Search

Abstract Background Prosthetic legs help individuals with an amputation regain locomotion. Recently, deep neural network (DNN)-based control methods, which take advantage of the end-to-end learning capability of the network, have been proposed. One prominent challenge for these learning-based approaches is obtaining data for the training, particularly for the training of a mid-level controller. In this study, we propose a method for generating synthetic gait patterns (vertical load and lower limb joint angles) using a generative adversarial network (GAN). This approach enables a mid-level controller to execute ambulation modes that are not included in the training datasets. Methods The conditional GAN is trained on benchmark datasets that contain the gait data of individuals without amputation; synthetic gait patterns are generated from the user input. Further, a DNN-based controller for the generation of impedance parameters is trained using the synthetic gait pattern and the corresponding synthetic stiffness and damping coefficients. Results The trained GAN generated synthetic gait patterns with a coefficient of determination of 0.97 and a structural similarity index of 0.94 relative to benchmark data that were not included in the training datasets. We trained a DNN-based controller using the GAN-generated synthetic gait patterns for level-ground walking, standing-to-sitting motion, and sitting-to-standing motion. Four individuals without amputation participated in bypass testing and demonstrated the ambulation modes. The model successfully generated control parameters for the knee and ankle based on thigh angle and vertical load. Conclusions This study demonstrates that synthetic gait patterns can be used to train DNN models for impedance control. We believe a conditional GAN trained on benchmark datasets can provide reliable gait data for ambulation modes that are not included in its training datasets. Thus, designing gait data using a conditional GAN could facilitate the efficient and effective training of controllers for prosthetic legs.

This paper delves into the pivotal role of water electrolysis (WE) in green hydrogen production, a process utilizing renewable energy sources through electrolysis. The term "green hydrogen" signifies its distinction from conventional "grey" or "brown" hydrogen produced from fossil fuels, emphasizing the importance of decarbonization in the hydrogen value chain. WE becomes a linchpin, balancing surplus green energy, stabilizing the grid, and addressing challenges in hard-to-abate sectors like long-haul transport and heavy industries. This paper navigates through electrolysis variants, technological challenges, and the crucial association between electrolytic hydrogen production and renewable energy sources (RESs). Energy consumption aspects are scrutinized, highlighting the need for optimization strategies to enhance efficiency. This paper systematically addresses electrolysis fundamentals, technologies, scaling issues, and the nexus with energy sources. It emphasizes the transformative potential of electrolytic hydrogen in the broader energy landscape, underscoring its role in shaping a sustainable future. Through a systematic analysis, this study bridges the gap between detailed technological insights and the larger energy system context, offering a holistic perspective. This paper concludes by summarizing key findings, showcasing the prospects, challenges, and opportunities associated with hydrogen production via water electrolysis for the energy transition. [ABSTRACT FROM AUTHOR]

Background: Prosthetic legs help individuals with an amputation regain locomotion. Recently, deep neural network (DNN)-based control methods, which take advantage of the end-to-end learning capability of the network, have been proposed. One prominent challenge for these learning-based approaches is obtaining data for the training, particularly for the training of a mid-level controller. In this study, we propose a method for generating synthetic gait patterns (vertical load and lower limb joint angles) using a generative adversarial network (GAN). This approach enables a mid-level controller to execute ambulation modes that are not included in the training datasets. Methods: The conditional GAN is trained on benchmark datasets that contain the gait data of individuals without amputation; synthetic gait patterns are generated from the user input. Further, a DNN-based controller for the generation of impedance parameters is trained using the synthetic gait pattern and the corresponding synthetic stiffness and damping coefficients. Results: The trained GAN generated synthetic gait patterns with a coefficient of determination of 0.97 and a structural similarity index of 0.94 relative to benchmark data that were not included in the training datasets. We trained a DNN-based controller using the GAN-generated synthetic gait patterns for level-ground walking, standing-to-sitting motion, and sitting-to-standing motion. Four individuals without amputation participated in bypass testing and demonstrated the ambulation modes. The model successfully generated control parameters for the knee and ankle based on thigh angle and vertical load. Conclusions: This study demonstrates that synthetic gait patterns can be used to train DNN models for impedance control. We believe a conditional GAN trained on benchmark datasets can provide reliable gait data for ambulation modes that are not included in its training datasets. Thus, designing gait data using a conditional GAN could facilitate the efficient and effective training of controllers for prosthetic legs. [ABSTRACT FROM AUTHOR]

