Model Code Research Articles

Testing Performance of Modeling and Simulation Code of Liquid Propellant Supply System Using Method of Characteristics

The primary objective of a liquid propellant supply system is to supply propellant to the engine, enabling stable mission performance. Therefore, the system design must aim at minimizing factors that may contribute to instability while satisfying the specified requirements. Modeling and simulation (M&S) techniques are typically used to achieve efficient design, offering advantages such as the ability to simulate various conditions without constraints, thereby reducing initial development costs and time. Consequently, the performance and accuracy of the M&S program can significantly impact the overall development process. In this study, the method of characteristics (MOC) is used to effectively calculate internal flow in pipes, thereby developing an M&S code with excellent accuracy and performance. The efficiency of the developed in-house code is evaluated by comparing the simulation results with those of a commercial program. Additionally, a water supply experiment is conducted, and the experimental results are compared with the two simulation outcomes for validation. The comparative analysis highlights the superior performance and accuracy of the in-house code over the commercial program, with the in-house code demonstrating an average accuracy improvement of approximately 2% to 3% and with computational performance being approximately 1.7 times faster.

Numerical analysis of the corrosion impact on the punching shear capacity of flat slabs with openings

AbstractThe effects generated by punching shear failure in flat slabs are catastrophic, as the failure of the slab‐column joint can cause a progressive collapse of the building. Furthermore, when exposed to aggressive environments, reinforced concrete structures can be degraded due to the penalization of the mechanical and geometric properties of the reinforcement and concrete. Therefore, this study intends to investigate the behavior of corroded reinforced concrete flat slabs with adjacent openings to the column. Three nonlinear finite element models of flat slabs based on experimental results were constructed and validated in the ABAQUS software and subsequently subjected to a parametric analysis to investigate the effects of different corrosion degrees on their punching shear capacity. The results demonstrated that corrosion of the reinforcement significantly affected the load capacity of the numerical models, reducing the ultimate load by up to 30%. Furthermore, the increase in corrosion caused an intensification of the cracking of the slabs, indicating that the cracking patterns increased as the degree of corrosion increased. The results obtained by experimental tests and numerical simulations were compared with predictions by the codes Eurocode, fib Model Code, American Concrete Institute 318, FprEN 1992 1‐1, and The Brazilian Association of Technical Standards Brazilian Regulatory Standard 6118. The results showed that most codes generate very conservative results.

Cross sections of the 226Ra(p,xn) reactions relevant for 225Ac production.

Limited availability constrains the implementation of 225Ac, the most promising α emitter for targeted therapy, in clinical practice. Proton activation of 226Ra is one of few realistic solutions to this problem. We have therefore measured cross sections of relevant 226Ra(p,xn) nuclear reactions in the energy range of 12.0 to 17.2MeV. The obtained results allowed us to deduce the yield of 224Ac, 225Ac, and 226Ac. The consequences of our data to predictions of production capacity and radionuclidic purity of 225Ac are thoroughly discussed, and their comparison with previous measurements and with the prediction of the TALYS 1.96 nuclear reaction model code run with default parameters are provided.

Low-energy incomplete fusion: A systematic study of entrance channel parameters

Research into heavy-ion fusion, a key area of modern nuclear reaction physics, has flourished in recent decades pursuant to developments in accelerator technology. The primary goal of studying heavy-ion reactions is to gain knowledge about the underlying processes and how they are affected by entrance channel parameters, such as beam energy, angular momentum and mass asymmetry. The fusion mechanism of non-[Formula: see text]-cluster projectiles, such [Formula: see text]N and [Formula: see text]F, has been studied in the low-energy zone. It has been challenging to analyze the contributing degrees of freedom in such reactions due to the absence of experimental data. The present study reports the measurement of residual cross-sections from the [Formula: see text]F induced reaction on [Formula: see text]Nb within the energy range of 3–6 MeV/A. The stack foil activation technique followed by offline [Formula: see text] spectroscopy was employed to measure the cross-sections of residues populated in the reaction. The experimental data were compared with theoretical predictions from statistical model code PACE4 to probe the underlying reaction dynamics. The imitation of xn and pxn channel data grossly by model code suggests the production of residues via the complete fusion (CF) mode, while the enhancement observation in [Formula: see text]-channel cross-sections hints at the signatures of incomplete fusion (ICF) in addition to the dominant CF. Thus, the ICF strength fraction ([Formula: see text]) was calculated. Moreover, the estimated incomplete fusion fraction has been used to study the effect of several entrance channel parameters on incomplete fusion reaction dynamics. The present analysis shows the presence of strong clustering in the [Formula: see text]F projectile as [Formula: see text] and [Formula: see text]N.

