Efficiency Improvement in Entrained Flow Gasification

This paper presents the development of a Digital Twin (DTwin) to detect and localize the leaks in water distribution networks (WDNs), using single-stage and two-stage data-driven models. In the single-stage model, we test the anomalies in the dataset using Logistic Regression and Random Forest. In the two-stage model, a linear regression model predicts pressure differences between sensor pairs in the first stage. Based on this, we compute the distribution of residuals. In the second stage, changes in the residual distribution are classified using Multinomial Logistic Regression and Random Forest models to compute possible leak locations’ posterior probabilities. We have tested these models on a real-time dataset.

This paper focuses on the determination of the maximum amplification of blade response due to mistuning in multi stage assemblies. The modal optimization strategy developed earlier in connection with single stage models is extended here to multi stage configurations. Theoretical developments are carried out first and lead to the new upper limit of (1+N1+N2(g2/g1)+…)/2, where Ni are the number of blades on the stages and gi = FiTMi−1Fi with Fi the force vector applied on a sector of stage i and Mi its mass matrix. For identical stages, this maximum equals the Whitehead limit observed with single stages but with the number of blades equal to sum of the numbers of blades of the coupled stages. A computational validation of the theoretical results is achieved next on both a single degree of freedom per blade model and a reduced order model of a blisk. These numerical optimization efforts confirm the theoretical developments and demonstrate that such high amplification factors can indeed be achieved with small levels of mistuning. The effects of the number of blades on the different stages, damping in the system, stage coupling strength, etc are discussed in details.

Pressurized gasification of lignite in a pilot scale bubbling fluidized bed reactor with air, oxygen, steam and CO2 agents

Analysis of three-dimensional single stage and twodimensional multistage models of flows in turbomolecular pumps

Single stage and multistage classification models for the prediction of liver fibrosis degree in patients with chronic hepatitis C infection

The objective of this article is to discuss: (a) the various theoretical perspectives on mental health and mental health disorders adopted in sport psychology, and (b) how the adoption of these various theoretical perspectives in studies might impact upon the interpretations and conclusions in research about the mental health of participants in elite sports. Well-being as a target construct, holistic models, the single continuum or stage models, and Keyes' dual-continuum model of mental health are described, together with a sports psychiatric view of mental health. The strengths and limitations of various mental health perspectives are discussed. We conclude that mental health is a complex construct and that the sport psychology literature, much like the clinical psychology literature, has struggled to reach a consensus regarding a definition or a feasible approach to investigating mental health. For the researcher, it becomes important to make explicit the underlying theoretical perspective adopted and the operationalization upon which conclusions about elite athletes' mental health are based so that an increased knowledge base with high scientific credibility can be established and consolidated over time.

A dynamic reduced order model for simulating entrained flow gasifiers. Part II: Model validation and sensitivity analysis

Evaluation of the energy efficiency of the shell coal gasification process by coal type

Advances toward validating examiner writership opinion based on handwriting kinematics

Past researchers, examining the duration of joblessness, have failed to include a variable to account for the unemployed’s migration decision. By employing a simultaneous structure with a censored dependent variable and an endogenous binary migration variable, the present study demonstrates that single stage models, which do not control for migration, produce selectivity bias in the estimated duration equation. The empirical model used in this study provides estimated coefficients that differ significantly from those of single stage models. For example, most past studies conclude that nonwhites suffer longer periods of unemployment than whites. The findings from the present study indicate that, after controlling for geographic mobility, there is no statistical difference in unemployment duration between whites and nonwhites.

Object detection is a computer vision technique.It plays a significant role in healthcare, autonomous vehicles, facial recognition and traffic management system, license plate recognition and so on.Our paper will focus on the existing deep learning techniques based on convolutional neural networks and how these techniques are different from each other.These techniques include single stage models such as -YOLOV7, YOLOV8, and SSD.Our paper will compare these techniques to draw meaningful conclusions.With the help of these single stage models, we are implementing vehicle detection systems and then comparing these model's performance.In comparative analysis, we are using Coco dataset, we evaluated the performance of these techniques based on parameters such as: mean average precision, inference time and recall time.

