Chlorpyrifos Degradation from Aqueous Solution Using Ultrasonic Treatment: A Comparative Study of RSM, ANN, and SVM Modeling

Chlorpyrifos (CPF) is a broad-spectrum organophosphate insecticide whose extensive use has led to widespread environmental contamination, threatening non-target organisms. This study examines the efficacy of ultrasound technology as an eco-friendly method for CPF degradation. The principal operational parameters - initial CPF concentration, ultrasonic power, pH, reaction time, and solution volume - were optimized using Response Surface Methodology (RSM) based on Central Composite Design (CCD), and the optimal conditions were determined. The process was further modeled using Artificial Neural Networks (ANN) and Support Vector Machines (SVM). The ANN was constructed as a two-layer MultiLayer Perceptron (MLP) and trained using the Levenberg–Marquardt backpropagation algorithm, with tangent sigmoid and linear transfer functions applied in the hidden and output layers, respectively. For the SVM, the Gaussian radial basis function (RBF) kernel was selected as the optimal kernel. Comparative analysis indicated superior predictive accuracy for the SVM (R۲= ۰.۹۹۶) compared with the ANN (R۲= ۰.۹۹) and RSM (R۲= ۰.۹۸). The results confirm that ultrasound-assisted remediation is a highly effective and sustainable strategy for the decontamination of CPF in aquatic ecosystems.

کلیدواژه ها:

Chlorpyrifos ، Pesticide degradation ، RSM ، ANN ، SVM ، Ultrasonic treatment