Quantitative structure‐activity relationship models (QSAR models) predict the physical properties or biological effects based on physicochemical properties or molecular descriptors of chemical structures. Our work focuses on the construction of optimal linear and nonlinear weighted mixes of individual QSAR models to more accurately predict their performance. How the splitting of the application domain by a nonlinear gating network in a “mixture of experts” model structure is suitable for the determination of the optimal domain‐specific QSAR model and how the optimal QSAR model for certain chemical groups can be determined is highlighted. The input of the gating network is arbitrarily formed by the various molecular structure descriptors and/or even the prediction of the individual QSAR models. The applicability of the method is demonstrated on the pK normala values of the OASIS database (1912 chemicals) by the combination of four acidic pK normala predictions of the OECD QSAR Toolbox. According to the results, the prediction performance was enhanced by more than 15% (root‐mean‐square error [RMSE] value) compared with the predictions of the best individual QSAR model.

Gas chromatography (GC) is an effective tool for the analysis of complex mixtures with a huge number of components. To keep tracking the chemical changes during the processes like plastic waste pyrolysis usually different sample states are profiled, but retention time drifts between the chromatograms make the comparability difficult. The aim of this study is to develop a fast and simple method to eliminate the time drifts between the chromatograms using easily accessible priori information. The proposed method is tested on GC chromatograms obtained by analysis of pyrolysis product (Mg/Y catalyst) of shredded real waste HDPE/PP/LDPE mixture. A modified k-means algorithm was developed to account the retention time drifts between samples (different sample states). The outcome of the retention time alignment is an averaged retention time for each peak from all the chromatograms which makes the comparison and further analysis (such as "fingerprinting") easier or possible.

No abstract

No abstract

In this work, the catalyst deactivation phenomenon in the case of special hydrocracking of sunflower oil and kerosene mixture was analyzed based on experiments and models. Alternative (bio/waste-originated) fuels are becoming more important to reduce the full life cycle environmental pollution of transportation. One of the production possibilities of these alternative fuels is the co-processing (i.e., catalytic quality improvement) of fossil and biobased feedstocks. The huge number of individual chemical components in the system increases the complexity of the investigation. Moreover, from the chemical analysis viewpoint, only some of these can be followed during the experiments. Hence, the models to describe the processes in this system (hydrocracker) are usually based on lumps (i.e., component groups). Experimental data are reported on the special catalytic hydrocracking of sunflower oil and kerosene mixture for the production of high-quality aviation fuel (JET). Experiments were carried out in a fixed-bed tubular reactor over temperatures between 533 and 613 K, pressure ranging from 30 to 70 bar, and liquid hourly space velocities of 1.0 and 2.0 h −1 , employing a Pt/H-mordenite catalyst. The objective of our work is to investigate the existing lumped models that are suitable for describing the experimental data; moreover, while maintaining the possible reaction pathways, model parameters were identified and validated against the measurements. As a result of the analysis of the measurement data, it has been established that in the case of lower liquid load/higher residence time, a deactivation phenomenon, the so-called catalyst fouling, takes place on the applied catalyst. Three catalyst deactivation models were developed and integrated into the kinetic model: the Levenspiel deactivation kinetic model, a simplified Eley−Rideal mechanism, and the last one based on competitive adsorption.