Bed Profile Research Articles

Abstract PD14-06: Dynamic changes of PD-L1 and T-cell activation in ECLIPSE: A phase II study investigating preoperative immune combination strategies in untreated, operable ER+ primary breast cancer

Abstract Background: Whilst Atezolizumab + nab-paclitaxel is the new standard of care in patients with PD-L1+ metastatic triple negative breast cancer (BC), the role of immunotherapy in ER+ BC has yet to be defined. ECLIPSE, an open label, phase II trial, aims to evaluate the effect of short-term pre-operative immune-modulatory agent combinations on ER+ tumour immune-phenotypes (NCT03395899). Here, we examined the first results of the dynamic changes of PD-L1 expression (≥1% positive immune cell by VENTANA PD-L1 SP142 IHC) and T-cell expansion and activation after 1 cycle of atezolizumab alone or in combination with AKT, MEK or VEGF inhibitors. Methods: Patients with histologically confirmed, untreated, operable ER+/HER2- primary BC received 1 cycle of atezolizumab alone or atezolizumab + ipatasertib, atezolizumab + cobimetinib, or atezolizumab + cobimetinib + bevacizumab 3 weeks prior to surgery or neoadjuvant therapy. Biopsies were obtained pre and on completion of study treatment. Dynamic changes in biomarkers were compared in 48 evaluable patients, including 41 paired samples. Results: PD-L1 IC expression significantly increased after 1 cycle of treatment (p < 0.0001). At a 1% IC cutoff, 21/45 patients (47%) were PD-L1 IC-positive (+) at baseline while 34/44 (77%) were PD-L1 IC + post-treatment. Increased tumour cell expression was also found in post-treatment samples (p < 0.01). All post-treatment PD-L1 negative patients (n = 9) were also negative at baseline. Overall, there was a small but significant increase in CD8 + T-cells post-treatment compared to baseline (p = 0.03). In parallel, the number of T-cells that were positive for both CD8 and granzyme B (GZMB) - a T-cell activation marker - increased substantially (p < 0.01). Single GZMB+ cells and single FOXP3+ cells did not change with treatment. In 2 cases, no tumour was detected post-treatment, suggesting a major pathological response. Baseline biopsies of both patients stood out as having the largest amount of CD8+GZMB+ T-cells. Post-treatment tumour bed profiling of these patients showed major infiltration of CD8+ and CD8+GZMB+ T-cells. Dynamic gene expression analysis showed upregulation of genes related to innate immunity and a cytotoxic T-cell transcriptional signature (tGE8), (p < 0.001). Patients were grouped as immune-responsive or non-responsive based on a dynamic increase of CD8+GZMB+ cells. The immune-response group was characterized by increased IFN response signatures and decreased luminal gene signatures at baseline. Conclusions: Our data indicate that short-term induction treatment with atezolizumab ± targeted treatments induce dynamic changes toward inflamed immune phenotypes both at the morphological and transcriptional level, in patients with operable ER+/HER2- primary BC. These findings might help identify novel immunotherapy combination strategies in this setting, alongside the most appropriate biomarkers that can aid in identifying patients most likely to benefit from these treatments. Citation Format: Peter Schmid, Mark Kockx, Sung-Bae Kim, Duncan Wheatley, Melissa Mary Phillips, Oliver Hoffmann, Jens U. Blohmer, Seock-Ah Im, Andreas Schneeweiss, Esther Zamora, Yannick Waumans, Pieter-Jan Van Dam, Dieter Peeters, Akhila Wimalasingham, Patricia Marosics, Aaron Prendergast, Kelly Mousa, Javier Cortes, Sherko Kuemmel. Dynamic changes of PD-L1 and T-cell activation in ECLIPSE: A phase II study investigating preoperative immune combination strategies in untreated, operable ER+ primary breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD14-06.

Effect of Varying Bottom Topography on the Radiation of Water Waves by a Floating Rectangular Buoy

In the present study, the effect of an undulated bottom topography on the radiation of water waves by a floating rectangular buoy is analyzed. Various physical quantities of interest such as the added mass and damping coefficients associated with the surge, heave, and pitch motions are analyzed for a variety of parameters associated with the incident waves and bottom undulations. The study reveals that the added mass and damping coefficients associated with the surge and pitch motions of the floating buoy vary in an oscillatory manner with the variation in wavenumber for a sinusoidally varying bottom topography. Moreover, the oscillation amplitude is higher around the primary Bragg value. Further, this oscillatory pattern and oscillation amplitude increase with an increase in the ripple amplitude and the number of ripples for a sinusoidally varying bottom. However, a reverse pattern is formed with an increase in the depth ratio. In the long-wave regime, the added mass and damping coefficient corresponding to the surge motion become higher for a protrusion-type bed profile and lower for a depression-type bed profile. However, a reverse pattern is observed in the intermediate- and short-wave regimes.

