Parameterization Strategy Research Articles

Abstract 17071: Cell Communication Inference in Macrophages

Introduction: Tissue macrophages play diverse roles in the cardiovascular system during health and disease. They have diverse functions within tissues, but our understanding of their dynamics is limited because most macrophage characterization assays are destructive and have low temporal resolution. We asked whether these cells are dynamic and interconnected. Methods: Here, we describe experimental and analytical methods for measuring cell dynamics and inferring communication between cells in vitro and in vivo. We created a mouse (Csf1r-Cre x GCaMP5) expressing the Cre-inducible genetically encoded calcium indicator GCaMP5 under the regulation of the innate immune promoter, Csf1r, to non-destructively quantify high-frequency cell dynamics and differentiated them in culture using m-CSF. We developed custom image analysis routines and parameterization strategies for classifying calcium responses. Results: Our studies revealed that calcium reporter BMDMs display minimal fluctuations at baseline but exhibit a dynamic response to immunogenic DNA sensing. DNA-induced isolated cell injury and death, which precipitated cell communication that spread with a velocity of [9μm/s], consistent with an extracellular diffusion mechanism. We developed quantitative image analysis methods that corrected for random calcium fluctuations and identified statistically significant areas of correlated calcium changes suggestive of communication. An analytical pipeline enabled quantification of calcium spike dynamics and correlations of dynamic calcium profiles of single cell sharing a local microenvironment. This resulted in an “improbable synchrony” metric that allowed localization of communication in time and space. We adapted the pipeline for in vivo studies and tested them in a dorsal window chamber model using intravital microscopy. At 2Hz sampling frequency, we identified 27 potential communication events as they responded to complex microenvironmental cues in vivo. Conclusion: The experimental and analytical methods for inferring cell communication provide a new quantitative toolkit for investigating known as-yet undiscovered cell communication pathways.

Modeling Human Steering Behavior in Teleoperation of Unmanned Ground Vehicles With Varying Speed

This paper extends a prior human operator model to capture human steering performance in the teleoperation of unmanned ground vehicles (UGVs) in path-following scenarios with varying speed. A prior study presented a human operator model to predict human steering performance in the teleoperation of a passenger-sized UGV at constant speeds. To enable applications to varying speed scenarios, the model needs to be extended to incorporate speed control and be able to predict human performance under the effect of accelerations/decelerations and various time delays induced by the teleoperation setting. A strategy is also needed to parameterize the model without human subject data for a truly predictive capability. This paper adopts the ACT-R cognitive architecture and two-point steering model used in the previous work, and extends the model by incorporating a far-point speed control model to allow for varying speed. A parameterization strategy is proposed to find a robust set of parameters for each time delay to maximize steering performance. Human subject experiments are conducted to validate the model. Results show that the parameterized model can predict both the trend of average lane keeping error and its lowest value for human subjects under different time delays. The proposed model successfully extends the prior computational model to predict human steering behavior in a teleoperated UGV with varying speed. This computational model can be used to substitute for human operators in the process of development and testing of teleoperated UGV technologies and allows fully simulation-based development and studies.

A parameterization strategy for hydrodynamic modelling of a cascade of poorly monitored reservoirs in Brazil

The scarcity of long time-series of water temperature for tropical reservoirs, usually available at a low frequency of measurements, represents a serious constraint for the understanding of their thermal behavior. Hydrodynamic models become a powerful tool for that understanding through a qualitative approach. In this study, the one-dimensional General Lake Model (GLM) was applied to hindcast multiannual temperature profiles of a cascade system of six reservoirs located in southeast Brazil over a period of eight years. The model parameters were calibrated and validated by field data for four reservoirs. To overcome the challenge of exploring hydrodynamics in two reservoirs with a lack of measured temperatures, a parameterization strategy was applied to accomplish a better simulation by the estimation of potential sensitive parameters through regression curves. It is expected that this approach can be extended to other tropical reservoirs whenever field data is not available for the calibration procedure.

Automated parameterization of quantum-mechanically derived force-fields including explicit sigma holes: A pathway to energetic and structural features of halogen bonds in gas and condensed phase.

