Ultrahigh pressure filtration dewatering of municipal sludge based on microwave pretreatment.

Municipal sludge is difficult to treat and dispose of because of its high moisture content (MC) and volume. In this study, a novel dewatering method that utilizes ultrahigh pressure filtration (UHPF) and thin-cake-press (TCP) coupled with microwave pretreatment was proposed to reduce the MC of sludge cake. The influence of microwave contact time (MCT), microwave intensity (MI), initial MC of sludge, dewatering time, applied pressure and sludge weight (Ws) on the MC of a sludge cake was investigated by the single-factor experiment. Moreover, a water discharge path model was developed to understand the dewatering mechanism and explain the relationship between thickness and MC of the cake. The key factors affecting the MC of the cake were explored by the orthogonal experiment. The experimental results showed that microwave irradiation could effectively improve the dewatering performance and reduce the MC of the sludge cake. The MC of the cake reached its lowest value of 28% at MCT of 120 s and MI of 400 W, which is much lower than the value obtained by the traditional dewatering method. Among the parameters mentioned above, Ws has the most significant influence on MC because a large amount of sludge leads to a thicker cake, which seriously hampers the flow of filtrate and inevitably increases the MC of the cake.

Vision transformer with masked autoencoders for referable diabetic retinopathy classification based on large-size retina image.

Computer-aided diagnosis systems based on deep learning algorithms have shown potential applications in rapid diagnosis of diabetic retinopathy (DR). Due to the superior performance of Transformer over convolutional neural networks (CNN) on natural images, we attempted to develop a new model to classify referable DR based on a limited number of large-size retinal images by using Transformer. Vision Transformer (ViT) with Masked Autoencoders (MAE) was applied in this study to improve the classification performance of referable DR. We collected over 100,000 publicly fundus retinal images larger than 224×224, and then pre-trained ViT on these retinal images using MAE. The pre-trained ViT was applied to classify referable DR, the performance was also compared with that of ViT pre-trained using ImageNet. The improvement in model classification performance by pre-training with over 100,000 retinal images using MAE is superior to that pre-trained with ImageNet. The accuracy, area under curve (AUC), highest sensitivity and highest specificity of the present model are 93.42%, 0.9853, 0.973 and 0.9539, respectively. This study shows that MAE can provide more flexibility to the input image and substantially reduce the number of images required. Meanwhile, the pretraining dataset scale in this study is much smaller than ImageNet, and the pre-trained weights from ImageNet are not required also.

A Study on the Through-Plane Permeability of Anisotropic Fibrous Porous Material by Fractal Stochastic Method.

The through-plane permeability is of great importance for understanding the transport phenomenon in anisotropic fibrous porous material. In this paper, a novel pore-scale model based on the equilateral triangle representative unit cell (RUC) and capillary bundle model is developed for the fluid flow through the anisotropic fibrous porous material according to fractal theory, and the effective through-plane permeability is presented accordingly. The digital structures of the fibrous porous material are generated by a fractal stochastic method (FSM), and the single-phase fluid flow through the 3D-reconstructed model is simulated by using the finite element method (FEM). It was found that the effective through-plane permeability depends on the fiber column size, porosity, and fractal dimensions for pore and tortuosity. The results show that the predicted through-plane permeability by the present fractal model indicates good agreement with numerical results and available experimental data as well as empirical formulas. The dimensionless through-plane permeability is positively correlated with the porosity and negatively correlated with the fractal dimensions for pore and tortuosity at certain porosity.

Pore-scale model and dewatering performance of municipal sludge by ultrahigh pressurized electro-dewatering with constant voltage gradient.

The disposal of huge municipal sludge with high moisture content has led to numerous energy consumption and brought extensive concerns in the world. In this paper, three dewatering modes, ultrahigh-pressure mechanical dewatering mode (UMDW), pressurized electro-dewatering (PEDW) with constant voltage mode (U-PEDW) and constant voltage gradient mode (G-PEDW) were performed on a self-designed pressurized electro-dewatering apparatus for municipal sludge. The pore structures and moisture distributions were detected by low-field nuclear magnetic resonance technology. Meanwhile, the moisture distributions and quantitative bound strength were analyzed by the thermogravimetric differential scanning calorimetry test. Moreover, a pore-scale electro-osmosis model was accordingly developed based on the fractal characteristics of pore size distribution. Finally, the effect of dewatering modes and operating parameters on moisture content and energy consumption were examined in detail. The results indicate that the pore-scale electro-osmosis model show good consistency with experimental data. The electric field can drive the middle-layers-water to overcome the capillary pressure, and make G-PEDW removing more water than UMDW. The moisture content of dewatered municipal sludge by G-PEDW and U-PEDW reaches to 28.41% and 27.33%, respectively. Furthermore, the energy consumption of G-PEDW is 189.62Wh/kg.H2O, it is much lower than that of U-PEDW. Therefore, the G-PEDW mode with low moisture content and less energy consumption indicates best dewatering performance compared with UMDW and U-PEDW modes. The present work is helpful to understand the dewatering mechanisms of G-PEDW and provides useful guidelines for G-PEDW dewatering technology.

Meso-mechanism of mechanical dewatering of municipal sludge based on low-field nuclear magnetic resonance.

Both huge volume and high moisture content of municipal sludge have brought great troubles and attracted extensive concerns in the world. In this paper, the mechanical press filtration (MPF) dewatering was performed under ultrahigh pressure in order to improve the dewatering performance of municipal sludge. Low-Field Nuclear Magnetic Resonance (NMR) technique was used to study the effect of MPF parameters on dewatering performance. Based on the pore characteristics of municipal sludge, a capillary bundle model was developed to explore the mesoscopic mechanisms of MPF dewatering. The results indicate that moisture content of sludge cake decreases gradually with the increase of compressed pressure and dewatering time as well as the decrease of sludge weight , and the moisture content of municipal sludge can be reduced to as low as 30% with MPF dewatering. According to the peak and envelope area of relaxation time curve in NMR, it can be found that the size of pores decreases and the percentage of small pores increases during the dewatering process. Therefore, the capillary water cannot be removed by MPF method because the capillary pressure in very small pores increases evidently. Furthermore, the compaction degree of the outmost layer is generally higher than that of the middle layer, and the difference of moisture content between the middle and outmost layers of cake is 25%-28.4%. The present work may be helpful to understand the MPF dewatering mechanism and shed light on the new dewatering techniques of municipal sludge.

The adsorption of NO, NH3, N2 on carbon surface: a density functional theory study.

To explore the adsorption mechanism of NO, NH3, N2 on a carbon surface, and the effect of basic and acidic functional groups, density functional theory was employed to investigate the interactions between these molecules and carbon surfaces. Molecular electrostatic potential, Mulliken population analyses, reduced density gradient, and Mayer bond order analyses were used to clarify the adsorption mechanism. The results indicate that van der Waals interactions are responsible for N2 physisorption, and N2 is the least likely to adsorb on a carbon surface. Modification of carbon materials to decorate basic or acidic functional groups could enhance the NH3 physisorption because of hydrogen bonding or electrostatic interactions, however, NO physisorption on a carbon surface is poor. Zig-zag sites are more reactive than armchair sites when these gas molecules absorb on the edge sites of carbon surface. Graphical abstract NH3, N2, NO adsortion on carbon surface.