Theoretical Investigation of Equilibrium Dynamics in Braided Gravel Beds for the Preservation of a Sustainable Fluvial Environment

Gravel bars have an important role in the exchange between surface and subsurface waters, in preventing and mitigating riverbank erosion, in allowing the recreational use of rivers, and in preserving fluvial or riparian habitats for species of fishes, invertebrates, plants, and birds. In many cases, gravel bars constitute an important substrate for the establishment and development of ground flora and woody vegetation and guarantee higher plant diversity. A sustainable management of braided rivers should, therefore, ensure their ecological potential and biodiversity by preserving a suitable braiding structure over time. In the present study, we propose an analytical–numerical model for predicting the evolution of gravel bars in conditions of dynamical equilibrium. The model is based on the combination of sediment balance equation and a regression formula relating dimensionless unit bedload rate and stream power. The results highlight the dependence of the evolving sediment particles’ pattern on the ratio of initial macro-bedforms longitudinal dimension to river width, which determines the gradual transition from advective and highly braiding to diffusive transport regime. Specifically, the tendency to maintain braiding and flow bifurcation is associated with equilibrium average bed profiles and, therefore, equilibrium average stream power characterized by the maximum period that does not exceed transverse channel dimension.

Migration of sand mining pit in rivers: An experimental, numerical and case study

Irregular, unauthorized and non-technical mining of sand and gravel from rivers play an important role in undesirable changes in morphology and environment. The present experimental study investigated sand mining pit modelling in the different scenarios. A two-dimensional flexible mesh, depth averaged model was set-up to simulate the experiments. The results exhibited an acceptable agreement between the numerical and experimental results, so that the maximum error was limited to less than 10%. After verifying the numerical model, a critical reach in Helleh River was selected as a case study. The effect of sand and gravel mining on a bridge at the site was studied. The results also indicated that for a flood event with a return period of 25 years, sand and gravel mining changed the bed profile up to 1.27 m at the bridge foundation. Moreover, initial signs of river meandering emerged in one scenario.

Flow Characteristics and Bed Morphology in a Compound Channel between Two Single Channels

In mountainous areas, a river can widen from a single channel to a compound channel under the influence of geological conditions or human impacts, bringing about challenges in terms of flood control and channel regulation. This paper reports the results of tests conducted in a 26 m long flume with a uniform sediment bed (grain size = 0.5 mm), investigating the flow characteristics and bed morphology in a compound channel between two single channels. The stage‒discharge relationship in the compound channel and the longitudinal and cross-sectional bed profile in the compound channel between two single channels are presented and analyzed. The experimental results indicate that the flow characteristics and bed morphology in a compound channel between two single channels are significantly different from those in a normal compound channel. Based on the experimental data and observations, the mechanisms of flow and sediment transport in the compound channel between two single channels are illuminated.

Wave-Induced Distribution of Microplastic in the Surf Zone

In this study, the wave-induced distribution of 13 microplastic (MP) samples of different size, shape, and density was investigated in a wave flume with a sandy mobile beach bed profile. The particle parameter were chosen based on an occurrence probability investigated from the field. MP abundances were analyzed in cross-shore and vertical direction of the test area after over 40,000 regular waves. It was found, that MP particles accumulated in more shallow waters with increasing size and density. Particles with high density (ρs>1.25 g/cm3) have been partly confined into deeper layers of the sloping beach during the formation of the bed profile. Particles with a density lower than that of water used in the experiments floated constantly in the surf zone or deposited on the beach caused by wave run-up. A correlation was found between the settling velocity of the MP particles and the flow velocity at the accumulation point and a power function equation developed. The obtained results were critically discussed with findings from the field and further laboratory studies.