In classical molecular dynamics, general purpose atomistic force-fields (FFs) often deliver inaccurate results when dealing with halogen bonds (XBs), notwithstanding their crucial role in many fields of science, ranging from material design to drug development. Given the large dimensions of the systems of interest, it would be therefore desirable to increase the FF accuracy maintaining the simplicity of the standard Lennard-Jones (LJ) plus point charge description to avoid an excessive computational cost. A simple yet effective strategy consists in introducing a number of virtual sites able to mimic the so-called "explicit σ-hole." In this work, we present an automated FF parameterization strategy based on a global optimization of both LJ and charge parameters with respect to accurate quantum mechanical data, purposely computed for the system under investigation. As a test case, we report on two homologue series, characterized either by weak or strong XBs, namely, the di-halogenated methanes and the mono-, di-, and tri-substituted acetonitriles, taking into consideration Cl, Br, and I substituents. The resulting quantum mechanically derived FFs are validated for each compound in the gas and in the condensed phase by comparing them to general purpose and specific FFs without virtual sites and to highly accurate reference quantum mechanical data. The results strongly support the adoption of the specific FFs with virtual sites, which overcome the other investigated models in representing both gas phase energetics and the structural patterns of the liquid phase structure related to the presence of XBs.

An Integrated Modelling Approach for Flood Simulation in the Urbanized Qinhuai River Basin, China

The accurate simulation and prediction of flood response in urbanized basins remains a great challenge due to the spatial and temporal heterogeneities in land surface properties. We hereby propose an integrated modelling approach that consists of a semi-distributed conceptual hydrological model and a novel parameterization strategy. The modelling approach integrates the Xinanjiang (XAJ) model, Taihu Basin (TB) model, and Nash instantaneous unit hydrograph (IUH) into a framework. Model parameters are calibrated by optimizing their relationships with corresponding physical factors. The proposed modelling approach is applied in the Qinhuai River basin (QRB), China. The modelling approach shows satisfactory performance in flood simulation both for calibration and validation of flood events in the QRB. The approach has temporal and spatial prediction capability by using the established relationships between parameter values and physical factors. Robustness analysis reveals that the different sets of flood events used for parameter relationship calibration led to similar model performance. Numerical experiments show that impervious coverage poses strong influences on the model performance and needs to be considered in flood routing simulations for small- or medium-intensity flood events.

Parameterization of the water stress reduction function based on soil–plant water relations

Rational parameterization of the soil water stress reduction function in root water uptake model is crucial for accurate description of root water uptake and simulation of soil water dynamics in a soil–plant system. In this study, we propose three improvements to a popular transpiration-based approach to parameterize the water stress reduction function in a widely used macroscopic root water uptake model. The improvements are based on the interdependent relationships between soil and plant water status and consideration of effects of (1) relative distribution of soil water to roots on transpiration; (2) differences in growth levels of plants exposed to different levels of water stresses on potential transpiration; and (3) hysteresis of water stress on parameter optimization through identifying and discarding the data involved in the recovery periods when the discrepancy between soil and plant water availability is significant. Lysimetric experiments with winter wheat planted alternatively in greenhouse soil columns and in a field were conducted to test the proposed improvements. Through minimizing the residuals between the measured and estimated actual transpiration, the optimized parameterization was used to set up the root water uptake model. Thereupon, actual transpiration and relative transpiration were estimated and soil water content distributions were simulated. The estimated actual (RMSE ≤ 0.09 cm day−1) and relative (RMSE = 0.06) transpiration agreed well with the measurements. The simulated soil water content distributions also matched the measured values well for both experiments (RMSE ≤ 0.023 cm3 cm−3). Omitting any of the three proposed improvements reduced the estimation accuracy of relative transpiration, as the individual contribution ratio for each improvement was between 21.2 and 51.2%. The improvements should be reasonable in providing rational parameter estimation for the water stress reduction function, from which root water uptake models can be established to accurately evaluate plant transpiration and simulate soil water flow in a soil–plant system. The parameterization strategy for the water stress reduction function of root water uptake not only benefits accurate evaluation of plant transpiration under drought conditions but also contributes to further study and description regarding the apparent hysteresis of root water uptake after re-watering.

