Application of Various Machine Learning Models for Process Stability of Bio-Electrochemical Anaerobic Digestion

The application of a machine learning (ML) model to bio-electrochemical anaerobic digestion (BEAD) is a future-oriented approach for improving process stability by predicting performances that have nonlinear relationships with various operational parameters. Five ML models, which included tree-, regression-, and neural network-based algorithms, were applied to predict the methane yield in BEAD reactor. The results showed that various 1-step ahead ML models, which utilized prior data of BEAD performances, could enhance prediction accuracy. In addition, 1-step ahead with retraining algorithm could improve prediction accuracy by 37.3% compared with the conventional multi-step ahead algorithm. The improvement was particularly noteworthy in tree- and regression-based ML models. Moreover, 1-step ahead with retraining algorithm showed high potential of achieving efficient prediction using pH as a single input data, which is plausibly an easier monitoring parameter compared with the other parameters required in bioprocess models.

Photocatalytic degradation of VOCs from air stream using Mo:TiO2/GAC nanocomposites

Modification of TiO2 is one of the techniques used to enhance its photodegradation efficiency and to make it visible-light-active. In this study, Mo-doped TiO2 nanoparticles were synthesized using a fast sol-gel technique, and then coated on granular activated carbon (GAC) as both substrate and adsorbent to obtain Mo:TiO2/GAC composite. The fabricated composite was characterized using powder XRD, SEM, EDAX, FTIR, and BET analysis. Then the composite was applied to photodegrade volatile organic compounds (VOCs) under both UV and visible light irradiation. The characterization results showed high crystallinity and purity. Mo:TiO2/GAC composite had higher photodegradation efficiency compared with bare TiO2 and bare GAC. Moreover, studying operational parameters showed that the optimum condition for photodegradation efficiency of VOCs was at flowrate of 1 l/min, VOCs concentration of 20 ppm, and light intensity of 400 and 600 W/m2 for UV and visible light respectively. The results suggest that Mo:TiO2/GAC is a visible-light-active composite and can be acceptably used to decompose VOCs under visible light with adequate efficiency and without the generation of harmful by-products such as O3 as compared with UV.

Statistical Method of Deducting Activation Energies for the Steam Methane Reforming Reactions

In this paper, a method of deducting activation energies for heterogeneous reactions of steam methane reforming is presented. The essence of the method lies in iterative evaluation of kinetic parameters, namely activation energies of reactions, for a given reactor. The novelty of the method lies in utilizing a statistical approach to reduce computational effort of numerical simulation. The method produces multivariable correlations between activation energies and operational parameters of the process: pressure, temperature, steam-to-methane ratio, residence time, and catalyst properties. These correlations can be used for numerical simulations of steam methane reforming to yield methane conversion rate, spatial and temporal distribution of reaction products, temperature and pressure within the reactor. An average computational effort is equal to a batch of 18 ([Formula: see text]) simulations for [Formula: see text] variables. The method was demonstrated by evaluating two-variable correlations of activation energies with pressure and temperature. The developed numerical model was validated against adopted experimental data.

Geomechanical Controls on Frac-Hits

The communication among the horizontal wells or "frac-hits" issue have been reported in several field observations. These observations show that the "infill" well fractures could have a tendency to propagate towards the "parent" well depending on reservoir in-situ conditions and operational parameters. Drilling the horizontal wells in a "staggered" layout with both horizontal and vertical offset could be a mitigation strategy to prevent the "frac-hits" issue. In this study, we present a detailed geomechanical modeling and analysis of the proposed solution. For numerical modeling, we used our state-of-the-art fully coupled poroelastic model "GeoFrac-3D" which is based on the boundary element method for the rock matrix deformation/fracture propagation and the finite element method for the fracture fluid flow. The "GeoFrac-3D" simulator fully couples pore pressure to stresses and allows for dynamic modeling of production/injection and fracture propagation. The simulation results demonstrate that production from a "parent’ well causes a non-uniform reduction of the reservoir pore pressure around the production fractures, resulting in an anisotropic decrease of the reservoir total stresses, which could affect fracture propagation from the "infill" wells. We examine the optimal orientation and position of the "infill" well based on the numerical analysis to reduce the "frac-hits" issue in the horizontal well refracturing. The posibility of "frac-hits" can be reduced by optimizing the direction and locations of the "infill" wells, as well as re-pressurizing the "parent" well. The results suggest that arranging the horizontal wells in a "staggered" or "wine rack" arrangement decreases direct well interference and could increase the drainage volume.

Optimization of Aggregate Production Circuit through Modeling of Crusher Operation

The paper concerns investigation of the effect of impact crusher operation on selected qualitative characteristics of mineral aggregate products. Qualitative characteristics of crushing products in terms of size reduction ratio and fine particles contents were analyzed from the point of view of operational parameters of the impact crusher. An investigative program was carried out on a plant scale and two primary parameters of the impactor were analyzed: velocity of the crusher rotor and the width of the outlet gap. The models of the crushing device operation were built separately for each type of the tested material, as well as for general conditions.

Investigations of The Hybrid Effects of The Operational Parameters In The Radial Forging Process Based On The Validation of The Hardness Test

In this study, considering all the parameters in radial forging and a three-dimensional model has been simulated using the finite element method. By implementing an elastoplastic state for the specimen tube, parameters such as friction type, residual stress distribution, effective strain distribution, material flow velocity and its effect on the neutral plate and the distribution of force in the die have been studied and analyzed. The effects of angle on the quality and characteristics of the specimen and the longevity of the die have also been obtained. Experimental results have been used to confirm the accuracy of the simulation. The results of the hardness test after forging were compared with the simulation results. Good agreement between the results indicates the accuracy of the simulation in terms of hardness. Therefore, this validation allows confirming the other obtained results for the analysis and prediction of various components in the forging process. After the validation and confirmation of the results through the hardness test, the hardness distribution was obtained by considering temperature changes and the effective strain on the specimen.