This paper presents experimental results of a lab-scale implementation of an extremum seeking control strategy for maximizing the biomass productivity of cultures of the micro-algae Dunaliella tertiolecta in a flat-panel photobioreactor operated in continuous mode. The real-time optimization is based on a recursive least squares adaptation where the input (Dilution rate)/output (Biomass concentration) relation is approximated by a linear Hammerstein regression from which a productivity gradient estimate can be inferred. Lab-scale instrumentation and operating conditions are described and the results of two experiments are presented. They demonstrate the fast convergence of the extremum seeking scheme and practical considerations related to stability are discussed.
All data in the Integrated Marine Information System (IMIS) is subject to the VLIZ privacy policy