Parametric representation of arrays of wave energy converters for motion simulation and unknown input estimation: A moment-based approach

When it comes to parameterisation of dynamical models for arrays of Wave Energy Converters (WECs), the most-used approach, within the wave energy literature, provides a state-space representation whose order (dimension) increases quadratically with the number of devices composing the WEC array. This represents a major drawback for key WEC design elements, such as motion simulation and unknown input estimation, being the latter essential to effectively maximise energy extraction from ocean waves. We present herein a multi-input, multi-output (MIMO) parameterisation strategy based on a system-theoretic interpretation of moments. The state-space representations computed with this moment-based approach exactly match the steady-state behavior of the target WEC system at specific (user-selected) interpolation points, providing efficient low dimensional models that can accurately represent the input-output dynamics of WEC arrays. Moreover, we show that there exists an intrinsic connection between wave excitation force estimation strategies and the moment-based parameterisation method proposed in this paper. We exploit this mathematical correlation to provide low order models that deliver the same degree of wave excitation force estimation accuracy to that obtained by implementing the currently-used parameterisation methods, with mild computational requirements. The performance of the strategy is analysed in terms of a case study, considering a WEC array composed of state-of-the-art CorPower-like devices, for both WEC motion simulation and wave excitation force estimation scenarios.

Multiobjective Optimal Power Flow Using a Semidefinite Programming-Based Model

In spite of the significant advance achieved in the development of optimal power flow (OPF) programs, most of the solution methods reported in the literature have considerable difficulties in dealing with different-nature objective functions simultaneously. By leveraging recent progress on the semidefinite programming (SDP) relaxations of OPF, in the present article, attention is focused on modeling a new SDP-based multiobjective OPF (MO-OPF) problem. The proposed OPF model incorporates the classical ε-constraint approach through a parameterization strategy to handle the multiple objective functions and produce Pareto front. This article emphasizes the extension of the SDP-based model for MO-OPF problems to generate globally nondominated Pareto optimal solutions with uniform distribution. Numerical results on IEEE 30-, 57-, 118-bus, and Indian utility 62-bus test systems with all security and operating constraints show that the proposed convex model can produce the nondominated solutions with no duality gap in polynomial time, generate efficient Pareto set, and outperform the well-known heuristic methods generally used for the solution of MO-OPF. For instance, in comparison with the obtained results of NSGA-II for the 57-bus test system, the best compromise solution obtained by SDP has 1.55% and 7.42% less fuel cost and transmission losses, respectively.

Synthesizing Talking Faces from Text and Audio: An Autoencoder and Sequence-to-Sequence Convolutional Neural Network

Synthesizing talking face from text and audio is increasingly becoming a direction in human-machine and face-to-face interactions. Although progress has been made, several existing methods either have unsatisfactory co-articulation modeling effects or ignore relations between adjacent inputs. Moreover, some of these methods often train models on shaky head videos or utilize linear-based face parameterization strategies, which further decrease synthesized quality. To address the above issues, this study proposes a sequence-to-sequence convolutional neural network to automatically synthesize talking face video with accurate lip sync. First, an advanced landmark location pipeline is used to accurately locate the facial landmarks, which can effectively reduce landmark shake. Then, a part-based autoencoder is presented to encode face images into a low-dimensional space and obtain compact representations. A sequence-to-sequence network is also presented to encode the relation of neighboring frames with multiple loss functions, and talking faces are synthesized through a reconstruction strategy with a decoder. Experiments on two public audio-visual datasets and a new dataset called CCTV news demonstrate the effectiveness of the proposed method against other state-of-the-art methods.

