A Deep Learning-Based Unbalanced Force Identification of the Hypergravity Centrifuge.

Accurate and quantitative identification of unbalanced force during operation is of utmost importance to reduce the impact of unbalanced force on a hypergravity centrifuge, guarantee the safe operation of a unit, and improve the accuracy of a hypergravity model test. Therefore, this paper proposes a deep learning-based unbalanced force identification model, then establishes a feature fusion framework incorporating the Residual Network (ResNet) with meaningful handcrafted features in this model, followed by loss function optimization for the imbalanced dataset. Finally, after an artificially added, unbalanced mass was used to build a shaft oscillation dataset based on the ZJU-400 hypergravity centrifuge, we used this dataset to train the unbalanced force identification model. The analysis showed that the proposed identification model performed considerably better than other benchmark models based on accuracy and stability, reducing the mean absolute error (MAE) by 15% to 51% and the root mean square error (RMSE) by 22% to 55% in the test dataset. Simultaneously, the proposed method showed high accuracy and strong stability in continuous identification during the speed-up process, surpassing the current traditional method by 75% in the MAE and by 85% in the median error, which provided guidance for counterweight and guaranteed the unit's stability.

Adsorption of NH4+-N on Chinese loess: Non-equilibrium and equilibrium investigations.

NH4+-N is a crucial pollutant in landfill leachate and can be in high concentrations for a long period of time due to anaerobic condition of landfills. The adsorption properties of NH4+-N on the Chinese loess were investigated using Batch test. The influences of ammonium concentration, temperature, reaction time, slurry concentration, and pH on the adsorption process are evaluated. Adsorption kinetics and isotherm behaviors were studied by applying different models to the test data to determine the adsorption parameters. The equilibrating duration was shown to be less than 60 min. The data on adsorption kinetics can be well fitted by the pseudo-second-order kinetics model. According to the Langmuir isotherm model, the adsorption capacity of Chinese loess about NH4+-N was predicted to be 72.30 mg g-1. The uptake of NH4+-N by Chinese loess was considered to be the type of physical adsorption on the basis of D-R isotherm analysis. The optimal pH and slurry concentration are 4 and 2 g/50 ml, respectively. According to the calculated values of free energy, enthalpy and entropy change, the adsorption process is determined to be exothermic. The disorder of the system appeared lowest at temperature of 308.15 K. The predicted Gibb's free energies also indicate the adsorption process is endothermic and spontaneous. The FTIR spectrum and EDX analysis showed the adsorption process of NH4+ involves cation exchange and dissolution of calcite.

Lead adsorption on loess under high ammonium environment.

Lead (Pb) is one of the most toxic, hazardous pollutants available in landfill leachate. Loess-amended soil buffers are found suitable and effective in attenuating migration of Pb and the other trace metals. High concentration of ammonium (NH4+ > 1000 mg/l) is also reported in landfill leachate, and therefore, it is essential to investigate the transport of lead under such condition. In this study, the mechanisms and the capacity of loess to adsorb Pb under high NH4+ concentration were investigated. Adsorption isotherm test data were obtained for 25 °C, 35 °C and 45 °C. The maximum adsorption capacity is estimated to be 2101.97 mg/g at 25 °C and 4292.8 mg/g at 45 °C under 1000 mg/l NH4+. The binding sites of Pb on loess are positively related to each other at low temperatures (25-35 °C). The thermodynamic analysis indicates that adsorption process is endothermic and non-spontaneous and the system randomness increases with reaction time. The kinetic test data, fitted with a pseudo-second-order kinetic model and an intraparticle diffusion model, suggests that removal of Pb is driven by both membrane and intraparticle diffusions. The SEM, XRD and FTIR analyses indicate flocculation, precipitations as well as some ion exchange processes, which perhaps combinedly increases adsorption of both NH4+ and Pb in loess. The two kinds of precipitations are involved for the removal of Pb. The precipitations of PbCO3, Pb(OH)2 and PbCO3·2H2O are formed by the reactions between calcite and lead. The other precipitation of white basic salt (Pb2O(NO3)2) is formed by the reactions among Pb2+, NO3- and aqueous ammonia under alkaline environment of loess slurry.