Optimizing a hybrid process of electrolysis ultrasound and persulfate for remediation of petroleum contaminated soils using AI models.

Research uses both RANSAC Regressor and Monte Carlo Optimization to improve the performance of electrolysis/ultrasound/persulfate system which detoxifies petroleum-contaminated soils. The Artificial Intelligence (AI) models used to optimize six process parameters showed X2 (humidity) and X3 (voltage) and X5 (surfactant) enhanced removal efficiency the most but X1 (pH) presented a robust negative impact. The selected optimal conditions for pollutant removal resulted from Monte Carlo simulations which specified X1 at 8.50 and X2 at 188.67 with X3 set to 2.45 and X4 at 0.64 and X5 at 0.07 and X6 at 198.02. The study supports AI-based models as strong tools which enable optimization of complex environmental remediation methods and enhance pollutant remediation procedures. The study hypothesis demonstrates that artificial intelligence models (RANSAC Regressor and Monte Carlo Optimization) precisely find crucial process parameters which enhance the efficiency of hybrid electrolysis/ultrasound/persulfate treatment in removing petroleum hydrocarbons from contaminated soil. Hybrid remediation technologies receive improved performance efficiency through the application of AI optimization with RANSAC and Monte Carlo models combined. These discoveries lead to worldwide applications that use affordable flexible methods for treating petroleum-contaminated soils to be deployed extensively in global contaminated sites.