Evaluation and Future Prospects of Data‐Driven Intelligence‐Based Framework for Predicting Cyclic Behavior of Reconstituted Sand

ABSTRACTMost of the robust artificial intelligence (AI)‐based constitutive models are developed with synthetic datasets generated from traditional constitutive models. Therefore, they fundamentally rely on the traditional constitutive models rather than laboratory test results. Also, their potential use within geotechnical engineering communities is limited due to the unavailability of datasets along with the model code files. In this study, the data‐driven constitutive models are developed using only laboratory test databases and deep learning (DL) techniques. The laboratory database was prepared by conducting cyclic direct simple shear (CDSS) tests on reconstituted sand, that is, PDX sand. The stacked long short‐term memory (LSTM) network and its variants are considered for developing the predictive models of the shear strain (γ [%]) and excess pore pressure ratio (ru) time histories. The suitable input parameters (IPs) are selected based on the physics behind the generation of ru and γ (%) of the liquefiable sands. The predicted responses of γ (%) and ru agree well in most cases and are used to predict the dynamic soil properties of the PDX sand. The same modeling framework is extended for other sand and compared with existing AI‐based constitutive models to verify its practical applicability. In summary, it is observed that though the trained models predicted the time histories of ru and γ reasonably well; however, they struggled to predict the hysteresis loops at higher cycles. Therefore, more research is needed to verify and enhance the predictability of existing AI‐based models in the future before using them in practice for simulating cyclic response.

TensorRight: Automated Verification of Tensor Graph Rewrites

Tensor compilers, essential for generating efficient code for deep learning models across various applications, employ tensor graph rewrites as one of the key optimizations. These rewrites optimize tensor computational graphs with the expectation of preserving semantics for tensors of arbitrary rank and size. Despite this expectation, to the best of our knowledge, there does not exist a fully automated verification system to prove the soundness of these rewrites for tensors of arbitrary rank and size. Previous works, while successful in verifying rewrites with tensors of concrete rank, do not provide guarantees in the unbounded setting. To fill this gap, we introduce TensorRight, the first automatic verification system that can verify tensor graph rewrites for input tensors of arbitrary rank and size. We introduce a core language, TensorRight DSL, to represent rewrite rules using a novel axis definition, called aggregated-axis , which allows us to reason about an unbounded number of axes. We achieve unbounded verification by proving that there exists a bound on tensor ranks, under which bounded verification of all instances implies the correctness of the rewrite rule in the unbounded setting. We derive an algorithm to compute this rank using the denotational semantics of TensorRight DSL. TensorRight employs this algorithm to generate a finite number of bounded-verification proof obligations, which are then dispatched to an SMT solver using symbolic execution to automatically verify the correctness of the rewrite rules. We evaluate TensorRight’s verification capabilities by implementing rewrite rules present in XLA’s algebraic simplifier. The results demonstrate that TensorRight can prove the correctness of 115 out of 175 rules in their full generality, while the closest automatic, bounded -verification system can express only 18 of these rules.

Performance of machine learning tools. Comparve analysis of libraries in interpreted and compiled programming languages

The article compares machine learning tools using the example of several popular programming languages. Existing tools in the following programming languages were tested and compared with each other: Python, Java, R, Julia, C#. For the needs of article, algorithms were created in each studied language, operating on the same test set and using algorithms from the same group. The collected results included the program's running time, number of lines of code and accuracy of trained model. Based on the obtained data, conclusions were drawn that interpreted language libraries in terms of creating machine learning solutions are more effective than compiled language libraries.

