Data-Driven Algorithm for Prediction of Atomization Effect and Quality Control of Spiked Feed Solution

In order to improve the atomization effect and quality control of the spiked feed solution, a data-driven prediction model is constructed to analyze the effects of physical properties such as viscosity, surface tension, and temperature on the formation of fog droplets, and a machine learning method is used to optimize the atomization quality. Experiments show that the neural network model is optimal in prediction accuracy, MSE is reduced to 1.15 × 10–3, MAE is reduced to 0.62 × 10–3, droplet uniformity coefficient is improved to 0.89, and spray coverage is expanded. The optimized spray particle size distribution was more uniform and quality stability was improved. The results show that the data-driven method can effectively enhance the fine control of the atomization process.