Different uncertainty distribution between high and low latitudes in modelling warming impacts on wheat

Global gridded climate–crop model ensembles are increasingly used to make projections of how climate change will affect future crop yield. However, the level of certainty that can be attributed to such simulations is unknown. Here, using currently available geospatial datasets and a widely employed simulation procedure, we created a wheat model ensemble of 1,440 global simulations of 20 climate scenarios, 3 crop models, 4 parameterization strategies and 3 management inputs of sowing date. We quantified the contributions of climate, model, parameterization and management to the overall uncertainty to predicted responses of yield to warming, then related the results to the latitude of the grid cells. For all warming scenarios, the total uncertainty for mid- and high latitudes is much larger than for low latitudes. Uncertainty arising from crop models was larger than that from the other sources combined. Parameterizing crop models with grid-specific information on wheat cultivars tended to decrease the crop model uncertainty, particularly for low latitudes. Crop model improvements and better-quality spatial input data more closely representing the wide range of growing conditions around the world will be needed to reduce the uncertainty of climate change impact assessment of crop yields. Global gridded crop models simulate the impact of climate change on crop yield, but uncertainty is difficult to quantify, which reduces their reliability. Xiong et al. explored how parameterization strategies can reduce uncertainty and identified greater uncertainty in future wheat yields grown at mid-to-high latitudes.

Inversion Technique for Quantitative Infrared Thermography Evaluation of Delamination Defects in Multilayered Structures

Inverse analysis is a promising tool for quantitative evaluation offering informative model-based prediction and providing accurate reconstruction results without preinspections for characterization criteria. For the traditional defect inverse reconstruction, a large number of parameters are required to reconstruct a complex defect, and the corresponding forward modeling simulation is very time-consuming. Such issues result in ill-posed and complex inverse reconstruction results, which further reduce its practical applicability. In this article, we propose and experimentally validate an inversion technique for the reconstruction of complexly shaped delamination defects in a multilayered metallic structure using signals derived from infrared thermography (IRT) testing. First, we employ a novel defect parameterization strategy based on Fourier series fitting to represent the profile of a complicated delamination defect with relatively few coefficients. Second, the multimedium element modeling method is applied to enhance a finite element method (FEM) fast forward simulator, in order to solve the mismatching mesh issue for mesh updating during inversion. Third, a deterministic inverse algorithm based on a penalty conjugate gradient algorithm is employed to realize a robust and efficient inverse analysis. By reconstructing delamination profiles with both numerically simulated IRT signals and those obtained through laser IRT experiments, the validity, efficiency, and robustness of the proposed inversion method are demonstrated for delamination defects in a double-layered plate. Based on this strategy, not only is the feasibility of the proposed method in IRT nondestructive testing is validated, but the practical applicability of inversion reconstruction analysis is significantly improved.

Computational design optimization of a smart material shape changing building skin tile

The development and evaluation of a computational approach for optimal design of a smart material shape changing building skin is presented and numerically evaluated. Specifically, a unique shape-based approach is utilized to create an optimization approach to identify the activation and actuation mechanisms to minimize the difference between a desired shape and the estimated morphed shape. Three potential metrics of shape difference are considered and their capability to facilitate an efficient optimization process leading to accurate shape matching is evaluated. Details of the optimal design framework are presented, particularly focusing on the shape difference metrics as well as the strategy to parameterize the activation of the smart material. In particular, the parameterization strategy is a unique approach to easily integrate controllable localized activation within a smart material structure in a generally applicable way that does not limit the design search space. A series of numerical design examples are presented based on the concept of a smart material (e.g., shape memory polymer) shape changing tile that can be activated and actuated in a variety of ways to achieve desirable surface wrinkle patterns. These numerical design examples are applied to both 2D and 3D problems and consider a variety of parameterizations and target shapes. Results indicate that the shape-based approach can consistently determine the mechanisms of morphing needed to accurately match a target shape. Furthermore, it is shown that localized material activation can lead to not only a more accurate shape but also requires less energy and actuation devices to do so.

Ancestral reconstruction of protein interaction networks.

The molecular and cellular basis of novelty is an active area of research in evolutionary biology. Until very recently, the vast majority of cellular phenomena were so difficult to sample that cross-species studies of biochemistry were rare and comparative analysis at the level of biochemical systems was almost impossible. Recent advances in systems biology are changing what is possible, however, and comparative phylogenetic methods that can handle this new data are wanted. Here, we introduce the term “phylogenetic latent variable models” (PLVMs, pronounced “plums”) for a class of models that has recently been used to infer the evolution of cellular states from systems-level molecular data, and develop a new parameterization and fitting strategy that is useful for comparative inference of biochemical networks. We deploy this new framework to infer the ancestral states and evolutionary dynamics of protein-interaction networks by analyzing >16,000 predominantly metazoan co-fractionation and affinity-purification mass spectrometry experiments. Based on these data, we estimate ancestral interactions across unikonts, broadly recovering protein complexes involved in translation, transcription, proteostasis, transport, and membrane trafficking. Using these results, we predict an ancient core of the Commander complex made up of CCDC22, CCDC93, C16orf62, and DSCR3, with more recent additions of COMMD-containing proteins in tetrapods. We also use simulations to develop model fitting strategies and discuss future model developments.