STOUT V2.0: SMILES to IUPAC name conversion using transformer models

Naming chemical compounds systematically is a complex task governed by a set of rules established by the International Union of Pure and Applied Chemistry (IUPAC). These rules are universal and widely accepted by chemists worldwide, but their complexity makes it challenging for individuals to consistently apply them accurately. A translation method can be employed to address this challenge. Accurate translation of chemical compounds from SMILES notation into their corresponding IUPAC names is crucial, as it can significantly streamline the laborious process of naming chemical structures. Here, we present STOUT (SMILES-TO-IUPAC-name translator) V2, which addresses this challenge by introducing a transformer-based model that translates string representations of chemical structures into IUPAC names. Trained on a dataset of nearly 1 billion SMILES strings and their corresponding IUPAC names, STOUT V2 demonstrates exceptional accuracy in generating IUPAC names, even for complex chemical structures. The model's ability to capture intricate patterns and relationships within chemical structures enables it to generate precise and standardised IUPAC names. While established deterministic algorithms remain the gold standard for systematic chemical naming, our work, enabled by access to OpenEye’s Lexichem software through an academic license, demonstrates the potential of neural approaches to complement existing tools in chemical nomenclature.Scientific contribution STOUT V2, built upon transformer-based models, is a significant advancement from our previous work. The web application enhances its accessibility and utility. By making the model and source code fully open and well-documented, we aim to promote unrestricted use and encourage further development.Graphical

Integrating single-cell multimodal epigenomic data using 1D convolutional neural networks.

Recent experimental developments enable single-cell multimodal epigenomic profiling, which measures multiple histone modifications and chromatin accessibility within the same cell. Such parallel measurements provide exciting new opportunities to investigate how epigenomic modalities vary together across cell types and states. A pivotal step in using these types of data is integrating the epigenomic modalities to learn a unified representation of each cell, but existing approaches are not designed to model the unique nature of this data type. Our key insight is to model single-cell multimodal epigenome data as a multichannel sequential signal. We developed ConvNet-VAEs, a novel framework that uses one-dimensional (1D) convolutional variational autoencoders (VAEs) for single-cell multimodal epigenomic data integration. We evaluated ConvNet-VAEs on nano-CUT&Tag and single-cell nanobody-tethered transposition followed by sequencing data generated from juvenile mouse brain and human bone marrow. We found that ConvNet-VAEs can perform dimension reduction and batch correction better than previous architectures while using significantly fewer parameters. Furthermore, the performance gap between convolutional and fully connected architectures increases with the number of modalities, and deeper convolutional architectures can increase the performance, while the performance degrades for deeper fully connected architectures. Our results indicate that convolutional autoencoders are a promising method for integrating current and future single-cell multimodal epigenomic datasets. The source code of VAE models and a demo in Jupyter notebook are available at https://github.com/welch-lab/ConvNetVAE.

Handling complexity of large high-level specifications in simulation environment: the case of transformational model-battle management

Modeling system requirements at a high level can significantly facilitate the communication of mental models across highly heterogeneous entities. Two major challenges may arise during such a process. The first is the degree to which details of the higher-level specifications can capture and effectively handle. Especially, when the model contains more nuanced descriptions of parallel execution and temporal pacing. The second challenge arises when realizing such specifications downstream due to ambiguities and incompleteness encountered in higher-level semiformal artifacts. We propose a formal specification and transformation of higher-level artifacts, particularly in the operational activity model of the Department of Defense Architecture Framework (OV-5 DoDAF). We create an isomorphic activity model corresponding to each activity in the Transformational Model-Battle Management. Then, we generate a code for simulating the models as Markov Discrete Event System Specifications. We demonstrate the size and other aspects of the experiments that can be conducted based on the generated simulation and the level of complexity in such an approach. We highlight aspects of managing complexity using model transformations and code generation techniques in a full-cycle model-based system engineering approach, expediting the process toward arriving at the concrete realization of simulation models that only originated from minimal high-level metamodels and artifacts.

Predicting Polyolefin Microstructure: A Parallelized Multidimensional Model for Metallocene‐Catalyzed Copolymerization of Propylene and 1‐Decene

AbstractThis work explores the copolymerization of propylene and 1‐decene using homogeneous metallocene catalysts to optimize polyolefin functionalization. A detailed mathematical model is developed with experimental validation employing the method of moments and probability generating functions to predict average molecular properties, the molecular weight distribution, the copolymer composition distribution, and the joint molecular weight distribution‐copolymer composition distribution. To efficiently handle computational resources, the model code is parallelized. This comprehensive model allows for explaining in detail the copolymer's microstructure under various semibatch reactor conditions. Moreover, the model is a powerful tool for selecting reaction conditions to synthesize materials with desired properties.

