ANFIS-Based Modelling and Optimal Operating Parameter Determination to Enhance Cocoa Beans Drying-Rate

This work aims to maximize the drying-rate of the cocoa beans by controlling the values of temperature, pressure, and moisture contents. The drying-rate is maximized by defining the optimal operating conditions. Using the experimental data, obtained from the drying-rate process, a robust model that describes the drying-rate of the cocoa beans is generated using fuzzy logic approach. Then, the optimal operating conditions of the system are determined using Particle Swarm Optimization (PSO) algorithm. Three different operating parameters that influence the drying rate of the cocoa beans are assessed. These parameters are the temperature (°C), pressure (in.Hg), and moisture contents. Accordingly, during the optimization process, these parameters are used as the decision variables for the PSO optimizer in order to maximize the drying-rate which is used as a cost function. The resulting plots demonstrated a well-fitting between the output of the fuzzy model and the experimental data. The mean square errors of the resulting predictions are found 0.000297 and 0.00012 for testing and whole datasets, respectively. Based on the built model, the optimization process achieved a significant increase in the drying-rate to a value of 2.12 relative to 1.315 which was obtained experimentally. This proved that the fuzzy modelling with PSO attained an increase of 61.14% without changing the system’s design or the materials used. Indeed, the originality of this research work promotes the proposed model to be a valuable tool for the design of future dryers.