Adversarial-residual-coarse-graining: Applying machine learning theory to systematic molecular coarse-graining.

We utilize connections between molecular coarse-graining (CG) approaches and implicit generative models in machine learning to describe a new framework for systematic molecular CG. Focus is placed on the formalism encompassing generative adversarial networks. The resulting method enables a variety of model parameterization strategies, some of which show similarity to previous CG methods. We demonstrate that the resulting framework can rigorously parameterize CG models containing CG sites with no prescribed connection to the reference atomistic system (termed virtual sites); however, this advantage is offset by the lack of a closed-form expression for the CG Hamiltonian at the resolution obtained after integration over the virtual CG sites. Computational examples are provided for cases in which these methods ideally return identical parameters as relative entropy minimization CG but where traditional relative entropy minimization CG optimization equations are not applicable.

Weibull grain-based model (W-GBM) for simulating heterogeneous mechanical characteristics of salt rock

This research presents a Weibull statistical grain-based model in two-dimensional Particle Flow Code (PFC2D) to describe the microstructural heterogeneity induced by grain shapes based on centroidal Voronoi tessellation technology. A consistent parameterization strategy based on macroscopically measured data of Qianjiang salt rock was implemented to determine the microparameters of the numerical model. Deformation response, strength characteristics as well as microcracking behavior are examined by varying heterogeneity of numerical specimens under uniaxial compression and compressive tests with various confining pressures. The simulated results indicate that the proposed model is capable of predicting the mechanical behavior of salt rock in the laboratory and numerical conclusions.

Turbine Passage Design Methodology to Minimize Entropy Production-A Two-Step Optimization Strategy.

Rapid aerodynamic design and optimization is essential for the development of future turbomachinery. The objective of this work is to demonstrate a methodology from 1D mean-line-design to a full 3D aerodynamic optimization of the turbine stage using a parameterization strategy that requires few parameters. The methodology is tested by designing a highly loaded and efficient turbine for the Purdue Experimental Turbine Aerothermal Laboratory. This manuscript describes the entire design process including the 2D/3D parameterization strategy in detail. The objective of the design is to maximize the entropy definition of efficiency while simultaneously maximizing the stage loading. Optimal design trends are highlighted for both the stator and rotor for several turbine characteristics in terms of pitch-to-chord ratio as well as the blades metal and stagger angles. Additionally, a correction term is proposed for the Horlock efficiency equation to maximize the accuracy based on the measured blade kinetic losses. Finally, the design and performance of optimal profiles along the Pareto front are summarized, featuring the highest aerodynamic performance and stage loading.

Quantifying peakflow attenuation/amplification in a karst river using the diffusive wave model with lateral flow

AbstractThe aim of this paper is to quantify peakflow attenuation and/or amplification in a river, investigating lateral flow from the intermediate catchment during floods. This is a challenge for the study of the hydrological response of permeable/intermittent streams, and our contribution refers to a modelling framework based on the inverse problem for the diffusive wave model applied in a karst catchment. Knowing the upstream and downstream hydrographs on a reach between two stations, we can model the lateral one, given information on the hydrological processes involved in the intermediate catchment. The model is applied to 33 flood events in the karst reach of the Iton River in French Normandy where peakflow attenuation is observed. The monitored zone consists of a succession of losing and gaining reaches controlled by strong surface‐water/groundwater (SW/GW) interactions. Our results show that despite a high baseflow increase in the reach, peakflow is attenuated. Model application shows that the intensity of lateral outflow for the flood component is linked to upstream discharge. A combination of river loss and overbank flow for highest floods is proposed for explaining the relationships. Our approach differentiates the role of outflow (river loss and overbank flow) and that of wave diffusion on peakflow attenuation. Based on several sets of model parameterization, diffusion is the main attenuation process for most cases, despite high river losses of up to several m3/s (half of peakflow for some parameterization strategies). Finally, this framework gives new insight into the SW/GW interactions during floods in karst basins, and more globally in basins characterized by disconnected river‐aquifer systems.

