Search Results (62)

To overcome the limitations of single-objective optimization in permanent magnet synchronous motor (PMSM) performance enhancement, this study proposes an adaptive moving least squares (AMLS) for a 12-pole/36-slot built-in PMSM. Through comprehensive exploration of the design space, a systematic approach is established for holistic motor performance improvement. The Gaussian weight function is modified to improve the model’s fitting accuracy, and the decay rate of the control weight is optimized. The optimal adaptive meta-model for the built-in PMSM is selected based on the coefficient of determination. Subsequently, sensitivity analysis is conducted to identify the parameters that most significantly influence key performance indicators, including torque ripple, stator core loss, electromagnetic force amplitude, and average output torque. These parameters are then chosen as the optimal design variables. A multi-objective optimization framework, built upon the optimal adaptive meta-model, is developed to address the multi-objective optimization problem. The results demonstrate increased output torque, along with reductions in stator core loss, torque ripple, and radial electromagnetic force, thereby significantly improving the overall performance of the motor. Full article

Machine predictive maintenance plays a critical role in reducing unplanned downtime, lowering maintenance costs, and improving operational reliability by enabling the early detection and classification of potential failures. Artificial intelligence (AI) enhances these capabilities through advanced algorithms that can analyze complex sensor data with high accuracy and adaptability. This study introduces an explainable AI framework for failure detection and classification using symbolic expressions (SEs) derived from a genetic programming symbolic classifier (GPSC). Due to the imbalanced nature and wide variable ranges in the original dataset, we applied scaling/normalization and oversampling techniques to generate multiple balanced dataset variations. Each variation was used to train the GPSC with five-fold cross-validation, and optimal hyperparameters were selected using a Random Hyperparameter Value Search (RHVS) method. However, as the initial Threshold-Based Voting Ensembles (TBVEs) built from SEs did not achieve a satisfactory performance for all classes, a meta-dataset was developed from the outputs of the obtained SEs. For each class, a meta-dataset was preprocessed, balanced, and used to train a Random Forest Classifier (RFC) with hyperparameter tuning via RandomizedSearchCV. For each class, a TBVE was then constructed from the saved RFC models. The resulting ensemble demonstrated a near-perfect performance for failure detection and classification in most classes (0, 1, 3, and 5), although Classes 2 and 4 achieved a lower performance, which could be attributed to an extremely low number of samples and a hard-to-detect type of failure. Overall, the proposed method presents a robust and explainable AI solution for predictive maintenance, combining symbolic learning with ensemble-based meta-modeling. Full article

Projected changes in outdoor environmental conditions are expected to significantly alter building energy demand across the United States. Yet, policymakers and designers lack typology and climate-zone-specific guidance to support long-term planning. We simulated 10 U.S. Department of Energy (DOE) prototype buildings across all 16 ASHRAE climate zones with EnergyPlus. Future weather files generated in Meteonorm from a CMIP6 ensemble reflected two emissions pathways (RCP 4.5 and RCP 8.5) and two planning horizons (2050 and 2080), producing 800 simulations. Envelope parameters and schedules were held at DOE reference values to isolate the pure climate signal. Results show that cooling energy use intensity (EUI) in very hot-humid Zones 1A–2A climbs by 12% for full-service restaurants and 21% for medium offices by 2080 under RCP 8.5, while heating EUI in sub-arctic Zone 8 falls by 14–20%. Hospitals and large hotels change by < 6%, showing resilience linked to high internal gains. A simple linear-regression meta-model (R² > 0.90) links baseline EUI to future percentage change, enabling rapid screening of vulnerable stock without further simulation. These high-resolution maps supply actionable targets for state code updates, retrofit prioritization, and long-term decarbonization planning to support climate adaptation and sustainable development. Full article

This study aims to improve the accuracy and interpretability of deep groundwater level forecasting in Cangzhou, a typical overexploitation area in the North China Plain. To address the limitations of traditional models and existing machine learning approaches, we develop a Stacking ensemble learning framework that integrates meteorological, spatial, and anthropogenic variables, including lagged groundwater levels to reflect aquifer memory. The model combines six heterogeneous base learners with a meta-model to enhance prediction robustness. Performance evaluation shows that the ensemble model consistently outperforms individual models in accuracy, generalization, and spatial adaptability. Scenario-based simulations are further conducted to assess the effects of the South-to-North Water Diversion Project. Results indicate that the diversion project significantly mitigates groundwater depletion, with the most overexploited zones showing water level recovery of up to 17 m compared to the no-diversion scenario. Feature importance analysis confirms that lagged water levels and pumping volumes are dominant predictors, aligning with groundwater system dynamics. These findings demonstrate the effectiveness of ensemble learning in modeling complex groundwater behavior and provide a practical tool for water resource regulation. The proposed framework is adaptable to other groundwater-stressed regions and supports dynamic policy design for sustainable groundwater management. Full article

