engleski

Microcapsules production with mathematical optimization coupled with CFD simulation and neural networks

Modern agricultural production uses more and more new and innovative methods for nutrition and protection of certain plant crops. One of the newer methods is the method of applying microcapsule formulations filled with bioactive components with time release. To achieve the production of microcapsule formulation with optimal properties and shape, it is necessary to optimize the preparation parameters (concentration, temperature, pressure-flow rate, nozzle size). We propose a method to select these parameters using a novel procedure. Firstly, a potentially promising parameter set will be chosen based on a neural network trained using previous data. Secondly, a mathematical optimization algorithm will refine the set using CFD simulations of the microcapsule equipment. This is all done to minimize the number of real-world experiments that need to be done. Lastly, the lab technician will receive instructions on which parameters to use and measure the results entering them into the algorithm which in turn will iteratively offer an improved set of parameters with regards to the chosen properties.

microcapsules optimization, encapsulation, computational fluid dynamics, neural networks, mathematical optimization

nije evidentirano