Central composite design, Pareto analysis, and artificial neural network for modeling of microwave processing parameters for tender coconut water

Polyphenol oxidases (PPO) and peroxidases (POD) are the major enzymes that affect the quality of tender coconut water (TCW). Advanced thermal treatment such as microwave treatment has the potential for the inactivation of food enzymes. The experiments were conducted at three different microwave power levels (450, 600, and 900 W) and five different exposure times (70, 80, 90, 100, 110, and 120 s). The modeling and optimization of process parameters were done using a central composite design and artificial neural network. The microwave power level of 600 W for 120 s exposure time was suitable for enzyme inactivation with minimal quality loss. Optimized treatment has pH=5.02, total soluble solids (TSS)=5.68 °Brix, turbidity=12.51 NTU, titratable acid (TA)=0.07% of malic acid, PPO=0, POD=0, phenolic content=37.238 mg GAE/L and overall acceptability (OA)=7.5. These results confirmed that microwave treatment could be the potential alternative to conventional thermal treatment for processing tender coconut water.