Coconut Water Microfiltration Optimization Using Response Surface Modeling, Neural Networks, and Genetic Algorithms:...

Area of Science:

• Food Science and Technology
• Chemical Engineering
• Membrane Science

Background:

• Coconut water's industrial application is limited by rapid post-extraction degradation.
• Effective processing methods are needed to preserve coconut water's quality and extend its shelf life.
• Crossflow microfiltration offers a potential solution for processing coconut water.

Purpose of the Study:

• To optimize the crossflow microfiltration of coconut water using silicon carbide membranes.
• To investigate the effects of pressure and temperature on permeate flux and membrane fouling.
• To evaluate the hydraulic resistance, fouling behavior, and retention of nutrients under optimized conditions.

Main Methods:

• Optimization using a face-centered design (FCD) and artificial neural network-genetic algorithm (ANN-GA) modeling.
• Analysis of permeate flux and fouling index as key process responses.
• Assessment of hydraulic resistance, fouling characteristics, and mineral/metabolite retention.

Main Results:

• Permeate flux exhibited nonlinear behavior with pressure and temperature, while the fouling index was primarily pressure-dependent.
• Both FCD and ANN-GA models identified a consistent optimal operating window (60 kPa, 35 °C).
• The optimized process achieved high permeate flux (1085.23 L h⁻¹ m⁻²) with manageable, predominantly reversible fouling, and high retention of minerals and metabolites.

Conclusions:

• Crossflow microfiltration with silicon carbide membranes is an effective method for processing coconut water.
• The optimized conditions balance high productivity, controlled fouling, and preservation of nutritional value.
• This approach enhances the industrial potential of coconut water by addressing quality deterioration issues.