Powered lower-limb assistive devices, such as prostheses and exoskeletons, are a promising option for helping mobility-impaired individuals regain functional gait. Gait phase prediction plays an important role in controlling these devices and evaluating whether the device generates a gait similar to that of individuals with intact limbs. This study proposes a gait phase prediction method based on a deep neural network (DNN). The long short-term memory (LSTM)-based model predicts a continuous gait phase from the 250 ms history of the vertical load, thigh angle, knee angle, and ankle angle, commonly available on powered lower-limb assistive devices. One unified model was trained using publicly available benchmark datasets containing intact limb gaits for level-ground walking (LGW) and ascending stairs (SA). A phase prediction error of 1.28% for all benchmark datasets was obtained. The model was subsequently applied to a state machine-controlled powered prosthetic leg dataset collected fromfour individuals with unilateral transfemoral amputation. The gait phase prediction results (a phase prediction error of 5.70%) indicate that the model trained on benchmark data can be used for a system not included in the training dataset with no post-processing, such as model adaptation. Furthermore, it provided information regarding evaluation of the controller: whether the prosthetic leg generated normal gait. In conclusion, the proposed gait phase predictionmodel will facilitate efficient gait prediction and evaluation of controllers for powered lower-limb assistive devices. [ABSTRACT FROM AUTHOR]

Powered lower-limb assistive devices, such as prostheses and exoskeletons, are a promising option for helping mobility-impaired individuals regain functional gait. Gait phase prediction plays an important role in controlling these devices and evaluating whether the device generates a gait similar to that of individuals with intact limbs. This study proposes a gait phase prediction method based on a deep neural network (DNN). The long short-term memory (LSTM)-based model predicts a continuous gait phase from the 250 ms history of the vertical load, thigh angle, knee angle, and ankle angle, commonly available on powered lower-limb assistive devices. One unified model was trained using publicly available benchmark datasets containing intact limb gaits for level-ground walking (LGW) and ascending stairs (SA). A phase prediction error of 1.28% for all benchmark datasets was obtained. The model was subsequently applied to a state machine-controlled powered prosthetic leg dataset collected fromfour individuals with unilateral transfemoral amputation. The gait phase prediction results (a phase prediction error of 5.70%) indicate that the model trained on benchmark data can be used for a system not included in the training dataset with no post-processing, such as model adaptation. Furthermore, it provided information regarding evaluation of the controller: whether the prosthetic leg generated normal gait. In conclusion, the proposed gait phase predictionmodel will facilitate effcient gait prediction and evaluation of controllers for powered lower-limb assistive devices. [ABSTRACT FROM AUTHOR]

The evaluation of clustering algorithms can involve running them on a variety of benchmark problems, and comparing their outputs to the reference, ground-truth groupings provided by experts. Unfortunately, many research papers and graduate theses consider only a small number of datasets. Also, the fact that there can be many equally valid ways to cluster a given problem set is rarely taken into account. In order to overcome these limitations, we have developed a framework whose aim is to introduce a consistent methodology for testing clustering algorithms. Furthermore, we have aggregated, polished, and standardised many clustering benchmark dataset collections referred to across the machine learning and data mining literature, and included new datasets of different dimensionalities, sizes, and cluster types. An interactive datasets explorer, the documentation of the Python API, a description of the ways to interact with the framework from other programming languages such as R or MATLAB, and other details are all provided at https://clustering-benchmarks.gagolewski.com.