This paper proposes and demonstrates a domain-adapted ensemble machine learning approach for enhanced prediction of surface roughness (Ra) during the machining of polymeric materials. The proposed model methodology employs a two-stage pipelined architecture, where classified data are fed into the model for regressive analysis. First, a classifier (Logistic Regression or XGBoost, selected based on performance) categorizes machining data into distinct regimes based on cutting Speed (Vc), feed rate (f), and depth of cut (ap) as inputs. This classification leverages output discretization to mitigate data imbalance and capture regime-specific patterns. Second, a regressor (Support Vector Regressor or XGBoost, selected based on performance) predicts Ra within each regime, utilizing the classifier’s output as an additional feature. This structured hybrid approach enables more robust prediction in small, noisy datasets characteristic of machining studies. To validate the methodology, experiments were conducted on Polyoxymethylene (POM), Polytetrafluoroethylene (PTFE), Polyether ether ketone (PEEK), and PEEK/MWCNT composite, using a L₂₇ Design of Experiments (DoEs) matrix. Model performance was optimized using k-fold cross-validation and hyperparameter tuning via grid search, with R-squared and RMSE as evaluation metrics. The resulting meta-model demonstrated high accuracy (R² > 90% for XGBoost regressor across all materials), significantly improving Ra prediction compared to single-model approaches. This enhanced predictive capability offers potential for optimizing machining processes and reducing material waste in polymer manufacturing. Full article

Vehicles are now rapidly transitioning from a conventional torsion bar suspension to an in-arm suspension unit (ISU), reflecting the growing industrial demand for more compact, high-performance systems. Although the ISU system can adapt well to rough terrain, the side forces produced when the piston moves can affect its reliability. Current models built on ideal gas assumptions fail to describe the complex nonlinear behavior of nitrogen under extreme pressure and temperature variations. This study incorporates the Beattie–Bridgeman real-gas equation into a dynamic force-displacement model to overcome this limitation. Furthermore, a bi-objective optimization strategy was devised that simultaneously minimizes the side forces and enhances acceleration stability across diverse environmental conditions. The optimized design based on a metamodel and a hybrid metaheuristic algorithm resulted in an 81.4% reduction in peak lateral forces and a 53.3% improvement in acceleration robustness, which marks a significant increase in suspension system durability. These findings not only advance ISU design methodologies but also offer viable solutions to existing reliability challenges. Full article

Facility layout design (FLD) is critical for optimizing manufacturing efficiency, yet traditional approaches struggle with complexity, dynamic constraints, and fragmented data integration. This study proposes a generative-AI-enabled facility layout design, a novel paradigm aligning with Industry 4.0, to address these challenges by integrating generative artificial intelligence (AI), semantic models, and data-driven optimization. The proposed method evolves from three historical paradigms: experience-based methods, operations research, and simulation-based engineering. The metamodels supporting the generative-AI-enabled facility layout design is the Asset Administration Shell (AAS), which digitizes physical assets and their relationships, enabling interoperability across systems. Domain-specific knowledge graphs, constructed by parsing AAS metadata and enriched by large language models (LLMs), capture multifaceted relationships (e.g., spatial adjacency, process dependencies, safety constraints) to guide layout generation. The convolutional knowledge graph embedding (ConvE) method is employed for link prediction, converting entities and relationships into low-dimensional vectors to infer optimal spatial arrangements while addressing data sparsity through negative sampling. The proposed reference architecture for generative-AI-enabled facility layout design supports end-to-end layout design, featuring a 3D visualization engine, AI-driven optimization, and real-time digital twins. Prototype testing demonstrates the system’s end-to-end generation ability from requirement-driven contextual prompts and extensively reduced complexity of modeling, integration, and optimization. Key innovations include the fusion of AAS with LLM-derived contextual knowledge, dynamic adaptation via big data streams, and a hybrid optimization approach balancing competing objectives. The 3D layout generation results demonstrate a scalable, adaptive solution for storage workshops, bridging gaps between isolated data models and human–AI collaboration. This research establishes a foundational framework for AI-driven facility planning, offering actionable insights for AI-enabled facility layout design adoption and highlighting future directions in the generative design of complex engineering. Full article

