Efficient Developments in Modeling and Optimization of Solid State Fermentation

Artificial intelligence techniques are important tools for modelling and optimizing the solid-state fermentation (SSF) factors. The performance of fermentation processes is affected by numerous factors, including temperature, moisture content, agitation, inoculum level, carbon and nitrogen sources, etc. In this paper, the identification of non-linear relationship between fermentation factors and targeted objectives is performed, first, using the learning capabilities of a neural network (NN). Then, this approach is coupled with various artificial intelligence techniques to optimize the fermentation process, such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The effectiveness of different approaches is compared with the classical statistical techniques, such as Response Surface Methodology (RSM), that are increasingly being used. This paper presents the first attempt to adapt these approaches on the solid state fermentation process. The obtained results prove the effectiveness of the proposed approach. Particularly, we show that this approach leads to a significant improvement on the fermentation process performance.