A kinetic model to describe lovastatin biosynthesis by Aspergillus terreus ATCC 20542 in a batch culture with the simultaneous use of lactose and glycerol as carbon sources was developed. In order to do this the kinetics of the process was first studied. Then, the model consisting of five ordinary differential equations to balance lactose, glycerol, organic nitrogen, lovastatin and biomass was proposed. A set of batch experiments with a varying lactose to glycerol ratio was used to finally establish the form of this model and find its parameters. The parameters were either directly determined from the experimental data (maximum biomass specific growth rate, yield coefficients) or identified with the use of the optimisation software. In the next step the model was verified with the use of the independent sets of data obtained from the bioreactor cultivations. In the end the parameters of the model were thoroughly discussed with regard to their biological sense. The fit of the model to the experimental data proved to be satisfactory and gave a new insight to develop various strategies of cultivation of A. terreus with the use of two substrates.

The purpose of this study was to validate the applicability of specialized microbial consortium for the degradation of lipids in wastewater. An experimental model of the process is proposed that enables prediction of the required batch length. This model can be used for supervision of the process and to control cycles of the batch reactor. The study involved 4 reactors with microbial consortium obtained by inoculation from a commercially available biopreparate. Each reactor was fed a different load of lipid containing substrate. The biodiversity, settling characteristics and COD reductions were measured. The biodiversity of the microbial consortium changed within a range of ±15% depending on lipids concentration, as shown by the Shannon index and increasing amount of β-proteobacteria. Higher concentrations of lipids increased the biodiversity suggesting higher growth of microorganisms capable of utilizing lipids as energy and carbon source by producing lipid hydrolyzing enzymes. High lipid concentrations degrade the settling capabilities of the biomass. Higher lipid concentrations (0.5–2.0 [g/l]) increase the final COD (1445–2160 [mg O2/l]). The time necessary for substrate degradation changes with the initial concentration and can be predicted using the proposed model. The study showed that specialized microbial consortium is capable of reducing the lipids containing substrate and maintains its biodiversity suggesting that utilization of such consortia in multiple cycles of a batch reactor is possible. Future research should concentrate on assessing the biodiversity and effectiveness of substrate reduction after an increased number of batch reactor cycles.

A new method for measurement of sludge blanket height (SBH) based on image analysis is presented. The proposed method uses a histogram back-projection algorithm to distinguish between the settling sludge and supernatant and can be used with sludge possessing different coloring characteristics both in the sludge color and the color of supernatant produced. Individual pixels in the acquired image are compared with a histogram of a representative sludge region. Therefore, the proposed method relies neither on the assumed shape of light intensity profile nor on the dominant sludge or supernatant color. Batch sedimentation tests are presented for different initial sludge concentrations and different background colors to simulate different sludge characteristics. Parameters of a settling velocity function are estimated based on the obtained results. Additionally, an algorithm is proposed that enables the zone settling velocity (ZSV) to be estimated before the batch sedimentation test is completed.

Biosynthesis of lovastatin (a polyketide metabolite of Aspergillus terreus) in bioreactors of different working volume was studied to indicate how the change of scale of the process influences the formation of this metabolite. The experiments conducted in shake flasks of 150 ml working volume allowed to obtain lovastatin titres at the level of 87.5 mg LOV l-1, when two carbon sources, namely lactose and glycerol were used. The application of the same components in a large stirred-tank bioreactor of 5.3-litre working volume caused a decrease of lovastatin production by 87% compared to the shake flask culture. The deficiency of nitrogen in this bioreactor did not favour the formation of lovastatin, in contrast to the small bioreactor of 1.95-litre working volume, in which lovastatin titres comparable to those in the shake flasks could be achieved, when organic nitrogen concentration was two-fold decreased. When the control of pH and/or pO2 was used simultaneously, an increase in lovastatin production was observed in the bioreactors. However, these results were still slightly lower than lovastatin titres obtained in the shake flasks.

