Hasil untuk "Machine design and drawing"

Elena-Daniela Lupu, Ștefan Lucian Tabacu

Environmental regulations, rapid technological advancements, and evolving mobility trends have led to a significant transformation of the automotive industry in recent years. The adoption of battery-electric vehicles (BEVs) has been accelerated by these developments, which are becoming increasingly efficient and widely deployed. Evaluating BEV energy consumption and performance is essential for optimizing energy efficiency, extending driving range, and developing effective control strategies under real-world operating conditions. The analysis is based on the WLTP Class 3 driving cycle, in which the vehicle operating points are projected onto the motor efficiency map to evaluate the influence of real-world operating conditions on overall propulsion efficiency. Two operating scenarios are considered: with regenerative braking and without regenerative braking. The inverter and battery are modeled using quasi-static energy-based representations to ensure system-level energetic consistency while maintaining computational efficiency. The results show that rotor topology significantly influences vehicle-level energy consumption. The dual-layer IPM configuration reduces net WLTP energy demand by approximately 9% and increases the estimated driving range from about 489 km to 535 km compared to the single-layer V-shaped configuration. Variations in rotor topology led to different efficiency distributions, which leads to systematic differences in battery energy demand and achievable driving range. The results highlight the importance of aligning traction motor design with realistic operating-point distributions rather than optimizing solely for peak efficiency or marginal improvements in regenerative braking performance.

Shen-Yung Lin, Yen-Ting Lai

As global manufacturing faces rising energy costs, environmental pressures, and machining precision, the development trends of the machine tools are moving towards lightweight and high-rigidity structures. While those approaches of increasing key component geometrical size or enhancing rib design do enhance rigidity performance, they also usually increase weight, which conflicts with the goals of achieving high performance and environmental sustainability. Therefore, how to achieve system lightweightness while maintaining or enhancing structural rigidity has become a key research challenge. This study adopts a biomimetic design approach, drawing inspiration from the natural growth features of biological structures. By integrating these natural structural features, the design aims to enhance rigidity while reducing weight. Static and modal analyses are conducted firstly by using FEM software to simulate the total deformation, natural frequency, and modal shape, respectively. The biomimetic designs are then performed on those subsystems in a grinding machine-tool, which exhibit larger deformation and weaker stiffness by incorporating the structural features of leaf veins, cacti, and bamboos. Single or multiple structural feature combinations are constituted during the biomimetic design processes for worktable, base, and column subsystems, and the natural frequencies and weight obtained from the numerical analysis were compared subsequently to identify the better bionic subsystems that replace the corresponding ones originally assembled in the grinding machine-tool finally. The results show that one of the first three mode natural frequencies of a better bionic worktable (leaf vein and cactus) is increased up to 7.07%, with a 1.12% weight reduction. A better bionic base (leaf vein) with corner trimming exhibits a 14.04% increase in natural frequency and a 2.04% weight reduction. Similarly, a better bionic column (bamboo) achieves a 5.58% increase in natural frequency and a 0.14% weight reduction. After these better bionic subsystems are substituted in the grinding machine-tool, one of the first three mode natural frequencies is increased up to 14.56%, the weight is reduced by 1.25%, and the maximum total deformation is decreased by 39.64%. The maximum total deformation for the headstock is reduced by 26.95% after the original grinding machine-tool is replaced by better bionic subsystems. The increases in the specific stiffness for these better bionic subsystems are also investigated in this study to illustrate the effectiveness of the biomimetic designs.

Dardzińska-Głębocka Agnieszka, Kasperczuk Anna, Gardocki Grzegorz

Occupational stress is a critical human factor that affects efficiency, safety, and the continuity of operational processes, particularly in high-risk professions such as uniformed services. The aim of this study was to develop and validate an interpretable decision support system (DSS) for the early pre-screening of employees exhibiting symptoms of post-traumatic stress disorder (PTSD). To this end, the J48 decision tree algorithm was applied to extract classification rules based on psychological symptoms defined in the DSM-5. The performance of the J48 model was compared with Support Vector Machine and Random Forest algorithms. Among the evaluated models, J48 demonstrated the highest overall effectiveness, achieving top results across all key metrics, including an F1-score of 0.976 and a ROC Area of 0.987. The generated classification rules ensure model transparency and interpretability – features essential for practical implementation in organizational occupational health procedures. The proposed tool contributes to the field of human factors engineering by offering a practical solution for managing mental health risks, ultimately supporting improved safety and operational performance in organizational settings.

