Engineering reports : open access最新文献

Dynamic process modeling is essential for simulating time-evolving biochemical systems, particularly those with multistate interactions and combinatorial complexity. Traditional Ordinary Differential Equation (ODE) models offer mechanistic clarity but struggle with scalability and context-sensitive encoding. Rule-Based Modeling (RBM) frameworks address these limitations through modular rule abstraction, yet require manual specification and lack adaptive learning. This study introduces algorithmic innovations within the Neural Ordinary Differential Equation (Neural ODE) paradigm to bridge the gap between mechanistic interpretability and scalable expressivity. Neural ODEs can be considered as a revolutionary approach in the field of modeling dynamic biochemical interactions. They have made it possible to create models of such interactions that are flexible enough to adapt to different scenarios and do so without requiring any manual intervention in terms of rule encoding or predefined reaction schemes. This is achieved by employing differential solvers within the framework of neural networks, thus enabling a learning process that is in accordance with the behavior of the system. Using the DARPP-32 signaling network—a benchmark system characterized by multivalent phosphorylation and dynamic perturbations—the proposed Neural ODE framework demonstrates the ability to replicate key dynamic behaviors observed in ODE and RBM models. Comparative simulations under baseline and perturbed conditions reveal that Neural ODEs maintain trajectory fidelity while offering enhanced modularity and computational efficiency. Feature importance analysis and latent space visualizations further validate the model's interpretability and robustness. Unlike ODEs and RBMs, Neural ODEs adapt to structural mutations and binding schemes through latent trajectory learning, enabling flexible simulation of biochemical variability without manual rule encoding. This work establishes Neural ODEs as a viable and scalable alternative for modeling complex biochemical systems, combining the strengths of data-driven learning with the interpretability of differential equations.

In this work, the influence of strut thickness on the deformation and failure mechanisms of new vascular bundle–inspired structures, which exhibit comparable or better mechanical properties than honeycomb and star-shaped lattices, is presented. The novelty of the work lies on the design of the structure; this is a new structure, and its behavior has not been reported elsewhere. Structures consisting of 0.2, 0.5, 1.0-, and 1.15-mm strut thicknesses were designed, modeled, fabricated, and tested. A finite element model of a quasi-static compression test is developed in ANSYS Explicit Dynamics to evaluate the deformation and failure mechanisms of the various structures. It is demonstrated that 0.2- and 0.5-mm structures exhibit stretch-dominated stress–strain behavior, whereas 1.0- and 1.15-mm structures show bend-dominated stress–strain characteristics. As the strut thickness increases, there is an increase in peak stresses (with reported peak stresses of 1.3, 1.4, 5, and 5.1 MPa for 0.2, 0.5, 1.0, and 1.15 mm, respectively) and energy absorption (reported values of 33.84, 31.48, 159.28, and 179.07 J for thicknesses of 0.2, 0.5, 1.0, and 1.15 mm, respectively) characteristics. Poisson's ratio values of the samples ranged between 0.6 and 1.2. Additionally, the deformation mechanisms transform from perpendicular collapse of the structure to 45° bending (shearing) of the structure from low to higher strut thickness. As the strut thickness increases, the failure mechanisms transform from ductile fracture to near-brittle failure of the structures. The findings in this paper provide key insights into the design and fabrication of next-generation vascular bundle–inspired multifunctional materials for lightweight structural applications. As a contribution, the energy absorption and peak stress values for the vascular bundle structures presented in this paper are comparable to published data on similar PLA lattice structures.

This study develops the Fractional Novel Analytical Method (FNAM), a Taylor-series–oriented approach for constructing approximate analytical solutions of NFDΔEs prevalent in control, integrability studies, and arithmetic modeling. Grounded in the Caputo fractional derivative, the method attains rapid convergence of truncated series and eliminates dependence on Adomian polynomial decompositions, multiplier methods, auxiliary parameters, perturbative schemes, and transform operators. Testing on three well-known NFDΔEs with fractional order $��\alpha �ϵ��(��0�,�1�]��\hspace{0.17em}��$ and combined delay terms reveal that FNAM secures high-fidelity approximations with limited series terms. The method proceeds by extracting a direct coefficient recurrence from the NFDΔE.A short convergence proof is outlined. Across all test cases, few-term truncations suffice to reach high accuracy, with absolute errors below those of ADTM/HATM/PIA/MHLM under matched truncation depth and reduced runtime due to analytic coefficient recurrences. Graphical overlays against exact benchmarks show strong concordance. In concert, the analytical framework and numerical results show that FNAM provides a robust and resource-efficient solution strategy for NFDΔEs, with competitive accuracy achieved through minimal machinery. The method's transform-independent design, elementary calculus basis, and reliable convergence characteristics make it an attractive option for a wide class of fractional models.