The field of remote sensing has undergone a remarkable shift where vast amounts of imagery are now readily available to researchers. New technologies, such as uncrewed aircraft systems, make it possible for anyone with a moderate budget to gather their own remotely sensed data, and methodological innovations have added flexibility for processing and analyzing data. These changes create both the opportunity and need to reproduce, replicate, and compare remote sensing methods and results across spatial contexts, measurement systems, and computational infrastructures. Reproducing and replicating research is key to understanding the credibility of studies and extending recent advances into new discoveries. However, reproducibility and replicability (R&R) remain issues in remote sensing because many studies cannot be independently recreated and validated. Enhancing the R&R of remote sensing research will require significant time and effort by the research community. However, making remote sensing research reproducible and replicable does not need to be a burden. In this paper, we discuss R&R in the context of remote sensing and link the recent changes in the field to key barriers hindering R&R while discussing how researchers can overcome those barriers. We argue for the development of two research streams in the field: (1) the coordinated execution of organized sequences of forward-looking replications, and (2) the introduction of benchmark datasets that can be used to test the replicability of results and methods. [ABSTRACT FROM AUTHOR]

Data analysis is a critical part of quantitative proteomics studies in interpreting biological questions. Numerous computational tools for protein quantification, imputation and differential expression (DE) analysis were generated in the past decade and the search for optimal tools is still going on. Moreover, due to the rapid development of RNA sequencing (RNA-seq) technology, a vast number of DE analysis methods were created for that purpose. The applicability of these newly developed RNA-seq-oriented tools to proteomics data remains in doubt. In order to benchmark these analysis methods, a proteomics dataset consisting of proteins derived from humans, yeast and drosophila, in defined ratios, was generated in this study. Based on this dataset, DE analysis tools, including microarray- and RNA-seq-based ones, imputation algorithms and protein quantification methods were compared and benchmarked. Furthermore, applying these approaches to two public datasets showed that RNA-seq-based DE tools achieved higher accuracy (ACC) in identifying DEPs. This study provides useful guidelines for analyzing quantitative proteomics datasets. All the methods used in this study were integrated into the Perseus software, version 2.0.3.0, which is available at https://www.maxquant.org/perseus. [ABSTRACT FROM AUTHOR]

The validation process is done to ensure the performance of nuclear codes compared to experimental results and to determine the degree of reliability of codes output. The development of nuclear power plants (NPPs) designs and the need for safer power plants has led to the increasing diversity/extension of nuclear codes and, as a result, the validation process has become more complicated. In this study, single physics validation including neutronic, core thermal-hydraulics (CTH), system thermal-hydraulics (STH), fuel performance, and multi-physics validation especially coupled neutronic and thermal-hydraulics (CNTH) are discussed. An international collective effort to provide benchmark data and test facilities including core test facility (CTF), separate effect test facility (SETF), and integral test facility (ITF) accompanied with addressing challenges related to scaling issues has been made. However, validation processes may suffer from some challenges. Non-quantifiable parameters cannot be given by experimental measurements. Moreover, for measurable parameters, faulty sensors can lead to incorrect results. Neural network, especially generative deep learning, with the ability to learn the probability distribution of training data and to generate unknown patterns may be a good candidate for the challenges mentioned above, e.g., detecting faulty sensors values and estimating non-quantifiable parameters based on quantifiable ones. Therefore, the application of artificial intelligence (AI) in the validation process may lead to a reduction of the volume of experimental measurements /construction of test facilities which alone is valuable. [ABSTRACT FROM AUTHOR]