Chlorophyll is an essential pigment for photosynthesis in tea plants, and fluctuations in its content directly impact the growth and developmental processes of tea trees, thereby influencing the final quality of the tea. Therefore, achieving rapid and non-destructive real-time monitoring of leaf chlorophyll content (LCC) is beneficial for precise management in tea plantations. In this study, derivative transformations were first applied to preprocess the tea hyperspectral data, followed by the use of the Stable Competitive Adaptive Reweighted Sampling (SCARS) algorithm for feature variable selection. Finally, multiple individual machine learning models and stacking models were constructed to estimate tea LCC based on hyperspectral data, with a particular emphasis on analyzing how the selection of base models and meta-models affects the predictive performance of the stacking models. The results indicate that derivative processing enhances the sensitivity of hyperspectral data to tea LCC; furthermore, compared with individual machine learning models, the stacking models demonstrate superior predictive accuracy and generalization ability. Among the 17 constructed stacking configurations, when the meta-model is fixed, the predictive performance of the stacking model improves continuously with an increase in the number and accuracy of the base models and with a decrease in the structural similarity among the selected base models. Therefore, when constructing stacking models, the base model combination should comprise various models with minimal structural similarity while ensuring robust predictive performance, and the meta-model should be chosen as a simple linear or nonlinear model. Full article

High-risk industries like mining must address occupational safety to reduce accidents and fatalities. Training through role-playing, simulations, and Serious Games (SGs) can reduce occupational risks. This study aims to conduct a systematic literature review (SLR) on SG methodologies for the mining sector. This review was based on a methodology inspired by the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Three research questions were formulated to explore how SGs contribute to immediate feedback, brain stimulation, and training for high-risk scenarios. The review initially identified 1987 studies, which were reduced to 30 relevant publications following a three-phase process: (1) A search string based on three research questions was defined and applied to databases. (2) Publications were filtered by title and abstract. (3) A full-text reading was conducted to select relevant publications. The SLR showed SG development methodologies with structured processes that are adaptable to any case study. Additionally, it was found that Virtual Reality, despite its implementation costs, is the most used technology for safety training, inspection, and operation of heavy machinery. The first conclusion of this SLR indicates the lack of methodologies for the development of SG for training in the mining field, and the relevance of carrying out specific methodological studies in this field. Additionally, the main findings obtained from this SLR are the following: (1) Modeling languages (e.g., GML and UML) and metamodeling are important in SG development. (2) SG is a significant mechanism for cooperative and participative learning strategies. (3) Virtual Reality technology is widely used in safe virtual environments for mining training. (4) There is a need for methodologies that integrate the specification of cognitive functions with the affective part of the users for SGs suitable for learning environments. Finally, this review highlights critical gaps in current research and underscores the need for more integrative approaches to SG development. Full article

With the increasing demand for time-series analysis, driven by the proliferation of IoT devices and real-time data-driven systems, detecting change points in time series has become critical for accurate short-term prediction. The variability in patterns necessitates frequent analysis to sustain high performance by acquiring the hyperparameter. The Cumulative Sum (CUSUM) method, based on calculating the cumulative values within a time series, is commonly used for change detection due to its early detection of small drifts, simplicity, low computational cost, and robustness to noise. However, its effectiveness heavily depends on the hyperparameter configuration, as a single setup may not be universally suitable across the entire time series. Consequently, fine-tuning is often required to achieve optimal results, yet this selection process is traditionally performed through trial and error or prior expert knowledge, which introduces subjectivity and inefficiency. To address this challenge, several strategies have been proposed to facilitate hyperparameter optimizations, as traditional methods are impractical. Meta-learning-based techniques present viable alternatives for periodic hyperparameter optimization, enabling the selection of configurations that adapt to dynamic scenarios. This work introduces a meta-modeling scheme designed to automate the recommendation of hyperparameters for the CUSUM algorithm. Benchmark datasets from the literature were used to evaluate the proposed framework. The results indicate that this framework preserves high accuracy while significantly reducing time requirements compared to Grid Search and Genetic Algorithm optimization. Full article