Comparison of Constrained Parameterisation Strategies for Aerodynamic Optimisation of Morphing Leading Edge Airfoil

In the context of ambitious targets for reducing environmental impact in the aviation sector, dictated by international institutions, morphing aircraft are expected to have potential for achieving the required efficiency increases. However, there are still open issues related to the design and implementation of deformable structures. In this paper, we compare three constrained parameterisation strategies for the aerodynamic design of a morphing leading edge, representing a potential substitute for traditional high-lift systems. In order to facilitate the structural design and promote the feasibility of solutions, we solve a multi-objective optimisation problem, including constraints on axial and bending strain introduced by morphing. A parameterisation method, inherently producing constant arc length curves, is employed in three variants, representing different morphing strategies which provide an increasing level of deformability, by allowing the lower edge of the flexible skin to slide and the gap formed with the fixed spar to be closed by a hatch. The results for the optimisation of a baseline airfoil show that the geometric constraints are effectively handled in the optimisation and the solutions are smooth, with a continuous variation along the Pareto frontier. The larger shape modification allowed by more flexible parameterisation variants enables an increase of the maximum lift coefficient up to 8.35%, and efficiency at 70% of stall incidence up to 4.26%.

Aerothermal Optimization of Fully Cooled Turbine Blade Tips

Optimal turbine blade tip designs have the potential to enhance aerodynamic performance while reducing the thermal loads on one of the most vulnerable parts of the gas turbine. This paper describes a novel strategy to perform a multi-objective optimization of the tip geometry of a cooled turbine blade. The parameterization strategy generates arbitrary rim shapes around the coolant holes on the blade tip. The tip geometry performance is assessed using steady Reynolds-averaged Navier–Stokes simulations with the k–ω shear stress transport (SST) model for the turbulence closure. The fluid domain is discretized with hexahedral elements, and the entire optimization is performed using identical mesh characteristics in all simulations. This is done to ensure an adequate comparison among all investigated designs. Isothermal walls were imposed at engine-representative levels to compute the convective heat flux for each case. The optimization objectives were a reduction in heat load and an increase in turbine row efficiency. The multi-objective optimization is performed using a differential evolution strategy. Improvements were achieved in both the aerodynamic efficiency and heat load reduction, relative to a conventional squealer tip arrangement. Furthermore, this work demonstrates that the inclusion of over-tip coolant flows impacts the over-tip flow field, and that the rim–coolant interaction can be used to create a synergistic performance enhancement.

Resource Allocation in Information-Centric Wireless Networking With D2D-Enabled MEC: A Deep Reinforcement Learning Approach

Recently, information-centric wireless networks (ICWNs) have become a promising Internet architecture of the next generation, which allows network nodes to have computing and caching capabilities and adapt to the growing mobile data traffic in 5G high-speed communication networks. However, the design of ICWN is still faced with various challenges with respect to capacity and traffic. Therefore, mobile edge computing (MEC) and device-to-device (D2D) communications can be employed to aid offloading the core networks. This paper investigates the optimal policy for resource allocation in ICWNs by maximizing the spectrum efficiency and system capacity of the overall network. Due to unknown and stochastic properties of the wireless channel environment, this problem was modeled as a Markov decision process. In continuous-valued state and action variables, the policy gradient approach was employed to learn the optimal policy through interactions with the environment. We first recognized the communication mode according to the location of the cached content, considering whether it is D2D mode or cellular mode. Then, we adopt the Gaussian distribution as the parameterization strategy to generate continuous stochastic actions to select power. In addition, we use softmax to output channel selection to maximize system capacity and spectrum efficiency while avoiding interference to cellular users. The numerical experiments show that our learning method performs well in a D2D-enabled MEC system.