The present paper gives an overview of liquid metal research activities performed in recent years at the Karlsruhe Institute of Technology (KIT). The work is motivated by applications in liquid metal blankets for a DEMO fusion reactor where lead lithium (PbLi), which serves as a neutron multiplier and tritium breeder, interacts with the plasma-confining magnetic field as it flows in the blanket covering the inner walls of the reactor. Liquid metal magnetohydrodynamic (MHD) research at KIT supports blanket design activities through theoretical and experimental investigations. Predictive computational tools are developed and validated by empirical data obtained for fundamental problems, such as flows in non-uniform magnetic fields or magneto-convective heat transfer from submerged obstacles. In addition, technological developments like pressure drop reduction by insulating flow channel inserts are pursued both theoretically and experimentally. Two complementary experimental facilities (MEKKA and MaPLE) provide a unique and versatile platform for MHD investigations at fusion relevant parameters. Using NaK as a model fluid in MEKKA allows experiments to be conducted at high Hartmann numbers in large complex geometries, such as scaled blanket mock-ups of ITER test blanket modules. For magneto-convection and heat transfer studies, MaPLE is well suited since it enables experiments with the prototypical fluid PbLi in test sections inclined at various orientations with respect to gravity. Some recent results have been selected to illustrate the broad spectrum of MHD activities at KIT. [ABSTRACT FROM AUTHOR]

The real resistance that a ship must face when it is navigating in ice floes is the key factor for knowing the necessary power and the required engine size. The aim of this work is to provide valuable data to help other research in which numerical frameworks will be developed to study ship navigation in broken ice. In this work, we used paraffin wax as an alternative to obtain affordable solutions, avoiding the high cost of ice tests. The experiments were carried out in a traditional basin facility and they consisted of towing tank tests with a ship model using different concentrations of blocks simulated by the use of paraffin wax. Photogrammetry was used as technique to determine the initial position of the ice blocks, which is important as starting data in the current development of numerical simulation code for studying the features of ship resistance in drift ice. These data are available for some ice concentrations in attached files. In addition, a procedure for testing in traditional towing facilities is presented and discussed. The results of the resistance obtained in the experiments in the presence of simulated floes are presented for three concentrations and three model speeds. Some findings may be applicable to ice sailing, under given circumstances.

Data-independent acquisition (DIA) mass spectrometry (MS) has emerged as a powerful technology for high-throughput, accurate, and reproducible quantitative proteomics. This review provides a comprehensive overview of recent advances in both the experimental and computational methods for DIA proteomics, from data acquisition schemes to analysis strategies and software tools. DIA acquisition schemes are categorized based on the design of precursor isolation windows, highlighting wide-window, overlapping-window, narrow-window, scanning quadrupole-based, and parallel accumulation-serial fragmentation-enhanced DIA methods. For DIA data analysis, major strategies are classified into spectrum reconstruction, sequence-based search, library-based search, de novo sequencing, and sequencing-independent approaches. A wide array of software tools implementing these strategies are reviewed, with details on their overall workflows and scoring approaches at different steps. The generation and optimization of spectral libraries, which are critical resources for DIA analysis, are also discussed. Publicly available benchmark datasets covering global proteomics and phosphoproteomics are summarized to facilitate performance evaluation of various software tools and analysis workflows. Continued advances and synergistic developments of versatile components in DIA workflows are expected to further enhance the power of DIA-based proteomics., Competing Interests: Conflict of interest The authors declare no competing interests., (Copyright © 2024 The Authors. Published by Elsevier Inc. All rights reserved.)

Recently, there has been a growing interest in applying machine learning methods to problems in engineering mechanics. In particular, there has been significant interest in applying deep learning techniques to predicting the mechanical behavior of heterogeneous materials and structures. Researchers have shown that deep learning methods are able to effectively predict mechanical behavior with low error for systems ranging from engineered composites, to geometrically complex metamaterials, to heterogeneous biological tissue. However, there has been comparatively little attention paid to deep learning model calibration, i.e., the match between predicted probabilities of outcomes and the true probabilities of outcomes. In this work, we perform a comprehensive investigation into machine learning model calibration across 7 open access engineering mechanics datasets that cover three distinct types of mechanical problems. Specifically, we evaluate both model and model calibration error for multiple machine learning methods, and investigate the influence of ensemble averaging and post hoc model calibration via temperature scaling. Overall, we find that ensemble averaging of deep neural networks is both an effective and consistent tool for improving model calibration, while temperature scaling has comparatively limited benefits. Looking forward, we anticipate that this investigation will lay the foundation for future work in developing mechanics specific approaches to deep learning model calibration. • We investigate machine learning model calibration for problems in mechanics. • This investigation includes 7 open access engineering mechanics datasets. • We investigate ensemble averaging and post hoc model calibration. • Ensemble averaging of deep neural networks improves model calibration. • Temperature scaling has comparatively limited benefits. [ABSTRACT FROM AUTHOR]

