Laser reflectance measurement for the online monitoring of Chlorella sorokiniana biomass concentration

•A method for the rapid online estimation of Chlorella sorokiniana biomass concentration is proposed.•The method is based on the processing of laser reflectance data through machine learning techniques.•The method requires low calibration effort and can be readily applied in industrial facilities.•It is applicable to other strains and to turbid suspensions of biomass with tendency to aggregate.

Fast and reliable methods to determine biomass concentration are necessary to facilitate the large scale production of microalgae. A method for the rapid estimation of Chlorella sorokiniana biomass concentration was developed. The method translates the suspension particle size spectrum gathered though laser reflectance into biomass concentration by means of two machine learning modelling techniques. In each case, the model hyper-parameters were selected applying a simulated annealing algorithm. The results show that dry biomass concentration can be estimated with a very good accuracy (R2 = 0.87). The presented method seems to be suited to perform fast estimations of biomass concentration in suspensions of microalgae cultivated in moderately turbid media with tendency to aggregate.