Determination of Photoneutron Production from Different Targets Irradiated by Electron Beam

In the present work, a study has been conducted to estimate the production of photoneutrons from different targets irradiated by a 70 MeV electron beam. The linear electron accelerator at the Yerevan Physics Institute in Armenia was used for the purpose. The reaction rates were determined through detailed off-line analysis of experimental data and subsequently compared to theoretical predictions by using the MCNP (Monte Carlo N-Particle Transport) and Cascade Exciton Physics Model (CEM03.03) codes.

A model for analytical calculations of synthetic neutron energy spectra from beam-target reactions

Abstract We present a fully analytical model for calculating energy spectra of neutrons generated by fusion reactions involving a fast ion, or beam, and a stationary ion, or target, in magnetic fusion plasmas. For neutrons moving along the line-of-sight of a detector, the neutron spectrum is given by an analytical expression and the usual differential cross section. This makes the model several orders of magnitude faster than ordinary Monte Carlo simulations and free of any related statistical noise. Additionally, the analytical description of the reaction physics provides much more insight into the formation of the spectrum. An example of this is the bias of beam-target spectra towards high-energy neutron counts, which corresponds to forward-emission events. On the other hand, the fast-ion uniform gyro-angle distribution has an opposite effect, but is ultimately weaker than the preferential forward emission of neutrons. The model is validated against numerical calculations from the forward model code GENESIS to verify its validity and it is furthermore derived from a probabilistic viewpoint, adding further insight.

Moral and ethical aspects in the professional activity of a judge

The article is devoted to the understanding of moral and ethical aspects in the professional activity of a judge, which are formed on the basis of the principles of integrity, impartiality, conscientious performance of professional duties and are the standards that form the basis of public trust in the judiciary and confidence in the proper protection of human rights and freedoms and a citizen. It is emphasized that the judge’s observance of moral and ethical standards in judging and everyday personal life is a guarantee of the formation of the authority of the judiciary, the trust of citizens in the judiciary, as well as faith in a fair court, which is an important component in the process of the establishment and development of a rule of law and a democratic society. It is noted that Ukraine continues to implement reforms of the judicial system, initiated before the full-scale invasion of Russia, focusing on the principles of respect for human dignity, integrity, impartiality, honesty, honor, dignity and professionalism of judges. Also, this approach will contribute to Ukraine’s full and comprehensive fulfillment of the requirements for European integration processes. Attention is focused on the importance of compliance with a number of normative legal acts: the Bangalore principles of the conduct of judges, the Model Code of Professional Conduct, the Code of Judicial Ethics, which confirm that the scope of moral and ethical norms covers the social and legal space, being specified through basic norms and principles in various aspects law enforcement and law enforcement activities. It is noted that when administering justice on behalf of Ukraine, the judge must first of all be aware that his violation of moral and ethical norms negatively affects not only his authority as a judge, but also forms a negative perception of the judiciary in the state, and therefore, raises doubts about the extent to which the state meets international requirements for the administration of justice, as far as it is legal, etc. It is emphasized that ensuring the right of every person to judicial protection based on the establishment of the principle of the rule of law, as well as the administration of justice on behalf of the state of Ukraine on the basis of the Constitution and laws of Ukraine set high requirements for the moral qualities of a judge.

A tutorial on pharmacometric Markov models.

The Markov chain is a stochastic process in which the future value of a variable is conditionally independent of the past, given its present value. Data with Markovian features are characterized by: frequent observations relative to the expected changes in values, many consecutive same-category or similar-value observations at the individual level, and a positive correlation observed between the current and previous values for that variable. In drug development and clinical settings, the data available commonly present Markovian features and are increasingly often modeled using Markov elements or dedicated Markov models. This tutorial presents the main characteristics, evaluations, and applications of various Markov modeling approaches including the discrete-time Markov models (DTMM), continuous-time Markov models (CTMM), hidden Markov models, and item-response theory model with Markov sub-models. The tutorial has a specific emphasis on the use of DTMM and CTMM for modeling ordered-categorical data with Markovian features. Although the main body of this tutorial is written in a software-neutral manner, annotated NONMEM code for all key Markov models is included in the Supplementary Information.

