Unified Modeling Framework Research Articles

Phase-field simulation tending to depict practical electrodeposition process in lithium-based batteries

Lithium dendrite growth due to uneven electrodeposition usually leads to the potential hazard of internal short circuit and shorter lifetime of lithium-based batteries. Extensive efforts have been devoted to explore the effects of single or two factors on dendrite growth, involving the diffusion coefficient, exchange current density, electrolyte concentration, temperature, and applied voltage. However, these factors interrelate during battery operation, signifying that a understanding of how they jointly influence the electrodeposition is of paramount importance for the effective suppression of dendrites. Here, we incorporate the dependent relationships among key factors into the phase-field model to capture their synergistic effects on electrodeposition. All the simulations are implemented in our self-written MATLAB code under a unified modeling framework. Following this, five groups of experimentally common dendrite patterns are reproduced and the corresponding electrodeposition driving forces are identified. Unexpectedly, we find that with the decrease of the ratio of exchange current density (or applied voltage) to diffusion coefficient, the electrodeposition morphology changes from needle-like dendrites to columnar dendrites and to uniform deposition. The present phase-field simulation tends to depict the practical electrodeposition process, providing important insights into synergistic regulation to suppress dendrite growth.

Spatial data analysis for intelligent buildings: Awareness of context and data uncertainty.

Intelligent buildings are among the most active Internet-of-Things (IoT) verticals, encompassing various IoT-enabled devices and sensing technologies for digital transformation. Analysis of spatial data, a very common type of data collected in intelligent buildings, offers a lot of insights for many purposes such as facilitating space management and enhancing the utilization efficiency of buildings. In this paper, we recognize two major challenges in spatial data analysis for intelligent buildings (SDAIB): (1) the complicated analytical contexts that are related to the building space and internal entities and (2) the uncertainty of spatial data due to the limitations of positioning and other sensing technologies. To address these challenges, we identify and categorize different kinds of analytical contexts and spatial data uncertainties in SDAIB, and propose a unified modeling framework for handling them. Furthermore, we showcase how the proposed framework and the associated modeling techniques are used to enable context-aware and uncertainty-aware SDAIB, in the tasks of hotspot discovery, path planning, semantic trajectory generation, and distance monitoring. Finally, we offer several research directions of SDAIB, in line with the emerging trends of the IoT.

Do carbon taxes affect economic and environmental efficiency? The case of British Columbia’s manufacturing plants

This paper evaluates the impact of British Columbia’s carbon tax on manufacturers’ economic and environmental performance in a unified modeling framework that allows for a critical distinction between efficiency, technical change, and total factor productivity as performance measures. In contrast to most papers that examine environmental policy impacts on either the economy or the environment, our approach combines a by-production model within a stochastic frontier framework to evaluate the tax’s impacts on both economic and environmental efficiency. Our findings suggest that a 1.0% increase in the carbon tax improved manufacturers’ efficiency in producing desirable output (real sales of manufactured goods) by 0.5%. In addition, the same 1.0% increase in the carbon tax improved manufacturers’ environmental efficiency for greenhouse gas (GHGs) and carbon monoxide (CO) emissions by the same amount, 0.2%. However, the carbon tax led to lower environmental efficiency for emissions of nitrogen oxides (NOX), −0.3%. In addition, our use of a rich plant-level dataset reveals considerable heterogeneity in manufacturers’ efficiency responses to the tax. Finally, we suggest that lower efficiency levels for undesirable outputs than desirable outputs indicate that the relative cost of adjusting production processes to improve efficiency favors economic efficiency over environmental efficiency.