PARAMETRY HYDRAULICZNE BYSTRZA O ZWIĘKSZONEJ SZORSTKOŚCI TYPU GRZEBIENIOWEGO: POTOK GRAJCAREK W SZCZAWNICY (KARPATY POLSKIE)

Aim of the study: The aim of this study is to analyze the hydrodynamic parameters in the area of one of the block ramp in the analyzed stream. Material and methods: The field research included geodetic measurements on the basis of which the longitudinal profile and cross-sections of the watercourse bed and structures were made. Numerical modelling of flood flows was also performed in the HEC-RAS program, from which hydrodynamic data were obtained. The data obtained from field measurements and numerical modelling were used to calculate the length of the energy dissipation basins of the structure and to determine the type of water movement using the Froude number. The calculations were performed for several variants. The real lengths of the energy dissipation basin were compared with the lengths of hydraulic jumps that are formed on them. Results and conclusions: The obtained results indicate that the length of the energy dissipation basin is too short because the hydraulic jumps go beyond them. As a result, a scour is formed below the structure, which in the event of further development may threaten the stability of the structure.

Bed compaction effect on dam break flow over erodible bed; experimental and numerical modeling

Disastrous floods result in severe loss of human lives and intense destruction to the infrastructure and economic activities. Furthermore, it leads to environmental and ecological damages located at the downstream area of the dam. The present study aims at investigating experimentally and numerically the sediment transport and morphological evolution of the erodible bed induced by dam-break flows. Experimental runs were conducted in a flume, specially designed for dam-break flow process, equipped with a thin vertical gate at its middle. Based upon an initial plane bed, the effect of three different compaction rates of the bed at both up-and downstream of the dam was investigated. The experimental data consistently suggested that increasing the bed compaction rate resulted in decreasing the scouring and sedimentation depth, as well as the sediment transport rate. Further, increasing the bed compaction rate led to increasing the wave-front celerity, reducing the scouring rate and the erosion depth. Both the experimental and numerical outputs highlighted the process of air entrainment at the leading edge of the wave-front. The numerical results showed that the maximum void fraction was associated with the maximum flow velocity at the zone of wave-front. Comparisons were made between the experimental results and those are provided from a numerical study using standard volume of fluid VOF method, where three different turbulence closure schemes; RNG, k-ε and k-ω, and three bed load sediment transport approaches, Van Rijn, Meyer-Peter-Muller and Nielson, were applied. Accordingly, by considering the error values in predicting the free surface height and the bed deformation, the k-ω closure model showed the highest accuracy in capturing the turbulence features while the Mayer Peter-Muller formula had the highest performance in predicting the bed deformation in non-compacted (NC) bed. Besides, the RNG closure model and Nielsen bed load formula revealed the highest accuracy in predicting the aforementioned features in a semi-compacted (SC) bed. The k-ɛ closure model and Mayer Peter-Muller formula showed the highest accuracy in reproducing the turbulence characteristics and bed profile in a fully-compacted (FC) bed.

Ballasted track aerodynamic optimization by wind tunnel test and CFD simulation

Ballast flight problem sets back high-speed railway development, which causing damages to train and rail. The present paper puts forward wind tunnel tests and CFD simulation to study ballast flight mechanisms and ballasted track aerodynamics. A 1:1 scale ballasted track model was set up to analyze ballast flight phenomena, where aerodynamics influencing factors and impacts on ballast bed were investigated. A numerical model of CRH3 train-ballast bed was built by combining the Realizable k- ϵ viscous model and sliding mesh technique. CFD model with the above method was used to investigate aerodynamic effects as well as the mechanism of ballast flight, thus providing the theoretical basis for optimizing ballast bed profile and preventing ballast flight. Results show that pressure distribution exhibits large value at the center of the ballasted track and decreases at track edge. The head, tail and bogie area of the train result in large fluctuation of wind pressure, which has a great probability of ballast flight. Appropriate countermeasures should be taken, such as reduce shoulder ballast height and crib ballast could be effective methods to decrease the probability of ballast flight.

Experimental investigation of conduction heat transfer in a rotary drum using infrared thermography

A technique for studying heat transfer in a rotary drum is proposed using infrared (IR) thermography, which is used to record images to give instantaneous thermal data. In this work, the calibration procedure and methodology used to extract temperatures of the particles and the drum wall from the IR camera is discussed. The technique provides insightful information on the flow regimes and heat transfer for varying drum fill levels and rotation rates. The flow profile of the particle bed for each run is analyzed both theoretically using Froude number and experimentally using recorded IR images. The results suggest that the flow of particles is influenced by particle-particle friction in the presence of heat transfer. A transition in the flow profile is observed as time progresses and the bed reaches a surging regime. Higher rotation rates result in maximum average bed temperatures, whereas fill level does not have a conclusive effect.

