Comparative ANN-TLBO and RSM optimisation approach for bioactive potential of microwave convective dried mango (Mangifera indica)

Mango (Mangifera indica) is a rich source of antioxidants like phenolic acids and flavonoids. Polyphenol oxidase (PPO) and peroxidase (POD) enzyme oxidises mono and/or diphenols, thus the antioxidant potential diminished. For cost-effective and sustainable preservation of this sub-tropical climacteric fruit, microwave convective dehydration is a prospective choice. Drying operation deactivates PPO and POD. Microwave convective drying was done at 100-300 watt of power level, the temperature range of 40-80°C, total soluble solid content of 20-30°B and puree load of 0.4-0.6 g/cm² to investigate the effect of process parameters on total phenolic content (TPC) of the final product. Response surface methodology (RSM) was employed to optimise the TPC value. Further, artificial neural network (ANN) with the back propagation-feed forward modelling approach was adopted for the experimental results obtained, teaching-learning based optimisation (TLBO) was then employed to acquire the optimised drying condition for the sample with maximum TPC value. The optimised process condition obtained from both the method was virtually compatible. From RSM analysis the maximum TPC value of 13.36 mg GAE/100g was observed at 170.27-watt power level, 57.84°C oven temperature, 0.60 g/cm² of puree load and 29.05°B of total soluble solid content. Whereas, from ANN-TLBO technique 14.29 mg GAE/100g of TPC was attained for the combination of drying parameters as follows: 175.08 watt, 57.15°C, 0.60 g/cm² and 29.25°B. The ANN-TLBO approach predicted better optimised result (TPC value) in comparison to the RSM method.