Future‐proofing photovoltaics module reliability through a unifying predictive modeling framework

AbstractSolar energy, especially photovoltaics (PV), plays a significant role in the global energy transition required for decarbonization. Technological advancements in increasing efficiencies coupled with cost reductions through economies of scale made PV a cost‐competitive energy source set to exceed the milestone of 1 TW globally deployed capacity in 2022. However, ongoing downward price pressure coupled with increasing service life expectations creates tremendous challenges for reliability engineers in ensuring the safe and reliable operation of PV modules and systems over the anticipated service life. Today's reliability research efforts aim for module lifetimes of up to 50 years. Still, our current tools fall short in accurately assessing degradation mechanisms and failure modes over such extended periods. We argue that the well‐established PV reliability learning cycle needs to be accelerated to keep up with the rapid technological advancements and high expectations that are put on PV and its role in the global energy transition. In this article, we explore the evolution of the PV reliability learning cycle and highlight the significance that predictive modeling capabilities will have on future PV module reliability. We propose creating a unifying modeling framework—which once established will enable the holistic assessment of PV module reliability, accelerate the PV reliability learning cycle, and bring us closer to quantitative service life predictions of PV modules.

Shunts vs networks: tuning and comparison of centralized and decentralized piezoelectric vibration absorbers

This work develops a unified modeling framework for piezoelectric structures controlled by passive shunts connected to a single transducer and/or networks interconnecting multiple transducers. A common tuning procedure for these different control approaches, termed decentralized and centralized approaches, respectively, is proposed. The generic model is then used to compare them in terms of vibration mitigation performance. It is shown that decentralization can be detrimental to performance in general. Digital vibration absorbers are then leveraged to realize the shunts and/or networks. In this regard, the proposed tuning procedure solely relies on characteristics that can be identified from the digital units of these absorbers. The theoretical developments are numerically and experimentally validated on piezoelectric beams.

Nonequilibrium statistical mechanics and optimal prediction of partially-observed complex systems

Only a subset of degrees of freedom are typically accessible or measurable in real-world systems. As a consequence, the proper setting for empirical modeling is that of partially-observed systems. Notably, data-driven models consistently outperform physics-based models for systems with few observable degrees of freedom; e.g. hydrological systems. Here, we provide an operator-theoretic explanation for this empirical success. To predict a partially-observed system’s future behavior with physics-based models, the missing degrees of freedom must be explicitly accounted for using data assimilation and model parametrization. Data-driven models, in contrast, employ delay-coordinate embeddings and their evolution under the Koopman operator to implicitly model the effects of the missing degrees of freedom. We describe in detail the statistical physics of partial observations underlying data-driven models using novel maximum entropy and maximum caliber measures. The resulting nonequilibrium Wiener projections applied to the Mori–Zwanzig formalism reveal how data-driven models may converge to the true dynamics of the observable degrees of freedom. Additionally, this framework shows how data-driven models infer the effects of unobserved degrees of freedom implicitly, in much the same way that physics models infer the effects explicitly. This provides a unified implicit-explicit modeling framework for predicting partially-observed systems, with hybrid physics-informed machine learning methods combining both implicit and explicit aspects.

Generic Solving of Physiologically-based Kinetic Models in Support of Next Generation Risk Assessment Due to Chemicals

Increasing confidence in using in vitro and in silico model-based data to aid the chemical risk assessment process is one of the most significant challenges faced by regulatory authorities. A crucial concern is taking full advantage of scientifically valid physiologically-based kinetic (PBK) models. Nevertheless, risk assessors remain unwilling to employ PBK models in their daily work. Indeed, PBK models are not often included in current official guidance documents. In addition, most users have limited experience with modeling in general. So, the complexity of PBK models and a lack of evaluation methods for their performance certainly contribute to their limited use in practical risk assessment. This paper proposes an innovative and unified modeling framework for writing PBK equations as matrix ordinary differential equations (ODE) and their solutions, expressed with matrix products. This generic PBK solution considers as many state variables as needed to quantify chemical absorption, distribution, metabolism, and excretion processes within living organisms when exposed to chemical substances. This generic PBK model makes possible any compartmentalization to be considered, as well as all appropriate interconnections between compartments and with the external medium. We first introduce our PBK modeling framework, with all the intermediate steps from the matrix ODE to the exact solution. Then we apply this framework to bioaccumulation testing before illustrating its concrete use through complementary case studies regarding species, compounds, and model complexity.

Concave likelihood-based regression with finite-support response variables.