Deep learning technique for fast inference of large-scale riverine bathymetry

Riverine bathymetry is of crucial importance for shipping operations and flood management. However, obtaining direct measurements of depth is not always easy. Conversely, with recent advances in sensor technology, indirect measurements can be obtained and used to estimate high-resolution river bed topography. Physics-based inverse modeling techniques have been used to estimate bathymetry using indirect measurements like flow velocity at the surface. However, these methods are computationally expensive for large-scale problems. Recently, deep learning has opened a new door toward knowledge representation and complex pattern identification in many fields; however, these techniques have not been used for high-dimensional riverine bathymetry problems since they require a large amount of data in the training phase to have a good estimation performance that can be generalized for new river profiles. Also, unless one reduces the dimension of the problem, these methods can have a computationally expensive similar to that of physics-based techniques. Here, we develop a new deep learning framework for riverine problems that can be trained using only a few river profiles and in a computationally efficient way that allows finding solutions on personal computers. The proposed method exploits the spatially local connection between the observations and river bed profile and combines a fully connected Deep Neural Network (DNN) with Principal Component Analysis (PCA) to image river bed topography using depth-averaged flow velocity observations. The new method is presented and applied to three riverine bathymetry identification problems. Results show that the proposed method achieves satisfactory performance in bathymetry estimation, providing a powerful data-driven technique for riverine bathymetry in terms of prediction quality, robustness, and computational cost that requires only a relatively small number of training samples.

An experimental study of mitigating coastal sand dune erosion by microbial- and enzymatic-induced carbonate precipitation

Due to more extreme weather events and accelerating sea-level rise, coastal sand dunes are subjected to more frequent storm wave inundation and surge impacts, which contribute to widespread coastal erosion problems. In this study, two novel bio-mediated methods, microbial-induced carbonate precipitation (MICP) and enzymatic-induced carbonate precipitation (EICP), were investigated and compared for their effectiveness in mitigating sand dune erosion under wave attack. Small-scale laboratory model tests were performed on MICP-treated, EICP-treated, and untreated sand dunes at dune slope angles and two wave intensities for up to 2 h. The cross-shore profile was captured continuously during the course of the erosion test. The erosion volume above the still water level (SWL) and landward retreat distance at the SWL were calculated based on the captured bed profiles. The results show that both EICP and MICP could substantially reduce sand dune erosion at mild-to-moderate wave and dune slope conditions. However, the effectiveness of MICP treatment deteriorated at steeper dune slopes with longer period of wave attack. Under the most adverse condition (i.e., steepest dune slope, biggest wave, and longest period of wave attack), neither EICP nor MICP could effectively mitigate erosion. Fundamentally, the variable effectiveness of MICP and EICP treatment for sand dune erosion control was attributed to the spatial distribution pattern of formed calcite precipitation, which was determined by the way how EICP and MICP were applied. The calcite precipitation was relatively uniform in EICP-treated sand dunes. In MICP-treated ones, however, substantial calcite precipitation concentrated in the shallow surface layer as confirmed by the surface penetration test and SEM observation.

Effect of undulating bottom on wave interaction with a floating flexible plate coupled with a flexible porous barrier

In the present study, under the assumption of small amplitude water wave theory and structural response, effect of bed undulation on the wave interaction with a combination of flexible porous barrier and a flexible floating plate, is studied. The flexible porous barrier is modelled using the porous wave maker-theory while the elastic floating plate is modelled using thin plate theory. The physical problem is handled for solution using eigenfunction expansion method by matching pressure and velocity at interface boundaries while finite difference method is used to deal with modified mild-slope equation. In the present study, two types of plate configurations are namely (a) finite plate and (b) semi-infinite plate. To understand the role of undulating seabed, wave and structural parameters, in attenuating wave force and plate deflection, numerical results are computed and compared with available literature. It is found that full wave reflection occurs for certain critical angle and wave force acting on the barrier vanishes for the same critical angle for the semi-infinite plate case while the wave reflection tends to attain maximum value at critical angle for the finite case and these maximum value increases with the increase of plate length. It is observed that less wave reflection occurs for sloping bed profile compare to rest of the bed profile. The study reveals that positioning a barrier between plate and undulated region, helps in the reduction of plate deflection significantly. As special case, wave interaction problem with a finite floating elastic plate is studied experimentally and the experimental result has shown good agreement with the numerical result. The findings of the present study are likely to be of immense help in the design of various types of marine structures for protecting very large floating structures (VLFS). The present theory can be extended to handle large class of acoustic wave interaction problems with flexible porous structures.

