Modeling of pressure drop in reverse osmosis feed channels using multilayer artificial neural networks

Gu, Jianghang; Luo, Jiu; Li, Mingheng; Huang, Chunyan; Heng, Yi

doi:10.1016/j.cherd.2020.04.019

Cited by 10 publications

(3 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, intelligent approaches have been widely utilized for modeling thermal and hydrodynamic characterizations of various systems 53 , 54 . The multilayer perceptron (MLP) is the most famous approach among several available neural networks, which has been broadly used for modeling complex systems 55 – 58 . Figure 1 shows the configuration of the present MLP network for modeling of two-phase FPD in coiled tubes.…”

Section: Methodsmentioning

confidence: 99%

Robust and universal predictive models for frictional pressure drop during two-phase flow in smooth helically coiled tube heat exchangers

Moradkhani

Hosseini

Mansouri

et al. 2021

Sci Rep

View full text Add to dashboard Cite

There is a lack of well-verified models in the literature for the prediction of the frictional pressure drop (FPD) in the helically coiled tubes at different conditions/orientations. In this study, the robust and universal models for estimating two-phase FPD in smooth coiled tubes with different orientations were developed using several intelligent approaches. For this reason, a databank comprising 1267 experimental data samples was collected from 12 independent studies, which covers a broad range of fluids, tube diameters, coil diameters, coil axis inclinations, mass fluxes, saturation temperatures, and vapor qualities. The earlier models for straight and coiled tubes were examined using the collected database, which showed absolute average relative error (AARE) higher than 21%. The most relevant dimensionless groups were used as models’ inputs, and the neural network approach of multilayer perceptron and radial basis functions (RBF) were developed based on the homogenous equilibrium method. Although both intelligent models exhibited excellent accuracy, the RBF model predicted the best results with AARE 4.73% for the testing process. In addition, an explicit FPD model was developed by the genetic programming (GP), which showed the AARE of 14.97% for all data points. Capabilities of the proposed models under different conditions were described and, the sensitivity analyses were performed.

show abstract

Section: Methodsmentioning

confidence: 99%

Robust and universal predictive models for frictional pressure drop during two-phase flow in smooth helically coiled tube heat exchangers

Moradkhani

Hosseini

Mansouri

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In some works, the ANN were used together with other methods, in particular with the solution-diffusion model [158,163], the surface force-pore flow model [79], the CFD methods [164], and also with the surface response methodology [166]. Moreover, in work [79], it was noticed, that the ANN predictions were more accurate than the surface force-pore flow model ones.…”

Section: F17mentioning

confidence: 99%

Simulation of Reverse Osmosis Process: Novel Approaches and Development Trends

Huliienko

Korniyenko

Muzyka

et al. 2022

JES

View full text Add to dashboard Cite

Reverse osmosis is an essential technological separation process that has a large number of practical applications. The mathematical simulation is significant for designing and determining the most effective modes of membrane equipment operation and for a deep understanding of the processes in membrane units. This paper is an attempt at systematization and generalizing the results of the investigations dedicated to reverse osmosis simulation, which was published from 2011 to 2020. The main approaches to simulation were analyzed, and the scope of use of each of them was delineated. It was defined that computational fluid dynamics was the most used technique for reverse osmosis simulation; the intensive increase in using of molecular dynamics methods was pointed out. Since these two approaches provide the deepest insight into processes, it is likely that they will further be widely used for reverse osmosis simulations. At the same time, for the simulation of the membrane plant, it is reasonable to use the models that required the simplest solutions methods. The solution-diffusion model appears to be the most effective and flexible for these purposes. Therefore, this model was widely used in considering the period. The practical problems solved using each of the considered approaches were reviewed. Moreover, the software used for the solution of the mathematical models was regarded.

show abstract

“…Sivanantham et al modeled and optimized the rejection of chlorophenol in spiral wound RO modules using ANNs [38]. However, the data to train the ANN is either derived by model simulation [39,40], experimental single-stage RO membrane data points [41] or water sampling [42].…”

Section: Introductionmentioning

confidence: 99%

A Neural Network Based Superstructure Optimization Approach to Reverse Osmosis Desalination Plants

2022

View full text Add to dashboard Cite

An ever-growing population together with globally depleting water resources pose immense stresses for water supply systems. Desalination technologies can reduce these stresses by generating fresh water from saline water sources. Reverse osmosis (RO), as the industry leading desalination technology, typically involves a complex network of membrane modules that separate unwanted particles from water. The optimal design and operation of these complex RO systems can be computationally expensive. In this work, we present a modeling and optimization strategy for addressing the optimal operation of an industrial-scale RO plant. We employ a feed-forward artificial neural network (ANN) surrogate modeling representation with rectified linear units as activation functions to capture the membrane behavior accurately. Several ANN set-ups and surrogate models are presented and evaluated, based on collected data from the H2Oaks RO desalination plant in South-Central Texas. The developed ANN is then transformed into a mixed-integer linear programming formulation for the purpose of minimizing energy consumption while maximizing water utilization. Trade-offs between the two competing objectives are visualized in a Pareto front, where indirect savings can be uncovered by comparing energy consumption for an array of water recoveries and feed flows.

show abstract

Modeling of pressure drop in reverse osmosis feed channels using multilayer artificial neural networks

Cited by 10 publications

References 34 publications

Robust and universal predictive models for frictional pressure drop during two-phase flow in smooth helically coiled tube heat exchangers

Robust and universal predictive models for frictional pressure drop during two-phase flow in smooth helically coiled tube heat exchangers

Simulation of Reverse Osmosis Process: Novel Approaches and Development Trends

A Neural Network Based Superstructure Optimization Approach to Reverse Osmosis Desalination Plants

Contact Info

Product

Resources

About