Mehdi Attioui, Mohamed Lahby

Traffic congestion continues to pose a significant challenge to contemporary urban transportation systems, exerting substantial effects on economic productivity, environmental sustainability, and the overall quality of life. This systematic literature review thoroughly explores the development of traffic congestion forecasting methodologies from 2014 to 2024 by analyzing 100 peer-reviewed publications according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We examine the technological advancements from traditional machine learning (achieving 75–85% accuracy) through deep learning approaches (85–92% accuracy) to recent large language model (LLM) implementations (90–95% accuracy). Our analysis indicates that LLM-based systems exhibit superior performance in managing multimodal data integration, comprehending traffic events, and predicting non-recurrent congestion scenarios. The key findings suggest that hybrid approaches, which integrate LLMs with specialized deep learning architectures, achieve the highest prediction accuracy while addressing the traditional limitations of edge case management and transfer learning capabilities. Nonetheless, challenges remain, including higher computational demands (50–100× higher than traditional methods), domain adaptation complexity, and constraints on real-time implementation. This review offers a comprehensive taxonomy of methodologies, performance benchmarks, and practical implementation guidelines, providing researchers and practitioners with a roadmap for advancing intelligent transportation systems using next-generation AI technologies.

Ananya Saha, Mahir Afser Pavel, Md Fahim Shahoriar Titu et al.

Efficient and explainable vehicle damage inspection is essential due to the increasing complexity and volume of vehicular incidents. Traditional manual inspection approaches are not time-effective, prone to human error, and lead to inefficiencies in insurance claims and repair workflows. Existing deep learning methods, such as CNNs, often struggle with generalization, require large annotated datasets, and lack interpretability. This study presents a robust and interpretable deep learning framework for vehicle damage classification, integrating Vision Transformers (ViTs) and ensemble detection strategies. The proposed architecture employs a RetinaNet backbone with a ViT-enhanced detection head, implemented in PyTorch using the Detectron2 object detection technique. It is pretrained on COCO weights and fine-tuned through focal loss and aggressive augmentation techniques to improve generalization under real-world damage variability. The proposed system applies the Weighted Box Fusion (WBF) ensemble strategy to refine detection outputs from multiple models, offering improved spatial precision. To ensure interpretability and transparency, we adopt numerous explainability techniques—Grad-CAM, Grad-CAM++, and SHAP—offering semantic and visual insights into model decisions. A custom vehicle damage dataset with 4500 images has been built, consisting of approximately 60% curated images collected through targeted web scraping and crawling covering various damage types (such as bumper dents, panel scratches, and frontal impacts), along with 40% COCO dataset images to support model generalization. Comparative evaluations show that Hybrid ViT-RetinaNet achieves superior performance with an F1-score of 84.6%, mAP of 87.2%, and 22 FPS inference speed. In an ablation analysis, WBF, augmentation, transfer learning, and focal loss significantly improve performance, with focal loss increasing F1 by 6.3% for underrepresented classes and COCO pretraining boosting mAP by 8.7%. Additional architectural comparisons demonstrate that our full hybrid configuration not only maintains competitive accuracy but also achieves up to 150 FPS, making it well suited for real-time use cases. Robustness tests under challenging conditions, including real-world visual disturbances (smoke, fire, motion blur, varying lighting, and occlusions) and artificial noise (Gaussian; salt-and-pepper), confirm the model’s generalization ability. This work contributes a scalable, explainable, and high-performance solution for real-world vehicle damage diagnostics.

Zhihui Yang, Qingyong Zhang, Wanfeng Chang et al.

Due to the regular influence of human activities, traffic flow data usually exhibit significant periodicity, which provides a foundation for further research on traffic flow data. However, the temporal dependencies in traffic flow data are often obscured by entangled temporal regularities, making it challenging for general models to capture the intrinsic functional relationships within the data accurately. In recent years, a plethora of methods based on statistics, machine learning, and deep learning have been proposed to tackle these problems of traffic flow forecasting. In this paper, the Transformer is improved from two aspects: (1) an Efficient Attention mechanism is proposed, which reduces the time and memory complexity of the Scaled Dot Product Attention; (2) a Generative Decoding mechanism instead of a Dynamic Decoding operation, which accelerates the inference speed of the model. The model is named EGFormer in this paper. Through a lot of experiments and comparative analysis, the authors found that the EGFormer has better ability in the traffic flow forecasting task. The new model has higher prediction accuracy and shorter running time compared with the traditional model.