Numerical modeling of local scour at a submerged weir with a downstream slope using a coupled moving-mesh and masked-element approach

This paper presents a model for local scour at submerged weirs with downstream slopes that uses a coupled moving-mesh and masked-element approach. In the developed model, the fluid-sediment interface is tracked using a moving-mesh technique, and the effects of the structure on the hydrodynamics and bed morphology are resolved using a masked-element technique. Compared to traditional sediment scour models, based on the moving-mesh technique, the present model has the advantage of allowing for a simpler setup of the computational grids and a larger-amplitude deformation. Laboratory experiments on local scour at a submerged weir with a downstream slope were conducted, which provided bed profiles at different time instants. The results obtained by the present model are compared to the experimental data. The comparisons demonstrate the performance of the model in satisfactorily predicting local scour at a submerged weir with a downstream slope. The model was further modified and employed to carry out additional computations to investigate the influence of various parameters and sub-models.

Numerical Investigation of the Effect of Suspended Load on the Morphology of a Beach under the Action of Tsunamis

Fan, F.; Zhu, Z.; Lu, Y.; Yang, B., and Zhao, C., 2020. Numerical investigation of the effect of suspended load on the morphology of a beach under the action of tsunamis. In: Zheng, C.W.; Wang, Q.; Zhan, C., and Yang, S.B. (eds.), Air-Sea Interaction and Coastal Environments of the Maritime and Polar Silk Roads. Journal of Coastal Research, Special Issue No. 99, pp. 174–180. Coconut Creek (Florida), ISSN 0749-0208.To study the morphology of a beach under the action of a tsunami, a beach evolution numerical model based on the open-source calculated fluid dynamics (CFD) code OpenFOAM is constructed. The bed load and suspended load models are fully coupled with the hydrodynamics model in the beach evolution model, and the moving mesh method is used to capture the change in bed morphology. Then, a laboratory experiment on the scour induced by solitary waves is adopted to validate the constructed numerical model, and the surface elevation, bed shear stress and bed profile of the model fit very well with the experimental observations. Finally, the effect of the suspended load on the evolution of beach morphology and the process of beach evolution under the action of a solitary wave are studied with a numerical model. The study results show that the beach evolution mainly occurred at the time the solitary wave ran down and that the suspended load played a major role in the beach evolution, contributing approximately 85% to the scour profile.

The effect of air bubbles on optical backscatter sensor measurements under plunging breaking waves

Accurate suspended sediment concentration measurements are key to understand and quantify the sediment transport patterns in the surf and swash zones. One of the most widely used instruments to collect suspended sediment concentrations is the Optical Backscatter Sensor (OBS). However, the OBS is known to give erroneous readings when deployed in bubbly environments like the surf zone. The present study aims to quantify the influence of an aerated wave breaking environment on the OBS sediment concentration measurements. Experiments are performed in a large wave flume, which ensures full air entrapment under plunging breaking waves, and avoids scale effects that could affect the volume of entrapped air, the air bubble penetration depth and the residence time of air bubbles in the post-breaking turbulent eddies. OBS measurements are obtained at 66 locations along a fixed bed profile for 14 regular breaking wave conditions. In the absence of suspended sediment particles, OBS voltage measurements are used as a proxy for air bubble content. The presented OBS results show peaks up to 1.49 V (31% of the OBS measurement range, corresponding to 16 g/l for sediment with d50 = 0.25 mm) produced by air bubbles in the most energetic tested wave breaking conditions, while the maximum time-averaged value obtained is 0.48 V (10% of the OBS measurement range, corresponding to 5 g/l). The results highlight the importance of considering the presence of air bubbles where OBS are deployed to measure suspended sediment concentrations. A good correlation is found between the breaker depth index and the air bubble distribution and two predictive formulas are derived to forecast the area of air bubble influence in the surf zone.

Long-term effects of dam operations for water supply to irrigation on downstream river reaches. The case of the Ribb River, Ethiopia

ABSTRACT This work investigates the applicability of an analytical method for quick assessments of the long-term morphological effects of different dam operations on downstream river reaches with the idea to apply the method in feasibility studies to identify the least morphologic-impacting operation scenario. The Ribb River (Ethiopia) is used as a study case. The analytical method estimates the idealized, new equilibrium of the river bed profile without considering the duration of the morphological evolution. We apply the analytical method distinguishing sand-bed from gravel-bed reaches. The outcome of the analytical method is compared to that of a calibrated one-dimensional river morphology computer model. The analytical method overestimated the morphological changes compared to the one-dimensional model. By establishing the upper limits of the impact, the analytical method identifies a theoretical maximum river bed degradation near the base of the dam. If all sediment is trapped in the reservoir, the method allows distinguishing the effects of different dam operation scenarios, but only for gravel-bed river reaches. However, the method can also be applicable for sand-bed reaches if there is sediment input from the upper reaches. Further research works should be done to validate both methods if they indeed allow to detect the least impacting scenario, considering that data showing the effects of long-term dam operations on the downstream river reaches are lacking.