Scale-up and energy estimations of single- and two-stage vinasse anaerobic digestion systems for hydrogen and methane production

Applications of artificial intelligence (AI) in medical imaging informatics have attracted broad research interest. In ophthalmology, for example, automated analysis of retinal fundus photography helps diagnose and monitor illnesses like glaucoma, diabetic retinopathy, hypertensive retinopathy, and cancer. However, building a robust AI model requires a large and diverse dataset for training and validation. While large number of fundus photos are available online, collecting them to create a clean, well-structured dataset is a difficult and manually intensive process. In this work, we propose a two-stage deep-learning system to automatically identify clean retinal fundus images and delete images with severe artifacts. In two stages, two transfer-learning models based the ResNet-50 architecture pre-trained using ImageNet data are built with Increased threshold values on SoftMax to reduce false positives. The first stage classifier identifies “easy” images, and the remaining “difficult” (or undetermined) images are further identified by the second stage classifier. Using the Google Search Engine, we initially retrieve 1,227 retinal fundus images. Using this two-stage deep-learning model yields a positive predictive value (PPV) of 98.56% for the target class compared to a single-stage model with a PPV of 95.74%. The two-stage model helps reduce by two-thirds the false positives for the retinal fundus image class. The PPV over all classes increases from 91.9% to 96.6% without compromising the number of images classified by the model. The superior performance of this two-stage model indicates that the building of an optimal training dataset can play an important role in increasing performance of deep-learning models.

About 200,000 bpd (barrel/day) vacuum residue oil is produced from oil refineries in Korea. These are supplied to use asphalt, high sulfur fuel oil, and upgrading residue hydro-desulfurization units. High sulfur fuel oil can be prepared by blending oil residue with light distillate to bring fuel oil characteristics in the range of commercial specifications, which will become more stringently restrictive in the near future in Korea. Vacuum residue oil has high energy content; however, due to its high viscosity, high sulfur content and high concentration of heavy metals are representative of improper low grade fuel, which is considered difficult to gasify. At present, over 20 commercial scale IGCC (Integrated Gasification Combined Cycle) plants using feedstocks with vacuum residue oil for gasification are under construction or operating stage worldwide. Recently, KIER (Korea Institute of Energy Research) has been studying the vacuum residue gasification process using an oxygen-blown entrained-flow gasifier. The experiment runs were evaluated under a reaction temperature of 1,200-1,250 C, reaction pressure of 1.0 kg/cm2, oxygen/V.R ratio of 0.8-1.2 and steam/V.R ratio of 0.4-0.7. Experimental results show a syngas composition (CO+H2), 77-88%; heating value, 2,300-2,600 kcal/Nm3; carbon conversion, 95-99, and cold gas efficiency, 68-72%. Also, equilibrium modeling was used to predict the vacuum residue gasification process and the predicted values reasonably well agreed with experimental data.

A novel system combining dynamic flash evaporation and vapor separation for desalination has been investigated in this work. The feed water passes through injection tubes which are connected to injector passages installed tangentially onto a separator tube. Flashing in the injection tubes is initiated from pressure drop due to friction and acceleration resulting in a two-phase mixture. Centrifugal force from tangential injection separates the two-phase mixture. In this compact system, vapor production and phase separation occur on the order of several milliseconds. Tap water and seawater were used as inlet feed water for the system. Thermal conversion efficiency to analyze the vapor production efficacy and phase separation efficiency to evaluate the purity of the condensate were investigated to evaluate the system. Thermal conversion and phase separation efficiencies up to 98% were obtained with the single stage system. Further improvement in purity of condensate was achieved with a two-stage system. In the two-stage system, the vapor captured along with some entrained droplets from the first stage was directed to a second set of injector passages connected in series to the first stage retrieval tube. With a two-stage system and seawater with salt concentration by mass of 2.5%, collected condensate with salt concentrations lower than 0.02% by mass was achieved, which is comparable to that of potable water. Thus, the novel dynamic flash evaporation and vapor separation system have been demonstrated to be very effective in producing potable water.