We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Prediction of Microplastic Removal Behavior via Gaussian Process Regression Based Residual Correction of a Reaction Kinetic Model
Summary
Scientists are using a special technology called low-temperature plasma (a zap of energized gas) to break down microplastics, and this study built a math model to predict exactly how well it will work under different power levels and treatment times. The model didn't just describe past results—it accurately predicted outcomes in new, untested conditions, helping researchers figure out the fastest, most energy-efficient way to destroy microplastics. This matters because it could speed up the development of practical, real-world tools to remove these tiny pollutants—which have been found in human blood, lungs, and organs—from water and the environment.
Low-temperature plasma technology has recently gained increasing attention as a promising approach for microplastic degradation. However, most previous studies have primarily focused on removal efficiency and byproduct analysis, with limited emphasis on quantitative prediction of removal behavior. In this study, we proposed a predictive model combining reaction kinetics and Gaussian process regression to describe microplastic removal behavior under varying power and treatment time conditions. The proposed model accurately reproduced the experimental removal behavior and demonstrated reliable predictive capability in untrained power and time domains. Furthermore, the model was used to identify favorable operating conditions for achieving rapid microplastic removal with high energy yield. These findings suggest that the kinetics–Gaussian process regression model can serve as a useful tool for predicting low-temperature plasma-based microplastic removal behavior and for optimizing process operating conditions.