Saccharamyces cerevisia known as baker’s yeast is a product used in various food industries. Worldwide economic competition makes it a necessity that industrial processes be operated in optimum conditions, thus maximisation of biomass in production of saccharamyces cerevisia in fedbatch reactors has gained importance. The facts that the dynamic fermentation model must be considered as a constraint in the optimisation problem, and dynamics involved are complicated, make optimisation of fed-batch processes more difficult. In this work, the amount of biomass in the production of baker’s yeast in fed-batch fermenters was intended to be maximised while minimising unwanted alcohol formation, by regulating substrate and air feed rates. This multiobjective problem has been tackled earlier only from the point of view of finding optimum substrate rate, but no account of air feed rate profiles has been provided. Control vector parameterisation approach was applied the original dynamic optimisation problem which was converted into a NLP problem. Then SQP was used for solving the dynamic optimisation problem. The results demonstrate that optimum substrate and air feeding profiles can be obtained by the proposed optimisation algorithm to achieve the two conflicting goals of maximising biomass and minimising alcohol formation.

In recent years, deep learning and especially deep neural networks (DNN) have obtained amazing performance on a variety of problems, in particular in classification or pattern recognition. Among many kinds of DNNs, the convolutional neural networks (CNN) are most commonly used. However, due to their complexity, there are many problems related but not limited to optimizing network parameters, avoiding overfitting and ensuring good generalization abilities. Therefore, a number of methods have been proposed by the researchers to deal with these problems. In this paper, we present the results of applying different, recently developed methods to improve deep neural network training and operating. We decided to focus on the most popular CNN structures, namely on VGG based neural networks: VGG16, VGG11 and proposed by us VGG8. The tests were conducted on a real and very important problem of skin cancer detection. A publicly available dataset of skin lesions was used as a benchmark. We analyzed the influence of applying: dropout, batch normalization, model ensembling, and transfer learning. Moreover, the influence of the type of activation function was checked. In order to increase the objectivity of the results, each of the tested models was trained 6 times and their results were averaged. In addition, in order to mitigate the impact of the selection of learning, test and validation sets, k-fold validation was applied.

An increasing number of municipal sewage treatment plants in Poland, desirable from an environmental perspective, raises the problem of managing the growing volume of sewage sludge. The thermal treatment of municipal sewage sludge (TTMSS) method, by greatly reducing the waste volume, increases the heavy metal concentration in fly ash (primary, end product of the treatment process), which may constitute a risk factor when attempting to utilize them economically. The research paper concentrates on determining the TTMSS fly ash heavy metal leaching level. For this purpose, ash samples were subjected to leaching with the batch and percolation tests, and the heavy metal content in eluates was determined by the FAAS method. The obtained results served as a base to determine the level of heavy metal immobilization in the ash, the element release mechanism (percolation test), and the impact of the L/S (liquid to solid) ratio and pH on the heavy metal leaching intensity (percolation test). The conducted research indicated high immobilization of heavy metals in TTMSS fly ash, regardless of the applied study method, which corresponds to the results of other researchers. Lead was the most intensively eluted metal.

In this work, a design equation was presented for a batch-recirculated photoreactor composed of a packed bed reactor (PBR) with immobilised TiO2-P25 nanoparticle thin films on glass beads, and a continuous-flow stirred tank (CFST). The photoreactor was studied in order to remove C.I. Acid Orange 7 (AO7), a monoazo anionic dye from textile industry, by means of UV/TiO2 process. The effect of different operational parameters such as the initial concentration of contaminant, the volume of solution in CFST, the volumetric flow rate of liquid, and the power of light source in the removal efficiency were examined. A rate equation for the removal of AO7 is obtained by mathematical kinetic modelling. The results of reaction kinetic analysis indicate the conformity of removal kinetics with Langmuir-Hinshelwood model (kL-H = 0.74 mg L-1 min-1, Kads = 0.081 mg-1 L). The represented design equation obtained from mathematical kinetic modelling can properly predict the removal rate constant of the contaminant under different operational conditions (R2 = 0.963). Thus the calculated and experimental results are in good agreement with each other.