Joint extraction of entity and relation based on Fine-tuning BERT for Long Biomedical literatures

Abstract Motivation Joint extraction of entity and relation is an important research direction in Information Extraction. The number of scientific and technological biomedical literature is rapidly increasing, so automatically extracting entities and their relations from these literatures are key tasks to promote the progress of biomedical research. Results The joint extraction of entity and relation model achieves both intra-sentence extraction and cross-sentence extraction, alleviating the problem of long-distance information dependence in long literature. Joint extraction of entity and relation model incorporates a variety of advanced deep learning techniques in this paper: (1) a fine-tuning BERT text classification pre-training model, (2) Graph Convolutional Network learning method, (3) Robust Learning Against Textual Label Noise with Self-Mixup Training, (4) Local regularization Conditional Random Fields. The model implements the following functions: identifying entities from complex biomedical literature effectively, extracting triples within and across sentences, reducing the effect of noisy data during training, and improving the robustness and accuracy of the model. The experiment results prove that the model performs well on the self-built BM_GBD dataset and public datasets, enabling precise large language model enhanced knowledge graph construction for biomedical tasks. Availability and implementation The model and partial code are available on GitHub at https://github.com/zhaix922/Joint-extraction-of-entity-and-relation.

AssertHE: an R package to improve quality assurance of HTA models

Background Health economic models are increasingly used to inform decisions about the allocation of healthcare resources. Ensuring the robustness and reliability of these models is critical. Currently, quality assurance is conducted by both technical and non-technical experts assessing different components of the model manually. This is resource intensive. Understanding how the different components of the model fit together is time consuming, and testing every part of the model is sometimes not feasible in the time available. To aid in this, we have developed the assertHE R package. Methods The open-source assertHE package provides testing functionality for those building and reviewing health economic models built in R programming language. It provides a series of checks which can be integrated into the model development workflow to reduce the probability of common errors. It also provides a suite of functions which allow users to better understand the network of R functions contained in a model, where they are defined, if (and where) they are tested, and provides a simple metric to quantify the extent to which they are tested. Results We applied the assertHE package to three open-source health economic models built in R, showing how to include check functions within the model code and how to visualise the network of functions, see the test coverage, and obtain a Generative Pretrained Transformer Large Language Model (GPT-LLM) generated summary of any function in the code-base. We have worked with collaborators from industry, regulators and academia to develop the package to be applicable to the widest possible range of models, making adaptations to the source code based upon feedback. Conclusions assertHE offers an open-source toolkit for health economists building and reviewing models, promoting collaborative development and facilitating a more robust and efficient quality assurance process.

Comparison between simplified and refined models for deflection in continuous reinforced concrete T-beams

Verification of the Serviceability Limit State for allowable displacements is a mandatory step in the design of a reinforced concrete (RC) beam. In order to proceed this verification, simplified methods recommended by Design Codes or analytical models can be employed. In these models, the variation in flexural stiffness due to the concrete cracking is the primary factor to be considered. A “T” cross-section beam present greater stiffness compared to a rectangular section of equal height and area. The concentration of more area at one end of the section raises questions about whether this type of section exhibits the same behavioral response as the rectangular section when applying the models for predicting deflection as indicated by Codes. In this work, a comparative study of different simplified methods and finite element (FE) models used for calculating deflections in reinforced concrete T-beams is performed. Simplified methods employed are the one recommended by Brazilian Code ABNT NBR 6118, and the Bilinear method recommended by the fib Model Code. Two FE models are adopted, in which the material nonlinearities are considered by means of moment-curvature diagrams, these based on the simplified methods. The deflections obtained by the different models are compared for several examples of continuous RC T-beams at service, by varying the beam geometry and the reinforcing ratio. The results are analyzed for short-term deflections and long-term deflections. In the end of this work, differences between the models are pointed and recommendations regarding the use of then are drawn for this specific section type.

TAC-UNet: transformer-assisted convolutional neural network for medical image segmentation.

