity knowledge is essential. In this work, we prepared eight solutions with different water and PVP (Polyvinylpyrrolidone) ratios, measured their different viscosity values, and produced videos of their droplets. We aimed to extract the droplets’ characteristics using image processing and to use these characteristics to train an Artificial Neural Network model to estimate the viscosity values of the solutions. The proposed model was able to predict the viscosity value of the samples using the characteristics of their droplets with an accuracy of 83.08% on the test dataset.