Hasil untuk "Machine design and drawing"

F. Longo, A. Padovano, Steven Umbrello

Although manufacturing companies are currently situated at a transition point in what has been called Industry 4.0, a new revolutionary wave—Industry 5.0—is emerging as an ‘Age of Augmentation’ when the human and machine reconcile and work in perfect symbiosis with one another. Recent years have indeed assisted in drawing attention to the human-centric design of Cyber-Physical Production Systems (CPPS) and to the genesis of the ‘Operator 4.0’, two novel concepts that raise significant ethical questions regarding the impact of technology on workers and society at large. This paper argues that a value-oriented and ethical technology engineering in Industry 5.0 is an urgent and sensitive topic as demonstrated by a survey administered to industry leaders from different companies. The Value Sensitive Design (VSD) approach is proposed as a principled framework to illustrate how technologies enabling human–machine symbiosis in the Factory of the Future can be designed to embody elicited human values and to illustrate actionable steps that engineers and designers can take in their design projects. Use cases based on real solutions and prototypes discuss how a design-for-values approach aids in the investigation and mitigation of ethical issues emerging from the implementation of technological solutions and, hence, support the migration to a symbiotic Factory of the Future.

Mohammed Shabbir Ali, Alexis Warsemann, Pierre Merdrignac et al.

Ensuring the safety of Autonomous Driving Systems (ADS) at urban intersections remains challenging due to complex interactions between vehicles and traffic management infrastructure. This study validates an ADS equipped with connected perception using Infrastructure-to-Vehicle (I2V) communication within a combined virtual and hybrid testing approach. The validation follows the overall structure and methodology of the SUNRISE Safety Assurance Framework (SAF), which is applied in detail where required by the scope of the study. Five representative urban intersection scenarios, covering both nominal driving conditions and safety-critical edge cases, are evaluated using virtual simulations in MATLAB/Simulink (2014b) and hybrid experiments integrating OMNeT++ (5.7.1)/Veins (5.2)/SUMO (1.12.0) with real-world components. Key Performance Indicators (KPIs) related to safety, decision-making, longitudinal control, passenger comfort, and V2X communication performance are analyzed. The results show strong consistency between virtual and hybrid testing, with ego vehicle speed deviations below 2 km/h and trigger distance differences under 3 m. V2X communication achieves a near-perfect Cooperative Awareness Message (CAM) delivery ratio, with an average latency of approximately 142 ms. While this latency remains within the tolerance of the deployed ADS, the overall end-to-end delay highlights opportunities for further optimization. The study demonstrates how the SUNRISE SAF can effectively structure ADS validation, identifies critical scenarios such as right-of-way violations by non-priority obstacles, and provides insights into improving connectivity handling and low-speed braking behavior for Cooperative, Connected, and Automated Mobility (CCAM) systems in urban environments.

Iris A. M. Huijben, W. Kool, Max B. Paulus et al.

The Gumbel-max trick is a method to draw a sample from a categorical distribution, given by its unnormalized (log-)probabilities. Over the past years, the machine learning community has proposed several extensions of this trick to facilitate, e.g., drawing multiple samples, sampling from structured domains, or gradient estimation for error backpropagation in neural network optimization. The goal of this survey article is to present background about the Gumbel-max trick, and to provide a structured overview of its extensions to ease algorithm selection. Moreover, it presents a comprehensive outline of (machine learning) literature in which Gumbel-based algorithms have been leveraged, reviews commonly-made design choices, and sketches a future perspective.

Ramiz Salama, Fadi Al-Turjman

Soulaimane Idiri, Mohammed Said Boukhryss, Abdellah Azmani et al.

This paper details the development of an embedded system for vehicle data acquisition using the On-Board Diagnostics version 2 (OBD2) protocol, with the objective of predicting power loss caused by exhaust gas backpressure (EBP). The system decodes and preprocesses vehicle data for subsequent analysis using predictive artificial intelligence algorithms. MATLAB’s 2023b Powertrain Blockset, along with the pre-built “Compression Ignition Dynamometer Reference Application (CIDynoRefApp)” model, was used to simulate engine behavior and its subsystems. This model facilitated the control of various engine subsystems and enabled simulation of dynamic environmental factors, including wind. Manipulation of the exhaust backpressure orifice revealed a consistent correlation between backpressure and power loss, consistent with theoretical expectations and prior research. For predictive analysis, two deep learning models—Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU)—were applied to the generated sensor data. The models were evaluated based on their ability to predict engine states, focusing on prediction accuracy and performance. The results showed that GRU achieved lower Mean Absolute Error (MAE) and Mean Squared Error (MSE), making GRU the more effective model for power loss prediction in automotive applications. These findings highlight the potential of using synthetic data and deep learning techniques to improve predictive maintenance in the automotive industry.

