H_∞ Control of Anaerobic Digester for Winery Industry Wastewater Treatment

Abstract

A robust H ∞ controller has been developed to regulate the chemical oxygen demand in an anaerobic digester from the winery industry. A sensitivity analysis was performed, and the parameter set having the most significant effect on the process behavior was identified. The parameters inducing the most sensitivity in the solutions were selected as uncertain; in addition, they were related to kinetic terms and he hydrodynamic regime. Then, a control problem was formulated as robust regulation, and a controller was designed using H ∞ theory to ensure robust stability. The actions of the H ∞ controller are illustrated through numerical simulations. The controller was found to execute robust regulation facing parametric uncertainties and load disturbances.

An adaptive observer for operation monitoring of anaerobic digestion wastewater treatment

“…The results contribute on deriving reliable information for monitoring and controlling AD. The biomass and kinetics can be used on improving the performance and robustness of feedback control schemes as that reported by [1,3]. This is an interesting advance towards the widespread implementation of advance control laws as green technology in AD.…”

“…Since such a model has allowed one the control synthesis [1][2][3], the model is a suitable one to design monitoring strategies based on the adaptive observers towards the operation decision (as, for example, tolerance margin in load disturbances or modifications of set-points).…”

“…The assumptions are general for AD wastewater treatment but needed for mathematical formalism in the sense of ensuring the global convergence of the adaptive observer, the persistent excitation of the AD process by inputs, and both boundedness and uniqueness of the solutions for model (1). Complementary, the theoretical framework of the adaptive observer requires the system is written in cascade form and the observability be ensured to derive main result.…”

“…This fact corresponds to the input persistent excitation. Now, the respect to Assumption A2, the fluctuation at feeding composition has the additional effect onto the output, which is persistently fluctuating around reference in a realistic implementation (for instance see experimental results at [1][2][3]. That is, the outlet substrate fluctuates and, consequently, Q ðsÞ; KðsÞ fluctuate and are bounded.…”

“…Technical alternatives have been reported in regard to handle with the first three issues. Among others, the control theory has been used to design feedback controllers capable of handling the nonlinear behavior and uncertainty on the kinetic terms [1][2][3]. But, although spectroscopy-based instrumentation schemes have been also explored as alternative for online measurements for COD, TOC and VFA [4], few theoretical frameworks can be found in regards to reliable schemes for advising AD monitoring and control.…”