A perspective-driven and technical evaluation of machine learning in bioreactor scale-up: A case-study for potential model developments.

Bioreactor scale-up and scale-down have always been a topical issue for the biopharmaceutical industry and despite considerable effort, the identification of a fail-safe strategy for bioprocess development across scales remains a challenge. With the ubiquitous growth of digital transformation technologies, new scaling methods based on computer models may enable more effective scaling. This study aimed to evaluate the potential application of machine learning (ML) algorithms for bioreactor scale-up, with a specific focus on the prediction of scaling parameters. Factors critical to the development of such models were identified and data for bioreactor scale-up studies involving CHO cell-generated mAb products collated from the literature and public sources for the development of unsupervised and supervised ML models. Comparison of bioreactor performance across scales identified similarities between the different processes and primary differences between small- and large-scale bioreactors. A series of three case studies were developed to assess the relationship between cell growth and scale-sensitive bioreactor features. An embedding layer improved the capability of artificial neural network models to predict cell growth at a large-scale, as this approach captured similarities between the processes. Further models constructed to predict scaling parameters demonstrated how ML models may be applied to assist the scaling process. The development of data sets that include more characterization data with greater variability under different gassing and agitation regimes will also assist the future development of ML tools for bioreactor scaling.

Biochemical production of bioenergy from agricultural crops and residue in Iran.

The present study assessed the potential for biochemical conversion of energy stored in agricultural waste and residue in Iran. The current status of agricultural residue as a source of bioenergy globally and in Iran was investigated. The total number of publications in this field from 2000 to 2014 was about 4294. Iran ranked 21st with approximately 54 published studies. A total of 87 projects have been devised globally to produce second-generation biofuel through biochemical pathways. There are currently no second-generation biorefineries in Iran and agricultural residue has no significant application. The present study determined the amount and types of sustainable agricultural residue and oil-rich crops and their provincial distribution. Wheat, barley, rice, corn, potatoes, alfalfa, sugarcane, sugar beets, apples, grapes, dates, cotton, soybeans, rapeseed, sesame seeds, olives, sunflowers, safflowers, almonds, walnuts and hazelnuts have the greatest potential as agronomic and horticultural crops to produce bioenergy in Iran. A total of 11.33million tonnes (Mt) of agricultural biomass could be collected for production of bioethanol (3.84gigaliters (Gl)), biobutanol (1.07Gl), biogas (3.15billion cubic meters (BCM)), and biohydrogen (0.90BCM). Additionally, about 0.35Gl of biodiesel could be obtained using only 35% of total Iranian oilseed. The potential production capacity of conventional biofuel blends in Iran, environmental and socio-economic impacts including well-to-wheel greenhouse gas (GHG) emissions, and the social cost of carbon dioxide reduction are discussed. The cost of emissions could decrease up to 55.83% by utilizing E85 instead of gasoline. The possible application of gaseous biofuel in Iran to produce valuable chemicals and provide required energy for crop cultivation is also studied. The energy recovered from biogas produced by wheat residue could provide energy input for 115.62 and 393.12 thousand hectares of irrigated and rain-fed wheat cultivation in Iran, respectively. The nitrogen requirement for 33.6% of the total wheat cultivation area could be supplied by the ammonia acquired from biohydrogen. A discussion of the logistics of collection and transportation of the biomass and sensitivity analysis are carried out to evaluate the effect of field cover factor, crop yield, and well-to-wheel GHG emission on collectable residue, biofuel production, and GHG emissions.

A combined model for large scale batch culture MFC-digester with various wastewaters through different populations.

In this study, a new model of microbial fuel cell (MFC) was obtained for the first time. The modeled MFC was made using a combination of two approaches; the conduction-based method and two-step anaerobic digestion. Performance of the MFC was based on calculations for current evolution and polarization curves with different subsequent variables of the biofilm and anolyte. The model was able to make predictions for performance of the MFC for a simple substrate to more complex ones. The model was successfully validated with a variety of substrates (acetate, glucose and dairy wastewater) and the results were compared with previously published measurements. The model polarization results showed that is able to predict overshoot as a dynamic phenomenon. The ratio of acetoclastic methanogens to electrogens in the biofilm increased from an average value of 0.63×10(-2) to 1.17×10(-2) by increasing external resistance from 50 Ω to 100Ω . The attached to planktonic cells ratio was computed 0.45 for the glucose-fed MFC and for the dairy wastewater-fed MFC at 50 Ω was 8.86 and at 100 Ω was 5.46.