Hiromu Mori, Takayuki Tanaka, Akihiko Murai et al.

Abstract This paper presents a simple mechanical model capable of evaluating a running effectiveness index $$\epsilon _{EI}$$ ϵ EI based on mechanical energy. We extended a spring-loaded inverted pendulum model, considering the biomechanical determinants of running economy (RE), to develop a simplified mechanical model that can accurately represent the $$\epsilon _{EI}$$ ϵ EI calculated by a detailed running model. To assess the accuracy of the proposed model in estimating RE, we computed the $$\epsilon _{EI}$$ ϵ EI using both the proposed and detailed models, based on data obtained from running experiments. A linear regression analysis using the least squares method was performed to analyze the relationship between the $$\epsilon _{EI}$$ ϵ EI values calculated by the two models. The regression analysis results of the $$\epsilon _{EI}$$ ϵ EI values obtained from the two models yielded significant F-statistics ( $$p < 0.01$$ p < 0.01 ) for all four participants, demonstrating that the proposed model can sufficiently represent the running economy index calculated by the detailed model.

K. Feng, Xander Koo, Lawrence Tan et al.

Social media feeds are deeply personal spaces that reflect individual values and preferences. However, top-down, platform-wide content algorithms can reduce users’ sense of agency and fail to account for nuanced experiences and values. Drawing on the paradigm of interactive machine teaching (IMT), an interaction framework for non-expert algorithmic adaptation, we map out a design space for teachable social media feed experiences to empower agential, personalized feed curation. To do so, we conducted a think-aloud study (N = 24) featuring four social media platforms—Instagram, Mastodon, TikTok, and Twitter—to understand key signals users leveraged to determine the value of a post in their feed. We synthesized users’ signals into taxonomies that, when combined with user interviews, inform five design principles that extend IMT into the social media setting. We finally embodied our principles into three feed designs that we present as sensitizing concepts for teachable feed experiences moving forward.

Ludi Wang, Xueqing Chen, Yi Du et al.

The field of catalysis holds paramount importance in shaping the trajectory of sustainable development, prompting intensive research efforts to leverage artificial intelligence (AI) in catalyst design. Presently, the fine-tuning of open-source large language models (LLMs) has yielded significant breakthroughs across various domains such as biology and healthcare. Drawing inspiration from these advancements, we introduce CataLM (Catalytic Language Model), a large language model tailored to the domain of electrocatalytic materials. Our findings demonstrate that CataLM exhibits remarkable potential for facilitating human-AI collaboration in catalyst knowledge exploration and design. To the best of our knowledge, CataLM stands as the pioneering LLM dedicated to the catalyst domain, offering novel avenues for catalyst discovery and development.

Manuela Vieira, Gonçalo Galvão, Manuel A. Vieira et al.

This study combines Visible Light Communication (VLC) and Artificial Intelligence (AI) to enhance traffic signal control, reduce congestion, and improve safety, through real-time monitoring and dynamic traffic management. Leveraging VLC technology, the system uses existing road infrastructure to transmit live data on vehicle and pedestrian positions, speeds, and queues. AI agents, employing Deep Reinforcement Learning (DRL), process this data to manage traffic flows dynamically, applying anti-bottleneck and rerouting techniques to balance pedestrian and vehicle waiting times. A centralized global agent coordinates the local agents controlling each intersection, enabling indirect communication and data sharing to train a unified DRL model. This model makes real-time adjustments to traffic light phases, utilizing a queue/request/response system for adaptive intersection management. Tested using simulations and real-world trials involving standard and rerouting scenarios, the approach demonstrates significantly better performance in regard to the rerouting configuration, reducing congestion and enhancing traffic flow and pedestrian safety. Scalable and adaptable to various intersection types, including four-way, T-intersections, and roundabouts, the system’s efficacy is validated using the SUMO urban mobility simulator, resulting in notable reductions to travel and waiting times for both vehicles and pedestrians.

