Gas–Liquid Two-Phase Flow Pattern Identification of a Centrifugal Pump Based on SMOTE and Artificial Neural Network

The accurate identification of the gas–liquid two-phase flow pattern within the impeller of a centrifugal pump is critical to develop a reliable model for predicting the gas–liquid two-phase performance of the centrifugal pump. The influences of the inlet gas volume fraction, the liquid phase flow rate and the pump rotational speed on the flow characteristics of the centrifugal pump were investigated experimentally. Four typical flow patterns in the impeller of the centrifugal pump, i.e., the bubble flow, the agglomerated bubble flow, the gas pocket flow and the segregated flow, were obtained, and the corresponding flow pattern maps were drawn. After oversampling based on the SMOTE algorithm, a four-layer artificial neural network model with two hidden layers was constructed. By selecting the appropriate network super parameters, including the neuron numbers in the hidden layer, the learning rate and the activation function, the different flow patterns in the centrifugal pump impeller were identified. The identification rate of the model increased from 89.91% to 94.88% when the original data was oversampled by the SMOTE algorithm. It is demonstrated that the SMOTE algorithm is an effective method to improve the accuracy of the artificial neural network model. In addition, the Kappa coefficient, the Macro-F1 and the Micro-F1 were 0.93, 0.95 and 0.95, respectively, indicating that the model established in this paper can well identify the flow pattern in the impeller of a centrifugal pump.

Study on Plant Growth and Nutrient Uptake under Different Aeration Intensity in Hydroponics with the Application of Particle Image Velocimetry

Aeration is considered beneficial for hydroponics. However, little information is available on the effects of aeration, and even less on solutions that use bubble flow and their agronomic effects. In this study, the effects of aeration intensity on plants were studied through cultivation experiments and flow field visualization. It was found that the growth of plants did not increase linearly with an increase in aeration intensity. From the results of this study, when the aeration intensity was within the low range (0.07–0.15 L·L−1 NS·min−1), increasing the aeration intensity increased the plant growth. However, after the aeration intensity reached a certain extent (0.15–1.18 L·L−1 NS·min−1), some indicators did not change significantly. When the aeration intensity continued to increase (1.18–2.35 L·L−1 NS·min−1), growth began to decrease. These results show that for increasing dissolved oxygen and promoting plant growth, the rule is not “the higher the aeration intensity, the better”. There is a reasonable range of aeration intensity within which crops grow normally and rapidly. In addition, increasing the aeration intensity means increasing energy utilization and operating costs. In actual hydroponics production, it is very important to find a reasonable aeration intensity range.

Experimental study of the local structure of a bubble flow in a flat channel with sudden expansion

The hydrodynamic structure of the flow in a flat channel with sudden expansion was studied at constant flow rates of liquid and gas in the vertical flow at Re = 6600 and gas content β = 0.03. The measurements were carried out using the PLIF method; and with this view, fluorescent particles for PIV studies and the dye Rhodamine G were added to distilled water. An optical threshold filter was installed on the lens of the video camera. When processing images to obtain data on the local gas content, only bubbles falling into the plane of the laser beam were considered (the boundary glows, casting a shadow).

Investigation on Bubble Diameter Distribution in Upward Flow by the Two-Fluid and Multi-Fluid Models

Bubble flow can be simulated by the two-fluid model and the multi-fluid model based on the Eulerian method. In this paper, the gas phase was further divided into several groups of dispersed phases according to the diameter by using the Eulerian-Eulerian (E-E) multi-fluid model. The diameters of bubbles in each group were considered to be the same, and their distributions were reorganized according to a specific probability density function. The experimental data of two kinds of bubble flow with different characteristics were used to verify the model. With the help of the open-source CFD software, OpenFOAM-7.x (OpenFOAM-7.0, produced by OpenFOAM foundation, Reading, England), the influences of the group number, the probability distribution function, and the parameters of different bubble diameters on the calculation results were studied. Meanwhile, the numerical simulation results were compared with the two-fluid model and the experimental data. The results show that for the bubble flow with the unimodal distribution, both the multi-fluid model and the two-fluid model can obtain the distribution of gas volume fraction along the pipe radius. The calculation results of the multi-fluid model agree with the experimental data, while those of the two-fluid model differ greatly from the experimental data, which verifies the advantage of the multi-fluid model in calculating the distribution of gas volume fraction in the polydisperse bubble flow. Meanwhile, the multi-fluid model can be used to accurately predict the distribution of the parameters of each phase of the bubble flow if the reasonable bubble diameter distribution is provided and the appropriate interphase force calculation model is determined.

A Novel On-Chip Liquid-Metal-Enabled Microvalve

A room temperature liquid metal-based microvalve has been proposed in this work. The microvalve has the advantages of easy fabrication, high flexibility, and a low leak rate. By designing a posts array in the channel, the liquid metal can be controlled to form a deformable valve boss and block the flow path. Besides, through adjustment of the pressure applied to the liquid metal, the microvalve can perform reliable switching commands. To eliminate the problem that liquid metal is easily oxidized, which causes the microvalve to have poor repeatability, a method of electrochemical cathodic protection has been proposed, which significantly increases the number of open/close switch cycles up to 145. In addition, this microvalve overcomes the shortcomings of the traditional microvalve that requires an alignment process to assemble all the parts. When the valve is closed, no leak rate is detected at ≤320 mbar, and the leak rate is ≤0.043 μL/min at 330 mbar, which indicates it has good tightness. As an application, we also fabricate a chip that can control bubble flow based on this microvalve. Therefore, this microvalve has great prospects in the field of microfluidics.

Dancing with Bubbles: Deterministic versus Probabilistic Bubble Models in Dense Phase Sand Fluidized Beds for Biomass Gasification

This study addresses the need to model bubble flow in a fluidized bed using a probabilistic approach, which includes intrinsic bubble flow randomness. It is shown that the proposed probabilistic predictive model (PPM) overcomes the limitations of deterministic correlations, commonly used to describe bubble dynamics in high-density (above 2000 kg/m3) sand-beds of Type B particles of the Geldart classification. It is proven that a PPM can describe the relationship between bubble axial chord and bubble rise velocity using minimum and maximum behavioral bands. This probabilistic model, which applies to a wide range of operating conditions, as shown in the present study, can be considered applicable to single bubbles injected at incipient fluidization, as well as to bubbly beds with and without loaded biomass pellets.