Modeling lipase production process using Artificial Neural Networks

Solving the fermentation process modeling represents a challenge for many industries. The reason is the nonlinearities in the bioprocess which makes traditional modeling techniques limited in their effectiveness. Mass production in biotechnology industries such as chemical, food, pharmaceutical, and health care industries are rapidly growing. It is urgently required to develop efficient, accurate, not expensive, and reliable computing models with more accurate computation of Lipase activities. In this paper, we propose the use of Artificial Neural Network (ANN) to develop a non-parametric model for the lipase activity production. The process depends on building ANN models which can estimate the Lipase activities based on set of experimental data produced at the laboratory. A comparison between the proposed ANN and the traditional polynomial models is provided. ANN was able to provide excellent results with respect to modeling performance.