We propose a unified framework for likelihood-based regression modeling when the response variable has finite support. Our work is motivated by the fact that, in practice, observed data are discrete and bounded. The proposed methods assume a model which includes models previously considered for interval-censored variables with log-concave distributions as special cases. The resulting log-likelihood is concave, which we use to establish asymptotic normality of its maximizer as the number of observations n tends to infinity with the number of parameters d fixed, and rates of convergence of L1 -regularized estimators when the true parameter vector is sparse and d and n both tend to infinity with . We consider an inexact proximal Newton algorithm for computing estimates and give theoretical guarantees for its convergence. The range of possible applications is wide, including but not limited to survival analysis in discrete time, the modeling of outcomes on scored surveys and questionnaires, and, more generally, interval-censored regression. The applicability and usefulness of the proposed methods are illustrated in simulations and dataexamples.

The GALPROP Cosmic-ray Propagation and Nonthermal Emissions Framework: Release v57

The past decade has brought impressive advances in the astrophysics of cosmic rays (CRs) and multiwavelength astronomy, thanks to the new instrumentation launched into space and built on the ground. Modern technologies employed by those instruments provide measurements with unmatched precision, enabling searches for subtle signatures of dark matter and new physics. Understanding the astrophysical backgrounds to better precision than the observed data is vital in moving to this new territory. A state-of-the-art CR propagation code, called GalProp, is designed to address exactly this challenge. Having 25 yr of development behind it, the GalProp framework has become a de facto standard in the astrophysics of CRs, diffuse photon emissions (radio to γ-rays), and searches for new physics. GalProp uses information from astronomy, particle physics, and nuclear physics to predict CRs and their associated emissions self-consistently, providing a unifying modeling framework. The range of its physical validity covers 18 orders of magnitude in energy, from sub-keV to PeV energies for particles and from μeV to PeV energies for photons. The framework and the data sets are public and are extensively used by many experimental collaborations and by thousands of individual researchers worldwide for interpretation of their data and for making predictions. This paper details the latest release of the GalProp framework and updated cross sections, further developments of its initially auxiliary data sets for models of the interstellar medium that grew into independent studies of the Galactic structure—distributions of gas, dust, radiation, and magnetic fields—as well as the extension of its modeling capabilities. Example applications included with the distribution illustrating usage of the new features are also described.

Incorporating dynamic crop growth processes and management practices into a terrestrial biosphere model for simulating crop production in the United States: Toward a unified modeling framework

Agricultural decision-making by different interest groups (e.g., farmers, development agents and policy makers) usually takes place on different scales (e.g., plot, landscape and country). Currently, tools to assist decision-making are either dedicated to small-scale management guidance or large-scale assessment, which ignore the cross-scale linkages and interactions and thus may not provide robust and consistent guidance and assessment. Here, we developed an advanced agricultural modeling framework by integrating the strengths of conventional crop models in representing crop growth processes and management practices into a terrestrial biosphere model (TBM), the Dynamic Land Ecosystem Model (DLEM), to meet the cross-scale application needs (e.g., adaptation and mitigation). Specifically, dynamic crop growth processes, including crop-specific phenological development, carbon allocation, yield formation, biological nitrogen fixation processes, and management practices such as tillage, cover cropping and genetic improvements, were explicitly represented in DLEM. The new model was evaluated against site-scale observations and the results showed that the model performed generally well, with an average normalized root mean square error of 19.91% for leaf area index and 17.46% for aboveground biomass at the seasonal scale and 14.42% for annual yield. Then the model was applied to simulate corn, soybean, and winter wheat productions in the conterminous United States from 1960 to 2018. The spatial patterns of simulated crop productions were consistent with ground survey data. Our model also captured both the long-term trends and interannual variations of the total national productions of the three crops. This study demonstrates the significance of fusing conventional crop modeling techniques into TBMs to establish a unified modeling framework, which holds the potential to address climate impacts, adaptation and mitigation across varied spatiotemporal scales.

Painters in chromatin: a unified quantitative framework to systematically characterize epigenome regulation and memory.