Shashwat Goel, Ameya Prabhu, Philip Torr et al.

Machine Learning models increasingly face data integrity challenges due to the use of large-scale training datasets drawn from the Internet. We study what model developers can do if they detect that some data was manipulated or incorrect. Such manipulated data can cause adverse effects including vulnerability to backdoored samples, systemic biases, and reduced accuracy on certain input domains. Realistically, all manipulated training samples cannot be identified, and only a small, representative subset of the affected data can be flagged. We formalize Corrective Machine Unlearning as the problem of mitigating the impact of data affected by unknown manipulations on a trained model, only having identified a subset of the corrupted data. We demonstrate that the problem of corrective unlearning has significantly different requirements from traditional privacy-oriented unlearning. We find most existing unlearning methods, including retraining-from-scratch without the deletion set, require most of the manipulated data to be identified for effective corrective unlearning. However, one approach, Selective Synaptic Dampening, achieves limited success, unlearning adverse effects with just a small portion of the manipulated samples in our setting, which shows encouraging signs for future progress. We hope our work spurs research towards developing better methods for corrective unlearning and offers practitioners a new strategy to handle data integrity challenges arising from web-scale training. Code is available at https://github.com/drimpossible/corrective-unlearning-bench.

Joe Watson, Chen Song, Oliver Weeger et al.

This survey examines the broad suite of methods and models for combining machine learning with physics knowledge for prediction and forecast, with a focus on partial differential equations. These methods have attracted significant interest due to their potential impact on advancing scientific research and industrial practices by improving predictive models with small- or large-scale datasets and expressive predictive models with useful inductive biases. The survey has two parts. The first considers incorporating physics knowledge on an architectural level through objective functions, structured predictive models, and data augmentation. The second considers data as physics knowledge, which motivates looking at multi-task, meta, and contextual learning as an alternative approach to incorporating physics knowledge in a data-driven fashion. Finally, we also provide an industrial perspective on the application of these methods and a survey of the open-source ecosystem for physics-informed machine learning.

Yining Fu, Xindong Ni, Jingxuan Yang et al.

The positional deviation of the in-vehicle Automatic Emergency Call System (AECS) under collision conditions brings difficulties for Intelligent Connected Vehicles (ICVs) post rescue operations. Currently, there is a lack of analysis on system operating conditions during collisions in the reliability assessment methods for the Global Navigation Satellite System (GNSS) deployed in the AECS. Therefore, this paper establishes an in-vehicle collision environment simulation model for emergency calls to explore the influence of parameters such as temperature and vibration on Signal-Based In-Vehicle Emergency Call Systems. We also propose environmental limits applicable to comprehensive tests, which can objectively evaluate reliability and provide data support for the AECS bench test through a satellite-signal-based semi-physical simulation, which is subjected to a bench test under different operating conditions. The findings of this study demonstrate that the occurrence of random vibration and impact stress, induced by vibration, exerts considerable disruptive effects on positional signal data during collisions. Consequently, it leads to substantial interference with the accurate detection of post-collision satellite positioning information. When the simulation operates under a 2.4 gRMS vibration condition, the maximum phase noise error in the positioning system is 8.95%, which does not meet the test accuracy requirements. On the other hand, the semi-simulation system is less affected by temperature changes, and at the maximum allowable temperature difference of the equipment, the maximum phase noise error in the simulated signal is 2.12%. Therefore, based on the influence of phase noise variation on the accuracy of the satellite signal simulation, necessary environmental conditions for the test are obtained, including a temperature that is consistent with the maximum operating temperature of the vector generator and a vibration power spectral density (PSD) lower than 1.2 gRMS.

Pirouz Nourian, Shervin Azadi, Roy Uijtendaal et al.

This chapter presents methodological reflections on the necessity and utility of artificial intelligence in generative design. Specifically, the chapter discusses how generative design processes can be augmented by AI to deliver in terms of a few outcomes of interest or performance indicators while dealing with hundreds or thousands of small decisions. The core of the performance-based generative design paradigm is about making statistical or simulation-driven associations between these choices and consequences for mapping and navigating such a complex decision space. This chapter will discuss promising directions in Artificial Intelligence for augmenting decision-making processes in architectural design for mapping and navigating complex design spaces.