Model-based systems engineering (MBSE) offers significant advantages over traditional document-based approaches, particularly in improving the clarity, traceability, and efficiency of requirements engineering (RE). However, MBSE also introduces challenges, particularly in maintaining consistent semantics and handling evolving system models. This paper presents ARMADE, an agile requirements management and definition environment developed at DLR, which aims to address these challenges. ARMADE enables the flexible, user-friendly modeling of system requirements using a data model that incorporates natural language patterns. The tool supports the dynamic adaptation of metamodels and facilitates collaborative, project-wide requirements management. A case study based on two systems of systems (SoS) from the EU-funded HE COLOSSUS project—firefighting and advanced aerial mobility (AAM)—demonstrates ARMADE’s ability to manage complex, interdisciplinary requirements. The study highlights the tool’s potential to reduce data inconsistencies, improve adaptability, and enhance the overall efficiency of the RE process. By enabling seamless updates and changes to requirements, ARMADE shows promise as a versatile solution for dynamic metamodeling in complex systems, with potential applications extending beyond aeronautics to various industries reliant on intricate requirements management. Full article

One of the challenging aspects of designing body-in-white stiffness test rigs is measuring test accuracy. This paper proposes a method of integrating the body-in-white stiffness test rig and the body-in-white into an overall model for the optimization design. It establishes an optimization mathematical model based on the overall structure of the stiffness test rig, taking into account the factors affecting the accuracy of the test results of the body-in-white stiffness test rig. The stiffness test rig’s testing accuracy can be significantly increased by designating the degrees of freedom at each connection position as discrete variables. The Hybrid and Adaptive Metamodeling Method (HAM) is used to optimize the mathematical model. This approach uses and integrates three distinct metamodels with various attributes. The body-in-white torsional stiffness test result error is only 1.1%, and the body-in-white bending stiffness test result error is only 3.4%, owing to the optimization result that was used to design and manufacture a set of body-in-white stiffness test rigs and use them for a body-in-white stiffness test verification. Full article

To achieve accurate predictions of tail strike events during the landing phase of flight training, we propose a stacking ensemble learning prediction model that uses Random Forest (RF), Support Vector Regression (SVR), K-Nearest Neighbors (KNN), and Adaptive Boosting (AdaBoost) as base models, with Logistic Regression (LR) serving as the meta-model. This model is built on non-exceedance flight data recorded on airborne SD cards. By evaluating the importance scores of the feature parameters influencing tail strike events, we identified the optimal set of features for model input while using the landing pitch angle as the model output. We then compared the R² and RMSE of each model. The results indicate that under a prediction horizon of 5 s prior to landing, the ensemble learning model demonstrates high predictive accuracy. This capability provides flight trainees with sufficient reaction time to adjust their flight attitudes, thereby helping to avoid the occurrence of tail strike events during landing. Full article

This manuscript presents the development and testing of a novel model designed to help organizations, particularly small and medium-sized enterprises (SMEs), address the challenges of integrating new technologies within the frameworks of Industry 4.0 and 5.0. The proposed model is a metamodel that evaluates organizational and contextual vulnerabilities concerning both existing technologies and potential external technologies under consideration for adoption. It synthesizes three foundational frameworks: the Viable System Model (VSM), the principles of viable and sustainable systems, and the Technology, Organization, and Environment (TOE) Model. The findings demonstrate the practical applicability of this model in an SME context, showcasing its ability to facilitate the gradual and sustainable adoption of new technologies. By aligning business needs with technological solutions and leveraging insights from computer science and organizational cybernetics, the model adapts to varying levels of technological adoption, integrating organizational dynamics and business evolution to support the implementation of emerging technologies. Full article

The design of complex aerospace systems requires a broad multidisciplinary knowledge base and an iterative approach to accommodate changes effectively. Engineering knowledge is commonly represented through engineering analyses and descriptive models with underlying semantics. While guidelines from systems engineering methodologies exist to guide the development of system models, creating a system model from scratch with every new application/system requires research into more adaptable and reusable modeling frameworks. In this context, this research demonstrates how a physics-based multidisciplinary analysis and optimization tool, SUAVE, can be leveraged to develop a system model. By leveraging the existing physics-based knowledge captured within SUAVE, the process benefits from the expertise embedded in the tool. To facilitate the systematic creation of the system model, an ontological metamodel is created in SysML. This metamodel is designed to capture the inner workings of the SUAVE tool, representing its concepts, relationships, and behaviors. By using this ontological metamodel as a modeling template, the process of creating the system model becomes more structured and organized. Overall, this research aims to streamline the process of building system models from scratch by leveraging existing knowledge and utilizing an ontological metamodel as a modeling template. This approach enhances formal knowledge representation and its consistency, and promotes reusability in multidisciplinary design problems. Full article