Medical image segmentation is crucial for improving healthcare outcomes. Convolutional neural networks (CNNs) have been widely applied in medical image analysis; however, their inherent inductive biases limit their ability to capture global contextual information. Vision transformer (ViT) architectures address this limitation by leveraging attention mechanisms to model global relationships; however, they typically require large-scale datasets for effective training, which is challenging in the field of medical imaging due to limited data availability. This study aimed to integrate the advantages of CNN and ViT architectures to improve segmentation performance on small-scale medical image datasets. In this study, we established a U-shaped network architecture based on a Transformer-assisted convolutional neural network (TAC-UNet). The TAC-UNet is primarily composed of a hybrid structure integrating CNN and Transformer components. Specifically, the hybrid architecture follows a dual-path design in which the Transformer branch continuously conveys global contextual information to the CNN backbone. This allows the CNN backbone to enhance its global perception while building on the local features it extracts, thereby improving its ability to comprehend complex image structures. A channel cross-attention (CCA) module is also incorporated as a bridge between the encoder and decoder to better reconcile the semantic discrepancies between them. Detailed experiments on three public datasets were conducted. Specifically, our model was trained on 30 images from the Multi-organ Nucleus Segmentation (MoNuSeg) training dataset, 85 images from the Gland Segmentation (GlaS) training dataset, and 551 images from the Computer Vision Center Colorectal Cancer-Clinic Database (CVC-ClinicDB) dataset. We evaluated the performance of our model on the corresponding test sets. Our TAC-UNet achieved the best Dice scores (80.36%, 90.70%, and 91.81% on the MoNuSeg, GlaS, and CVC-ClinicDB datasets, respectively) of all the models. Compared to other CNN-based, Transformer-based, and hybrid methods, the TAC-UNet demonstrated significantly superior segmentation performance. Our TAC-UNet model showed advanced segmentation performance on small-scale medical image datasets. The detailed experimental results showed the effectiveness of the method. Our model's code is available at: https://github.com/hejlhello/TAC-UNet.

Flexural tensile behaviour of alkali-activated slag-based concrete and Portland cement-based concrete incorporating single and multiple hooked-end steel fibres

In this study, three-point bending tests on notched beams according to EN 14651 have been performed to evaluate the flexural post-cracking behaviour of alkali-activated slag-based concrete (AASC) and Portland cement-based concrete (PCC) incorporating single (3D) and multiple (4D, 5D) hooked-end steel fibres in different volume fractions up to 0.75%. According to the experimental results, the post-cracking residual flexural strength increases with the increase in the fibre volume fraction for each fibre and concrete matrix type. AASC mixes incorporating 3D and 4D fibres show higher values of residual flexural strength for the same crack opening than PCC mixes with the same fibre type and dosage. Only for the mixes incorporating 5D fibres, steel fibre-reinforced PCC (SFRPCC) mixes outperform steel fibre-reinforced AASC (SFRAASC) mixes in terms of post-cracking behaviour. According to EN 14651, the values of the residual strengths and , corresponding to a crack mouth opening displacement (CMOD) of 0.5 mm and 2.5 mm, respectively, and their corresponding characteristic values and , respectively, can be derived from the experimental load-CMOD curves. Following the fib Model Code 2020, each mix can then be classified according to the values of and the ratio. As a result, empirical models have been developed for SFRPCC to predict the values of and and the applicability of such models to SFRAASC is evaluated in this study. Once the values of and are known, tensile constitutive models can be derived according to the fib Model Code 2020 and used as input parameters for finite element modelling. In this study, the accuracy of the code-based constitutive model to predict the flexural behaviour of SFRAASC and SFRPCC is evaluated using the concrete damage plasticity (CDP) model available in ABAQUS. The numerical model based on the tensile stress-strain curve in the fib Model Code 2020 can qualitatively capture the post-cracking behaviour of SFRAASC and SFRPCC incorporating 3D, 4D and 5D at 0.25% fibre volume fraction, despite overestimating their tensile strength. For higher fibre volume fractions, the CDP model, in conjunction with the mode I parameters derived from the fib Model Code 2020, is unable to adequately describe the post-hardening behaviour exhibited by the composites.