• Graph Neural Networks are an effective tool for metamodels of complex structures. • Point cloud-based geometry representation outperformed tested alternatives. • Ensemble learning enhanced metamodel performance. • Performance is benchmarked with open source Asymmetric Buckling Columns dataset. From designing architected materials to connecting mechanical behavior across scales, computational modeling is a critical tool for understanding and predicting the mechanical response of deformable bodies. In particular, computational modeling is an invaluable tool for predicting global emergent phenomena, such as the onset of geometric instabilities, or heterogeneity induced symmetry breaking. Recently, there has been a growing interest in both using machine learning based computational models to learn mechanical behavior directly from experimental data, and using machine learning (ML) methods to reduce the computational cost of physics-based simulations. Notably, machine learning approaches that rely on Graph Neural Networks (GNNs) have recently been shown to effectively predict mechanical behavior in multiple examples of particle-based and mesh-based simulations. However, despite this initial promise, the performance of graph based methods have yet to be investigated on a myriad of solid mechanics problems. In this work, we examine the ability of neural message passing to predict a fundamental aspect of mechanically driven emergent behavior: the connection between a column's geometric structure and the direction that it buckles. To accomplish this, we introduce the Asymmetric Buckling Columns (ABC) dataset, a dataset comprised of three types of asymmetric and heterogeneous column geometries (sub-dataset 1, sub-dataset 2, and sub-dataset 3) where the goal is to classify the direction of symmetry breaking (left or right) under compression after the onset of the buckling instability. Notably, it is difficult to parameterize these structures into a feature vector for typical ML methods. Essentially, because the geometry of these columns is discontinuous and intricate, local geometric patterns will be distorted by the low-resolution "image-like" data representations that are required to implement convolutional neural network based metamodels. Instead, we present a pipeline to learn global emergent properties while enforcing locality with message passing neural networks. Specifically, we take inspiration from point cloud based classification problems from the computer vision research field and use PointNet++ layers to perform classification on the ABC dataset. In addition to investigating GNN model architecture, we study the effect of different input data representation approaches, data augmentation, and combining multiple models as an ensemble. Overall, we were able to achieve good performance with this approach, ranging from 0.952 prediction accuracy on sub-dataset 1, to 0.913 prediction accuracy on sub-dataset 2, to 0.856 prediction accuracy on sub-dataset 3 for training dataset sizes of 20 , 000 points each. However, these results also clearly indicate that predicting solid mechanics based emergent behavior with these methods is non-trivial. Because both our model implementation and dataset are distributed under open-source licenses, we hope that future researchers can build on our work to create enhanced mechanics-specific machine learning methods. Furthermore, we also intend to provoke discussion around different methods for representing complex mechanical structures when applying machine learning to mechanics research. [ABSTRACT FROM AUTHOR]

The real resistance that a ship must face when it is navigating in ice floes is the key factor for knowing the necessary power and the required engine size. The aim of this work is to provide valuable data to help other research in which numerical frameworks will be developed to study ship navigation in broken ice. In this work, we used paraffin wax as an alternative to obtain affordable solutions, avoiding the high cost of ice tests. The experiments were carried out in a traditional basin facility and they consisted of towing tank tests with a ship model using different concentrations of blocks simulated by the use of paraffin wax. Photogrammetry was used as technique to determine the initial position of the ice blocks, which is important as starting data in the current development of numerical simulation code for studying the features of ship resistance in drift ice. These data are available for some ice concentrations in attached files. In addition, a procedure for testing in traditional towing facilities is presented and discussed. The results of the resistance obtained in the experiments in the presence of simulated floes are presented for three concentrations and three model speeds. Some findings may be applicable to ice sailing, under given circumstances. [ABSTRACT FROM AUTHOR]