Luis Oala, Manil Maskey, Lilith Bat-Leah et al.

Drawing from discussions at the inaugural DMLR workshop at ICML 2023 and meetings prior, in this report we outline the relevance of community engagement and infrastructure development for the creation of next-generation public datasets that will advance machine learning science. We chart a path forward as a collective effort to sustain the creation and maintenance of these datasets and methods towards positive scientific, societal and business impact.

Rebecca Potts, Rick Hackney, Georgios Leontidis

Predicting emissions for gas turbines is critical for monitoring harmful pollutants being released into the atmosphere. In this study, we evaluate the performance of machine learning models for predicting emissions for gas turbines. We compare an existing predictive emissions model, a first principles-based Chemical Kinetics model, against two machine learning models we developed based on SAINT and XGBoost, to demonstrate improved predictive performance of nitrogen oxides (NOx) and carbon monoxide (CO) using machine learning techniques. Our analysis utilises a Siemens Energy gas turbine test bed tabular dataset to train and validate the machine learning models. Additionally, we explore the trade-off between incorporating more features to enhance the model complexity, and the resulting presence of increased missing values in the dataset.

Sonny Achten, Arun Pandey, Hannes De Meulemeester et al.

We propose a unifying setting that combines existing restricted kernel machine methods into a single primal-dual multi-view framework for kernel principal component analysis in both supervised and unsupervised settings. We derive the primal and dual representations of the framework and relate different training and inference algorithms from a theoretical perspective. We show how to achieve full equivalence in primal and dual formulations by rescaling primal variables. Finally, we experimentally validate the equivalence and provide insight into the relationships between different methods on a number of time series data sets by recursively forecasting unseen test data and visualizing the learned features.

Sohum Thakkar, Skander Kazdaghli, Natansh Mathur et al.

Quantum algorithms have the potential to enhance machine learning across a variety of domains and applications. In this work, we show how quantum machine learning can be used to improve financial forecasting. First, we use classical and quantum Determinantal Point Processes to enhance Random Forest models for churn prediction, improving precision by almost 6%. Second, we design quantum neural network architectures with orthogonal and compound layers for credit risk assessment, which match classical performance with significantly fewer parameters. Our results demonstrate that leveraging quantum ideas can effectively enhance the performance of machine learning, both today as quantum-inspired classical ML solutions, and even more in the future, with the advent of better quantum hardware.

Maksym Diachuk, Said M. Easa

This paper proposes an alternative model for describing the hydro-mechanical drive operation of the automatic transmissions. The study is aimed at preparing a reliable model that meets the requirements of sufficient informativeness and rapidity to, basically, be used as a model for optimized control. The study relevance is stipulated by the need for simple and precise models ensuring minimal computational costs in model predictive control (MPC) procedures. The paper proposes a method for coordinating the engine’s control and operating modes, with the torque converter (TC) operating mode, based on the criteria of angular acceleration derivative (jerk), which fosters including the angular acceleration in the state vector for using the optimal control. The latter provides stronger prediction quality than using only the crankshaft angular speed criterion. This moment comprises a study novelty. Additionally, the proposed approach can be helpful in tasks of powertrain automation, autonomous vehicles’ integrated control, elaboration of control algorithms, co-simulations, and real-time applications. The paper material is structured by the modeling stages, including mathematics and simulations, data preparation, testing and validation, virtual experiments, analysis of results, and conclusions. The essence of the problem, goals, and objectives are first presented, followed by the overview of main approaches in modeling the automatic transmission elements. The internal combustion engine (ICE), torque converter, drivetrain, tires, and translational dynamics mathematical models are determined and discussed in detail. The proposed approach convergence on decomposing the indicators of powertrain aggregates by derivatives is demonstrated. The considered method was simulated by using the data of the Audi A4 Quattro. The gear shifting control algorithm was described in detail, and a series of virtual tests for the composed model were carried out. The comparative analysis of the results for the conventional and advanced models of the automatic transmission’s hydro-mechanical drive were performed. The differences of the model outputs were discussed. The approach advantages were noted, as well as the options for extending the proposed technique.

Mochamad Cholik

Georgios Giannarakis, Vasileios Sitokonstantinou, Roxanne Suzette Lorilla et al.