Charles Patrick Bounds, Shishir Desai, Mesbah Uddin

Computational fluid dynamic (CFD) models and workflows are often developed in an ad hoc manner, leading to a limited understanding of interaction effects and model behavior under various conditions. Machine learning (ML) and explainability tools can help CFD process development by providing a means to investigate the interactions in CFD models and pipelines. ML tools in CFD can facilitate the efficient development of new processes, the optimization of current models, and enhance the understanding of existing CFD methods. In this study, the turbulent closure coefficient tuning of the SST <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>k</mi><mo>−</mo><mi>ω</mi></mrow></semantics></math></inline-formula> Reynolds-averaged Navier–Stokes (RANS) turbulence model was selected as a case study. The objective was to demonstrate the efficacy of ML and explainability tools in enhancing CFD applications, particularly focusing on external aerodynamic workflows. Two variants of the Ahmed body model, with 25-degree and 40-degree slant angles, were chosen due to their availability and relevance as standard geometries for aerodynamic process validation. Shapley values, a concept derived from game theory, were used to elucidate the impact of varying the values of the closure coefficients on CFD predictions, chosen for their robustness in providing clear and interpretable insights into model behavior. Various ML algorithms, along with the SHAP method, were employed to efficiently explain the relationships between the closure coefficients and the flow profiles sampled around the models. The results indicated that model coefficient <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>β</mi><mo>*</mo></msup></semantics></math></inline-formula> had the greatest overall effect on the lift and drag predictions. The ML explainer model and the generated explanations were used to create optimized closure coefficients, achieving an optimal set that reduced the error in lift and drag predictions to less than 7% and 0.5% for the 25-degree and 40-degree models, respectively.

Andrea Stefanelli, Marco Aprea, Fabio Carbone et al.

The accurate prediction of tire and brake thermomechanical behavior is crucial for various applications in the automotive industry, including vehicle dynamics analysis, racing performance optimization, and driver assistance system development. The temperature of the brakes plays a crucial role in determining the performance of rubber by altering its temperature. This change impacts the rim and the air within the tire, leading to variations in temperature and tire pressure, which consequently affect the vehicle’s overall performance. Traditionally, these components have been modeled separately, neglecting the crucial thermal interaction between them, thereby losing a lot of important information from the outside that influences the tire. This paper presents a novel method that overcomes this limitation by coupling the thermomechanical models of the tire and brake, enabling a more comprehensive understanding of their combined behavior. Therefore, the present work could be an interesting starting point to understand how a control system can be influenced by the thermodynamic of the wheel–brake system.

Olga Cherednichenko, Dmytro Chernyshov, Dmytro Sytnikov et al.

This research paper delves into the innovative integration of Shannon entropy and rough set theory, presenting a novel approach to generalize the evaluation approach in machine learning. The conventional application of entropy, primarily focused on information uncertainty, is extended through its combination with rough set theory to offer a deeper insight into data's intrinsic structure and the interpretability of machine learning models. We introduce a comprehensive framework that synergizes the granularity of rough set theory with the uncertainty quantification of Shannon entropy, applied across a spectrum of machine learning algorithms. Our methodology is rigorously tested on various datasets, showcasing its capability to not only assess predictive performance but also to illuminate the underlying data complexity and model robustness. The results underscore the utility of this integrated approach in enhancing the evaluation landscape of machine learning, offering a multi-faceted perspective that balances accuracy with a profound understanding of data attributes and model dynamics. This paper contributes a groundbreaking perspective to machine learning evaluation, proposing a method that encapsulates a holistic view of model performance, thereby facilitating more informed decision-making in model selection and application.

E. Capobianco

Rushil Samir Patel, Harshal D. Akolekar

Biomimicry involves drawing inspiration from nature’s designs to create efficient systems. For instance, the unique herringbone riblet pattern found in bird feathers has proven effective in minimizing drag. While attempts have been made to replicate this pattern on structures like plates and aerofoils, there has been a lack of comprehensive optimization of their overall design and of their constituent individual repeating structures. This study attempts to enhance the performance of individual components within the herringbone riblet pattern by leveraging computational fluid dynamics (CFD) and supervised machine learning to reduce drag. The paper outlines a systematic process involving the creation of 107 designs, parameterization, feature selection, generating targets using CFD simulations, and employing regression algorithms. From CFD calculations, the drag coefficients (C d ) for these designs are found, which serve as an input to train supervised learning models. Using the trained transformed target regressor model as a substitute to CFD, C d values for 10,000 more randomly generated herringbone riblet designs are predicted. The design with the lowest predicted C d is the optimized design. Notably, the regressed model exhibited an average prediction error rate of 6% on the testing data. The prediction of C d for the optimized design demonstrated an error of 4% compared to its actual C d value calculated through CFD. The study also delves into the mechanics of drag reduction in herringbone riblet structures. The resulting optimized microstructure design holds the potential for reducing drag in various applications such as aerospace, automotive, and marine crafts by integrating it onto their surfaces. This innovative approach could significantly transform drag reduction and open pathways to more efficient transportation systems.

A. Madsen, A. Munk, Johan Irving Søltoft

This paper provides a theoretical alternative to the prevailing perception of machine learning as synonymous with speed and efficiency. Inspired by ethnographic fieldwork and grounded in pragmatist philosophy, we introduce the concept of “data friction” as the situation when encounters between held beliefs and data patterns possess the potential to stimulate innovative thinking. Contrary to the conventional connotations of “speed” and “control,” we argue that computational methods can generate a productive dissonance, thereby fostering slower and more reflective practices within organizations. Drawing on a decade of experience in participatory data design and data sprints, we present a typology of data frictions and outline three ways in which algorithmic techniques within data science can be reimagined as “friction machines”. We illustrate these theoretical points through a dive into three case studies conducted with applied anthropologist in the movie industry, urban planning, and research.

