Optimal fault tolerant control strategy of a continuous fermentative bioprocess for biogas production

This is the authors' abstract. We don't add key points for this paper.

Fermentative system for biogas production is a nonlinear multistep degradation process where complex organic wastes are progressively transformed into low molecular weight metabolites by different bacteria. This process is economic viable due to high rates of biogas production, recovery of high-added value by-products and the low-cost renewable sources such as organic waste and wastewater. Due to the nonlinear nature of the process, the efficiency of biogas production could be affected by the pH, metabolic pathways and the associated microbiome. Optimal operation of the hydraulic retention time and organic loading rate are relevant for the activity of biogas producing bacteria. In this work, a fault tolerant control strategy for a fermentative system subject to fault actuators is proposed. A recurrent neural network model for unknown nonlinear systems is used to estimate dynamical sates and fault magnitude. An inverse optimal control law for trajectory tracking based on the neural model is proposed for fault compensation. Adequate closed loop control action and reference trajectories are selected to lead the system to optimal operating order, with respect to desirable performances and degree of priority. The main objective is to guarantee the productivity in a safe and a profitable plant operation. The strategy is applied to the fermentative process in presence of fault actuators and their efficiency is demonstrated via simulations.