To fight climate change and accommodate the increasing population, global crop production has to be strengthened. To achieve the "sustainable intensification" of agriculture, transforming it from carbon emitter to carbon sink is a priority, and understanding the environmental impact of agricultural management practices is a fundamental prerequisite to that. At the same time, the global agricultural landscape is deeply heterogeneous, with differences in climate, soil, and land use inducing variations in how agricultural systems respond to farmer actions. The "personalization" of sustainable agriculture with the provision of locally adapted management advice is thus a necessary condition for the efficient uplift of green metrics, and an integral development in imminent policies. Here, we formulate personalized sustainable agriculture as a Conditional Average Treatment Effect estimation task and use Causal Machine Learning for tackling it. Leveraging climate data, land use information and employing Double Machine Learning, we estimate the heterogeneous effect of sustainable practices on the field-level Soil Organic Carbon content in Lithuania. We thus provide a data-driven perspective for targeting sustainable practices and effectively expanding the global carbon sink.

Dario Niermann, Alexander Trende, Klas Ihme et al.

The quickly rising development of autonomous vehicle technology and increase of (semi-) autonomous vehicles on the road leads to an increased demand for more sophisticated human–machine-cooperation approaches to improve trust and acceptance of these new systems. In this work, we investigate the feeling of discomfort of human passengers while driving autonomously and the automatic detection of this discomfort with several model approaches, using the combination of different data sources. Based on a driving simulator study, we analyzed the discomfort reports of 50 participants for autonomous inner city driving. We found that perceived discomfort depends on the driving scenario (with discomfort generally peaking in complex situations) and on the passenger (resulting in interindividual differences in reported discomfort extend and duration). Further, we describe three different model approaches on how to predict the passenger discomfort using data from the vehicle’s sensors as well as physiological and behavioral data from the passenger. The model’s precision varies greatly across the approaches, the best approach having a precision of up to 80%. All of our presented model approaches use combinations of linear models and are thus fast, transparent, and safe. Lastly, we analyzed these models using the SHAP method, which enables explaining the models’ discomfort predictions. These explanations are used to infer the importance of our collected features and to create a scenario-based discomfort analysis. Our work demonstrates a novel approach on passenger state modelling with simple, safe, and transparent models and with explainable model predictions, which can be used to adapt the vehicles’ actions to the needs of the passenger.

Sidun Iurii, Vollis Oleksiy, Hidei Volodymyr et al.

Bitumen emulsions for slurry surfacing mix technology using oxidized bitumen and hydrochloric and orthophosphoric acids on laboratory DenimoTech bitumen-emulsion plant are made in the work. Methylene blue adsorption index of granite aggregate for use in slurry surfacing mix was investigated. Comparatively mix time and cohesion strength build-up of selected compositions slurry surfacing mix with hydrochloric and orthophosphoric acids depending on the variable content of bitumen emulsion. The advantage of using orthophosphoric acid in slurry surfacing mix according to the cohesion strength build-up criterion has been proved. The importance of correct distribution of bitumen drops in the emulsion was confirmed using a sedimentograph Mastersizer 2000. Two emulsions of the same component composition were compared, which differed in particle size. It has been established that it is not possible to design a slurry surfacing mix using the mix time criterion with the help of coarse bitumen emulsion.

Alex Glushkovsky

Design of experiments (DOE) is playing an essential role in learning and improving a variety of objects and processes. The article discusses the application of unsupervised machine learning to support the pragmatic designs of complex experiments. Complex experiments are characterized by having a large number of factors, mixed-level designs, and may be subject to constraints that eliminate some unfeasible trials for various reasons. Having such attributes, it is very challenging to design pragmatic experiments that are economically, operationally, and timely sound. It means a significant decrease in the number of required trials from a full factorial design, while still attempting to achieve the defined objectives. A beta variational autoencoder (beta-VAE) has been applied to represent trials of the initial full factorial design after filtering out unfeasible trials on the low dimensional latent space. Regarding visualization and interpretability, the paper is limited to 2D representations. Beta-VAE supports (1) orthogonality of the latent space dimensions, (2) isotropic multivariate standard normal distribution of the representation on the latent space, (3) disentanglement of the latent space representation by levels of factors, (4) propagation of the applied constraints of the initial design into the latent space, and (5) generation of trials by decoding latent space points. Having an initial design representation on the latent space with such properties, it allows for the generation of pragmatic design of experiments (G-DOE) by specifying the number of trials and their pattern on the latent space, such as square or polar grids. Clustering and aggregated gradient metrics have been shown to guide grid specification.