Experimental modeling of wave load on a pile-supported wharf with pile breakwater

A hybrid structure of pile-supported wharf connecting with pile breakwater is tested in a flume with a scale factor of 1:30. Based on generating regular and irregular waves and accurately reproducing bed profile by using pulverized coal, both wave, scour and load characteristic of this structure are investigated. When structure under wave impact, violent wave reflection is observed and wave transmission is small. With the large wave height and water depth, wave overtopping is significant and induces bigger green water height. Compared with bed without covering layer, rubble covering layer obviously decreases the scour range of bed profile. Pressures of the structure have distinct spatial distribution characteristics. The existing plates between piles of breakwater can easily trap the air during wave attack and induce significant impact pressure. Besides, superstructure of pile-supported wharf is mainly suffered from uplift pressure and attains the maximum pressure when air gap becomes zero. Due to dynamic response of structure, measured forces have smaller values and larger periods than integrated forces, and both horizontal and vertical measured forces decrease with increasing relative wave height. The findings and data presented here may be used by oceanic numerical modelers and engineers, for future research as well as predesign.

High-temperature CO2 methanation over structured Ni/GDC catalysts: Performance and scale-up for Power-to-Gas application

This paper investigates the performance of 25 wt% Ni/GDC catalyst supported on ceramic monolith (MO) and open-cell foam (FO) towards the conversion of CO2 to CH4 at bench-scale level (CO2 flow rate = 78–319 NL·h−1). Both structured catalysts were prepared by the In Situ-Solution Combustion Deposition (IS-SCD) method to coat thin, uniform and high-resistance catalytic layers (Ni/GDC). Catalysts were characterized by XRD, TEM, SEM, and adhesion tests. Temperature profiles in axial and radial direction were registered and discussed. The experimental results highlighted the effect of the geometry of the support on the thermal profile of the catalytic bed and, consequently, on the catalytic performance towards CO2 methanation. For the investigated quasi-adiabatic conditions, the monolith-supported sample enabled a favourable temperature distribution, leading to higher performances (χCO2 = 70.3%; FCH4, OUT = 2.6 NL·cm−3·h−1) compared to the foam-based catalyst (χCO2 = 52.0%, FCH4, OUT = 2.2 NL·cm−3·h−1). Besides, Ni/GDC-MO catalyst showed high stability for 50 h of time-on-stream under daily start-up and shut-down cycles.

Experimentally validated machine learning frameworks for accelerated prediction of cyclic steady state and optimization of pressure swing adsorption processes

Machine learning-based surrogate models are presented to accelerate the optimization of pressure swing adsorption processes. Various supervised machine learning algorithms, such as decision trees, random forests, support vector machines, Gaussian process regression, and artificial neural networks, are tested for their ability to predict key performance indicators for a given set of operating conditions. Among the algorithms studied, Gaussian process regression-based surrogate models were found to be the best at predicting process outputs, with minimal training effort. The adjusted coefficient of determination for predictions using the surrogate model is greater than 0.98 using a sampled training set of 400 operating conditions. A surrogate model based on artificial neural networks is also presented to predict the bed profiles of the intensive variables at cyclic steady state. The surrogate models show very good agreement with the detailed model simulations. Experiments performed on a lab-scale two-column rig, for the concentration of CO2 from a mixture of CO2+N2 on Zeolite-13X, confirm performance indicators such as purity, recovery and axial profiles predicted by the surrogate models. Two new optimization frameworks are presented: Surrogate optimization in which the trained surrogate model is used to provide the process performance; and cyclic steady state optimization in which the predicted cyclic steady state profiles are provided as an initial condition for detailed model in order to accelerate convergence. Both techniques are shown to accurately predict Pareto fronts of Purity-Recovery and Energy-Productivity calculated from optimization that uses a detailed process model. The Surrogate optimization and the cyclic steady state accelerated Detailed optimization show ≈23× and 6× reduction in computational load, respectively, when compared to the traditional optimization using detailed models.