This study examines the fragmented and rapidly evolving body of knowledge on the application of quantum computing to NP-hard decision problems in Operations Management (OM) and Operations Research (OR). It aims to systematically map how quantum optimization approaches are formulated and applied across core OM/OR problem classes, highlighting current advances and unresolved challenges for research and practice. A systematic mapping review was conducted using peer-reviewed studies indexed in Scopus and Web of Science from 2014 to 2026. The literature was classified by problem type, mathematical formulation, quantum technique, and application domain, with attention to the alignment between quantum models and established OM/OR decision frameworks. The review reveals a strong predominance of QUBO-based formulations and annealing-oriented approaches, mainly applied to logistics, manufacturing, and financial optimization problems. Applications remain largely exploratory, with limited empirical validation, weak theoretical integration with OM/OR decision-making models, and persistent challenges related to scalability and hybrid quantum-classical performance. Only a small subset of studies demonstrates how quantum formulations can support real-world scheduling and resource coordination problems. This study proposes an analytical framework linking quantum optimization paradigms to canonical OM/OR NP-hard problem classes, identifying key research gaps and methodological tensions to support more theory-driven and empirically grounded future applications, especially in complex decision environments.

The growing energy demand has intensified reliability challenges in power distribution systems. This study employs network reconfiguration to minimize power loss, enhance voltage stability, and improve system reliability. A modified particle swarm optimization algorithm incorporating a chaotic map and a linearly decreasing inertia weight (PSO-CLDIW) is proposed to address the limitations of conventional PSO. IEEE 33 and 69 bus systems were used to test the proposed method for power loss and reliability improvement. For the IEEE 33-bus system, the power loss of PSO-CLDIW decreased from 202.6771 to 139.5513 kW after reconfiguration under normal load. The reliability indices include SAIFI, SAIDI, CAIDI, ASAI, EENS, and AENS, with the corresponding values of 2.0261, 1.4291, 0.7053, 0.9998, 1522.683, and 0.0837, respectively. Under light load, PSO-CLDIW achieved 33.433 kW loss compared to 47.071 kW in the base case, reflecting a 28.98% reduction. At heavy load, the power loss was 381.14 kW, resulting in 33.73% improvement. Reliability indices also showed significant enhancement, with SAIFI, SAIDI, CAIDI, EENS, and AENS values improving to 1.984, 1.368, 0.68953, 377.1413, and 0.020722, respectively, under light load, and 1.994, 1.386, 0.695, 3882.871, and 0.213 under heavy load. The IEEE 69-bus system power loss decreased from 224.9606 to 98.5902 kW after reconfiguration under normal load, and SAIFI, SAIDI, CAIDI, ASAI, EENS, and AENS values are 1.112, 0.7613, 0.68464, 0.9999, 2913.084, and 0.2339, respectively, under normal load. PSO-CLDIW achieved a power loss of 23.827 kW at light load, corresponding to a 53.83% reduction. Reliability indices improved, with SAIFI, SAIDI, CAIDI, ASAI, EENS, and AENS reaching 1.0246, 0.73701, 0.719321, 0.99992, 728.281, and 0.0585, respectively. At the heavy load, the power loss was 276.12 kW. Overall, the proposed PSO-CLDIW method outperforms the baseline and other optimization techniques, demonstrating the superior ability of PSO-CLDIW to minimize power loss, enhance reliability, and improve the voltage profile in distribution systems.