In eukaryotes, many stable and heritable phenotypes arise from the same DNA sequence, owing to epigenetic regulatory mechanisms relying on the molecular cooperativity of ‘reader–writer’ enzymes. In this work, we focus on the fundamental, generic mechanisms behind the epigenome memory encoded by post-translational modifications of histone tails. Based on experimental knowledge, we introduce a unified modeling framework, the painter model, describing the mechanistic interplay between sequence-specific recruitment of chromatin regulators, chromatin-state-specific reader–writer processes and long-range spreading mechanisms. A systematic analysis of the model building blocks highlights the crucial impact of tridimensional chromatin organization and state-specific recruitment of enzymes on the stability of epigenomic domains and on gene expression. In particular, we show that enhanced 3D compaction of the genome and enzyme limitation facilitate the formation of ultra-stable, confined chromatin domains. The model also captures how chromatin state dynamics impact the intrinsic transcriptional properties of the region, slower kinetics leading to noisier expression. We finally apply our framework to analyze experimental data, from the propagation of γH2AX around DNA breaks in human cells to the maintenance of heterochromatin in fission yeast, illustrating how the painter model can be used to extract quantitative information on epigenomic molecular processes.

Response Model for Urban Area Source Pollution and Water Environmental Quality in a River Network Region.

With the development of cities, urban area source pollution has become more severe and a significant source of water pollution. To study the relationship between urban area source pollution and water environmental quality in a river network, this study uses a city in the Yangtze River Delta, China, as an example. The Storm Water Management Model (SWMM) model and the MIKE11 model were combined into a unified modeling framework and used to simulate dynamic changes in the water quality of a river network under light rain, moderate rain, and heavy rain. In the study period, the annual urban area source input loads of potassium permanganate (CODMn), total phosphorus (TP), and ammonia nitrogen were 29.8, 0.9, and 4.8 t, respectively. The influence of light rain on the water quality of the river network was lagging and temporary, and rainfall area pollution was the primary contributor. Under the scenario of moderate rain, overflow from a pipeline network compounded rainfall runoff, resulting in a longer duration of impact on the water quality in the river. Additionally, the water quality in the river course was worse under moderate rain than under light or heavy rain. Under the scenario of heavy rain, rain mainly served a dilutive function. This research can provide support for urban area source pollution control and management.

Association between community psychological label and user portrait model based on multimodal neural network.

By analyzing traditional deep learning multimode retrieval methods, an optimized multimode retrieval model based on convolutional neural network is established. This article proposes an innovative semi-supervised social network user portrait analysis model (UPAM) based on user portrait model, which integrates users’ social information and some known user attribute information (such as educational background and residence) into a unified topic model framework. Finally, a semi-supervised user portrait analysis method based on user social information and partial known user attribute information is proposed. According to the correlation of user attributes, the cross-validation method is used to train model prediction task and improve the prediction effect. In the first-level model, using a different model to extract the features in the user query, the basis of the second hierarchy model, Stacking is used to further integrate characteristics, finally realizing the attribute population forecast, and experimental verification showing the proposed model’s effectiveness in various attributes of a population.

Unified modelling framework of flowslide triggering and runout

Hydro-mechanical forcing, including sediment deposition, seismic loading, rainfall events and phreatic level rise, can alter the pore water pressure regime and trigger ground failure. In loose sands, such phenomena can manifest with rapid downslope movements displaying signatures of liquefaction and commonly referred to as ‘flowslides’. These can have catastrophic outcomes, yet the use of constitutive models for soils able to quantify the likelihood of liquefaction is usually restricted to triggering analyses, while the quantification of runout distance and velocity tends to be conducted through computationally expensive models based on rheological laws for fluids. To enable a seamless transition from the triggering to the runout stage, in this paper, a spatially condensed modelling framework of low computational cost is proposed, which allows for straightforward use of soil constitutive models in the analysis of the entire life cycle of a flowslide. The proposed framework couples the dynamic motion of a landslide mass with the pore fluid transients taking place within a pre-defined liquefiable zone. The constitutive law modulates the feedback between pore fluid diffusion and landslide movement, encompassing elastoplastic and viscoplastic soil models within the same formulation. Simulations show that partial drainage in the liquefiable zone has considerable effects on the triggering stresses initiating a flowslide. Specifically, the results imply that liquefaction risks are sharply suppressed when the ratio between the characteristic times of loading and pore pressure diffusion falls above a critical threshold influenced by the inelastic deformation response of the soil. Furthermore, it is shown that all the dynamics of flowslide propagation are dominated by the ratio between the characteristic timescales related to fluid consolidation, wave propagation and soil reaction time (i.e. soil viscosity). Although viscosity is not necessary to carry out propagation analyses, ignoring it leads to unrealistic estimates of peak velocity and runout distance. As a result, the model predicts that sand viscosity reflecting the emergence of a microscopic inertial regime is required to explain the flowslide propagation characteristics reported in the literature.

