Advanced Soft Computing Techniques in Biogas Production Technology

Biogas production from organic wastes is a complex, dynamic, nonlinear, multivariable, and uncertain biological process whose underlying mechanisms are still unclear. Accordingly, this process is not amenable to conventional mathematical and phenomenological modeling and optimization approaches. Advanced soft computing techniques are considered as powerful tools for dealing with the complexity, nonlinearity, dimensionality, and uncertainties of complicated ill-defined biological processes like biogas production. For this reason, advanced soft computing techniques are extensively employed in biogas applications due to their higher efficiency, generalization, and simplicity. In this chapter, after introducing the soft computing techniques and briefly describing their theoretical backgrounds, an overview is presented of the most important applications of these approaches for modeling and optimization of biogas production processes. This chapter is arranged into four main sections. In the first section, artificial neural network (ANN) is introduced and its applications in biogas production processes are reviewed and discussed. In the second part, fuzzy logic systems like Sugeno and Mamdani systems are presented in detail and their related applications in biogas production processes are summarized and analyzed. The third section covers evolutionary algorithms including Genetic Algorithm (GA), Ant Colony Optimization (ACO), and Particle Swarm Optimization (PSO) and their applications for optimizing biogas production processes. Hybrid models like Neuro-fuzzy, Fuzzy-Neural, and Neuro-Evolutionary approaches are discussed in the last section and their applications in anaerobic digestion systems are also summarized and scrutinized.