Ivan Puhachov, Cedric Martens, P. Kry et al.

We propose a method of reconstructing 3D machine-made shapes from bitmap sketches by separating an input image into individual patches and jointly optimizing their geometry. We rely on two main observations: (1) human observers interpret sketches of man-made shapes as a collection of simple geometric primitives, and (2) sketch strokes often indicate occlusion contours or sharp ridges between those primitives. Using these main observations we design a system that takes a single bitmap image of a shape, estimates image depth and segmentation into primitives with neural networks, then fits primitives to the predicted depth while determining occlusion contours and aligning intersections with the input drawing via optimization. Unlike previous work, our approach does not require additional input, annotation, or templates, and does not require retraining for a new category of man-made shapes. Our method produces triangular meshes that display sharp geometric features and are suitable for downstream applications, such as editing, rendering, and shading.

Yiwei Ma, Y. Tian, Xianping Liu

Tutak Magdalena, Brodny Jarosław, Małkowski Piotr et al.

The process of underground mining is one of the most complex and hazardous activities. In order to maintain the continuity and efficiency of this process, it is necessary to take measures to reduce this hazard. The paper addresses this issue by presenting a developed methodology for using model studies and numerical simulations to support the process of monitoring methane hazards. Its basis is the developed model of the region of underground mining exploitation along with the ventilation phenomena occurring in it. To develop it, the ANSYS Fluent program was used, based on the finite volume method classified as computational fluid mechanics. The model reflects both the geometries and physical and chemical phenomena occurring in the studied area, as well as the auxiliary ventilation equipment used during operation. The research was conducted for two variants of methane emissions from goaf zones, the first of which concerned the actual state of the mining area, and the second of which concerned increased methane emissions from these goaf zones. The purpose of the study was to determine the distribution of methane concentrations in the most dangerous part of the studied area, which is the intersection of the longwall and the tailgate, as well as the distribution of ventilation air flow velocities affecting them. The studies for both variants made it possible to determine places particularly exposed to the occurrence of dangerous concentrations of methane in this region. The methodology developed represent a new approach to studying the impact of methane emissions from goaf zones into mine workings.

Kasemset Chompoonoot, Opassuwan Takron, Tangsittikhun Thanakit et al.

This study aims to design and improve the plant layout of a ceramic factory by adopting Systematic Layout Planning (SLP) and the simulation technique. A ceramic company in northern Thailand is selected as a case study. Three ceramic products including roof tiles, wall tiles and dishware are studied due to their highest production volume. Through the SLP approach, information regarding the number of departments and machines, the area of the plant, the frequency of movement and the distance between each department is collected for the analysis of the relationship between departments. Two plant layout designs are then proposed; the first one is derived from the Computerized Relationship Layout Planning algorithm (CORELAP), and the second one is the process layout. For selecting the most appropriate layout design, five criteria are considered including total distance, the average total process time of each unit produced, ease of movement, material flow and safety. To determine the distance and the average total process time per unit, Distance-Based Scoring and simulation techniques are conducted while the ease of movement, material flow and safety are rated based on whether the company satisfies each criterion. Employing the weight scoring technique, the results report that the CORELAP layout is the most suitable for further implementation due to its highest weighted score equal to 2.536 while the process layout receives 2.386. Implementing the CORELAP layout can reduce the total distance by 16.76% while the average total process time per unit of the CORELAP layout is not significantly different at the significance level of 0.05 as compared to the existing layout.

Eric Töpel, Alexander Fuchs, Kay Büttner et al.

In this work, a method is developed for the component design of chassis bushings with contoured inner cores, aided by artificial neural networks (ANNs) and design optimization. First, a model of a physical chassis bushing is generated using the finite element method (FEM). To determine the material parameters of the material model, a material parameter optimization is conducted. Based on the bushing model, different samples for a design study are generated using the design of experiments method. Due to invalid areas of the geometrical model definitions, constraints are established and the design parameter space is cleaned up. From the cleaned design parameter space, a database of several design parameter samples and three associated quasi-static stiffnesses, calculated with FEM simulations, is generated. The database is subsequently used for the training and hyper-parameter optimization of the ANN. Subsequently, the feed-forward ANN is employed in a design study, where stiffnesses are prescribed and design parameters identified. The design process is inverted with the help of a constrained design parameter optimization (DO), based on particle swarm optimization (PSO). Two usecases are defined for the evaluation of the design accuracy of the entire method. The design parameters found are validated by corresponding FEM simulations.