Human–Computer Interaction Cognitive Behavior Modeling of Command and Control Systems

Human–computer interaction cognitive behavior (HCICB) modeling faces four deficiencies: 1) lack of a standard framework model; 2) large simulation error; 3) single simulation dimension; and 4) lack of a simulation software. To solve these deficiencies, we have carried out work in four aspects. First, we construct an HCICB model with the user, system device, and environment as the core elements, which provides a unified framework for the subsequent HCICB modeling in the Military Internet of Things (MIoT) command and control (C2) system. Second, we correct visual and motion parameters in the adaptive control of thought rational module of the Cogtool model by the commander in the loop (CIL) experiment. Third, we construct a mental workload (MW) prediction model of the MIoT C2 system based on improved visual auditory cognitive psychomotor, which realizes fast, high-precision, and quantitative MW prediction. It is added as a simulation dimension for the HCICB. Fourth, we develop MwCogtool, an HCICB prediction software that can rapidly simulate typical tasks at the design and usage stages of the MIoT C2 system, and also can output six parameters, including task completion time (TCT), MW, eye movement prepare time, eye movement execution time, motion time, and cognitive time in the whole process quickly and visually. In addition, we select 20 real users and 9 typical tasks of the MIoT C2 system to carry out the CIL verification experiment. Compared with Cogtool, MwCogtool reduces the maximum simulation error in TCT of the C2 system from 45.00% to 5.58%. The consistency of simulation results with real user data reaches 0.99. The results of the MW prediction model can significantly and negatively predict the change of real users’ eye movement, and can accurately predict the trend of MW change. Simultaneously, we build a fitting model between the mean MW prediction value and eye movement parameters.

A unified effective medium modeling framework for quantitative characterization of hydrate reservoirs

Quantitative characterization of hydrate reservoirs is essential for evaluating hydrate resources and monitoring environmental hazards. Rock-physics models provide the theoretical basis for linking the hydrate saturation and distribution to the elastic properties. However, many of the existing models ignore the coupled effect of hydrate saturation and complex morphologies on the elastic behaviors and thereby fail to accurately describe hydrate reservoirs. To address these problems, the authors have proposed a unified effective medium modeling framework honoring multiple hydrate morphologies. In this framework, the manners in which the different types of hydrates interact with the sediment grains are characterized by different effective medium models. In addition to the hydrate saturation, the modeling results also reveal the effects of the cement factor, friction coefficient, incidence angle, lithology, and hydrate morphology on the elastic properties of the hydrate reservoirs. Then, five experimental data sets for synthetic methane and tetrahydrofuran hydrates formed under different conditions are used to validate the feasibility of the unified rock-physics modeling scheme. The good agreement between the theoretical predictions and laboratory data indicates that this method can effectively determine the hydrate occurrence mechanisms from the acoustic responses in different hydrate formation environments. On the basis of the constructed elastic rock-physics templates, multiple hydrate morphologies are determined with cluster analysis and then used for constraining the modeling scheme. Finally, the proposed modeling method is tested using sonic logs from Ignik Sikumi, North Slope of Alaska, and Hole U1518B, the northern Hikurangi margin, offshore New Zealand. The hydrate saturations estimated from the velocities are consistent with the interpretations based on resistivity and nuclear magnetic resonance logs, as well as chloride measurements, confirming the applicability and feasibility of our unified theoretical modeling framework.

Continuum Robots for Medical Interventions.

