Artificial neural networks (ANN)-genetic algorithm (GA) optimization on thermosonicated achocha juice: kinetic and thermodynamic perspectives of retained phytocompounds

Abstract

The extraction of phytocompounds from Achocha (Cyclanthera pedata) vegetable juice using traditional methods often results in suboptimal yields and efficiency. This study aimed to enhance the extraction process through the application of thermosonication (TS). To achieve this, an artificial neural network (ANN) and a genetic algorithm (GA) were utilized to simulate and optimize the process parameters. The study investigated the influence of ultrasonic amplitude (30%–50%), temperature (30 °C–50 °C), and sonication duration (15–60 min) on total polyphenolic content (TPC), total flavonoid content (TFC), antioxidant activity (AOA), and ascorbic acid content (AA). Remarkably, the ANN-GA optimization resulted in optimal TS conditions: an ultrasonic amplitude of 40%, a temperature of 40 °C, and a sonication duration of 30 min. Subsequent analysis of extraction kinetics and thermodynamics across various temperatures (30 °C–50 °C) and extraction times (0–30 min) demonstrated R 2 (0.98821) and χ2 (1.74773) for TPC with activation energy (Ea) 26.0456, R 2 (0.99906) and χ2 (0.07215) for TFC with Ea 26.2336, R 2 (0.99867) and χ2 (0.03003) for AOA with Ea 26.0987, R 2 (0.99731) and χ2 (0.13719) for AA with Ea 26.0984, validating the pseudo second-order kinetic model. These findings indicate that increased temperature enhances the saturation concentration and rate constant of phytochemical extraction.