Abstract This paper investigates the use of machine learning (ML) techniques to study the effect of different process conditions on ethanol production from lignocellulosic sugarcane bagasse biomass using S. cerevisiae in a simultaneous hydrolysis and fermentation (SHF) process. The effects of temperature, enzyme concentration, biomass load, inoculum size and time were investigated using artificial neural networks, a C5.0 classification tree and random forest algorithms. The optimization of ethanol production was also evaluated. The results clearly depict that ML techniques can be used to evaluate the SHF (R2 between actual and model predictions higher than 0.90, absolute average deviation lower than 8.1% and RMSE lower than 0.80) and predict optimized conditions which are in close agreement with those found experimentally. Optimal conditions were found to be a temperature of 35 ºC, an SHF time of 36 h, enzymatic load of 99.8%, inoculum size of 29.5 g/L and bagasse concentration of 24.9%. The ethanol concentration and volumetric productivity for these conditions were 12.1 g/L and 0.336 g/L.h, respectively.

AbstractBiosurfactants have a wide range of applications in emulsions, separations, and solubilization because these chemicals reduce the surface tension and viscosity of solutions. This work studied rhamnolipid production using a batch bioreactor with a working volume of 1.5 liters, Pseudomonas aeruginosa ATCC (American Type Culture Collection) 10145 strain, and the greasy effluent from the slaughter of poultry and pigs as the substrate. The main goal of this research was to evaluate the level of aeration, agitation speed and inoculum concentration using a Central Composite Design (CCD). Experimental conditions were selected using the surface response technique obtained from the CCD, and the results were validated to test the reproducibility. The following operating conditions were selected: 1.2 vvm level of aeration, 600 rpm agitation speed, and 1.0 g/L biomass inoculum concentration. Under these conditions, the following results were obtained: the rhamnose production, surface tension and emulsifying index were 5.37 g/L, 25.6 dyne/cm and 100%, respectively.

An analysis was made of the use of whey fermentation by Lactobacillus acidophilus LA-5 for encapsulated probiotic bacteria cell production. Fermentation was done in a 2-liter Biostat B Fermentor at 28±1 ºC without air supply and agitation maintained at 200 rpm. Different processing conditions were studied using Center Composite Design applied to Surface Response Methodology. Maximum cell yield (2.7 x10(10) NMP/mL for 36 hours) was achieved with 30.85 g/L of lactose, a pH value of 6.45 and 1.04 g/L of inoculum. Cell growth was evaluated using reconstituted and fresh whey after 144 hours of fermentation in pre-optimized conditions. Cell concentration after fermentation was 10(10) MPN/mL in all the assays. The Verhulst model proved to be satisfactory to fit the experimental results, providing a stationary cell concentration of 6.0 g/L and a specific growth rate of 0.09 h-1. Cells were collected by centrifugation at 15000g for 5 minutes at 4 ºC, immobilized in 2% alginate, and dried to a constant weight at 50 ºC. Immobilized probiotic cells presented 10(11) MPN/g, a time required to kill 90% of the organisms (D value) of 18 h (70 ºC), an activation energy of 76.04 kJ/mol for thermal inactivation, and an in vitro resistance to low pH (D value of 62.5 min at 37 ºC, pH 2.5).