Continuum robots are not constructed with discrete joints but, instead, change shape and position their tip by flexing along their entire length. Their narrow curvilinear shape makes them well suited to passing through body lumens, natural orifices, or small surgical incisions to perform minimally invasive procedures. Modeling and controlling these robots are, however, substantially more complex than traditional robots comprised of rigid links connected by discrete joints. Furthermore, there are many approaches to achieving robot flexure. Each presents its own design and modeling challenges, and to date, each has been pursued largely independently of the others. This article attempts to provide a unified summary of the state of the art of continuum robot architectures with respect to design for specific clinical applications. It also describes a unifying framework for modeling and controlling these systems while additionally explaining the elements unique to each architecture. The major research accomplishments are described for each topic and directions for the future progress needed to achieve widespread clinical use are identified.

A 3-D virtual human model for simulating heat and cold stress.

In this study, we extended our previously developed anatomically detailed three-dimensional (3-D) thermoregulatory virtual human model for predicting heat stress to allow for predictions of heat and cold stress in one unified model. Starting with the modified Pennes bioheat transfer equation to estimate the spatiotemporal temperature distribution within the body as the underlying modeling structure, we developed a new formulation to characterize the spatial variation of blood temperature between body elements and within the limbs. We also implemented the means to represent heat generated from shivering and skin blood flow that apply to air exposure and water immersion. Then, we performed simulations and validated the model predictions with experimental data from nine studies, representing a wide range of heat- and cold-stress conditions in air and water and physical activities. We observed excellent agreement between model predictions and measured data, with average root mean squared errors of 0.2°C for core temperature, 0.9°C for mean skin temperature, and 27 W for heat from shivering. We found that a spatially varying blood temperature profile within the limbs was crucial to accurately predict core body temperature changes during very cold exposures. Our 3-D thermoregulatory virtual human model consistently predicted the body's thermal state accurately for each of the simulated hot and cold environmental conditions and exertional heat stress. As such, it serves as a reliable tool to assess whole body, localized tissue, and, potentially, organ-specific injury risks, helping develop injury prevention and mitigation strategies in a systematic and expeditious manner.NEW & NOTEWORTHY This work provides a new, unified modeling framework to accurately predict the human body's thermal response to both heat and cold stress caused by environmental conditions and exertional physical activity in one mathematical model. We show that this 3-D anatomically detailed model accurately predicts the spatiotemporal temperature distribution in the body under extreme conditions for exposures to air and water and could be used to help design medical interventions and countermeasures to prevent injuries.

A Novel Unified Data Modeling Method for Equipment Lifecycle Integrated Logistics Support.

Integrated logistics support (ILS) is of great significance for maintaining equipment operational capability in the whole lifecycle. Numerous segments and complex product objects exist in the process of equipment ILS, which gives ILS data multi-source, heterogeneous, and multidimensional characteristics. The present ILS data cannot satisfy the demand for efficient utilization. Therefore, the unified modeling of ILS data is extremely urgent and significant. In this paper, a unified data modeling method is proposed to solve the consistent and comprehensive expression problem of ILS data. Firstly, a four-tier unified data modeling framework is constructed based on the analysis of ILS data characteristics. Secondly, the Core unified data model, Domain unified data model, and Instantiated unified data model are built successively. Then, the expressions of ILS data in the three dimensions of time, product, and activity are analyzed. Thirdly, the Lifecycle ILS unified data model is constructed, and the multidimensional information retrieval methods are discussed. Based on these, different systems in the equipment ILS process can share a set of data models and provide ILS designers with relevant data through different views. Finally, the practical ILS data models are constructed based on the developed unified data modeling software prototype, which verifies the feasibility of the proposed method.

Document Summarization with Latent Queries

AbstractThe availability of large-scale datasets has driven the development of neural models that create generic summaries for single or multiple documents. For query-focused summarization (QFS), labeled training data in the form of queries, documents, and summaries is not readily available. We provide a unified modeling framework for any kind of summarization, under the assumption that all summaries are a response to a query, which is observed in the case of QFS and latent in the case of generic summarization. We model queries as discrete latent variables over document tokens, and learn representations compatible with observed and unobserved query verbalizations. Our framework formulates summarization as a generative process, and jointly optimizes a latent query model and a conditional language model. Despite learning from generic summarization data only, our approach outperforms strong comparison systems across benchmarks, query types, document settings, and target domains.1