The study simulation of delamination and damage prediction in composite materials is an important area of research that impacts various engineering applications, such as aerospace structures, significantly influencing their static, dynamic, buckling, fatigue, and crack resistance. To this end, this review presents a comprehensive analysis of modern approaches to predicting delamination and damage in polymer composites. In this regard, the review summarizes the views of 200 published studies and examines various computational methods, including finite element formulations, adaptive hierarchical kinematics, extended finite element methods (xFEM), cohesive zone models (CZM), and experimental methods, to evaluate their performance in damage prediction and delamination analysis. This review examines the strengths and weaknesses of each approach and highlights the need for further improvement in delamination analysis and damage prediction in the static, dynamic, buckling, and fatigue response and fracture behavior of composite structures. Dynamic analysis reveals that delamination substantially alters natural frequencies, damping characteristics, and vibration modes while influencing aeroelastic stability and impact resistance. Fatigue and fracture analyses demonstrate the critical roles of residual stress, fiber bridging, and material architecture in governing delamination propagation behavior. Experimental validation techniques, ranging from piezoelectric sensors to laser vibrometry and Lamb wave-based structural health monitoring, are increasingly integrated with machine learning algorithms to enable real-time damage detection and localization. Through systematic synthesis of these multidisciplinary advancements, this review identifies emerging trends including transitions toward higher-order theories, physics-informed machine learning approaches, and experiment-numerical hybrid paradigms, collectively offering promising directions for enhancing advanced composite reliability and structural integrity.

Diseases during the growth phases have a significant impact on the production of citrus fruits, degrading the quality of the fruits. To prevent significant losses, early detection of the disease severity and an accurate diagnosis are crucial. In this study, we emphasized the citrus fruit disease classification, then we tested it on a non-citrus apple leaf dataset for better confirmation. The research uses deep learning models (VGG16, VGG19, ResNet50) along with machine learning algorithms (KNN, Naive Bayes, Random Forest, SVM, Logistic Regression) for disease classification to increase accuracy. Accordingly, we achieved the highest accuracy for disease classification, with 99.69% for orange using ResNet50 paired with Logistic Regression, and 95% for lemon and 99.20% for apple leaf using ResNet50 combined with SVM, along with Recall of 96.60%, Precision of 95.80%, F1-Score of 95.90%, MCC of 96.70%, Kappa of 96.60% and GDR of 97.40% for lemon, and Recall of 99.20%, Precision of 99.10%, F1-Score of 99.10%, MCC of 98.80%, Kappa of 98.80% and GDR of 99.10% for apple leaf, while the ResNet50 with Logistic Regression model achieved Recall of 99.60%, Precision of 99.60%, F1 score 99.60%, MCC of 99.20%, Kappa of 99.20%, and GDR of 99.30% for orange. The proposed model also outperforms the existing models in which most of them classified the diseases using the Softmax classifier without using any individual classifiers. Furthermore, k-means clustering is used to find the infected region of fruits, and a ResNet50-based fuzzy logic control system is used to evaluate degrees of severity, especially of lemon and orange diseases. Moreover, K-fold cross-validation has been employed to ensure the model's robustness and validate its performance.

This research investigates the performance, combustion, and emission characteristics of a novel binary biodiesel blend synthesized from rice bran oil (RBO) and waste cooking oil (WCO), addressing the critical need for sustainable, nonedible second-generation feedstocks. The primary objective was to evaluate the synergistic effects of combining these two distinct oils through transesterification and magnetic stirring to optimize fuel properties. The study uniquely identifies a 50:50 ratio of WCO to RBO as the optimum precursor for secondary blending with mineral diesel. Experimental results reveal that while biodiesel blends exhibit a slight reduction in Brake Thermal Efficiency (BTE) and an increase in Brake Specific Fuel Consumption (BSFC), specifically a 22.2% increase for the B70 blend, they provide superior safety profiles with flash and fire points significantly exceeding those of conventional diesel. The research demonstrates a substantial environmental benefit: B30 blend (30% biodiesel, 70% diesel) achieved a 23.5% reduction in hydrocarbon (HC) emissions and a 13.6% reduction in carbon monoxide (CO) compared to standard diesel. The uniqueness of this work lies in the strategic binary coupling of a high-viscosity by-product (RBO) with a post-consumer waste (WCO) to achieve a balanced fuel profile that meets international standards without requiring engine modifications.

This research has introduced a hybrid model that integrates the long short-term memory (LSTM) and extreme gradient boosting (XGBoost) models to assess students' mental health states, particularly to identify students' levels of stress, mood, and fatigue. The physiological measures measured were heart rate (HR), heart rate variability (HRV), electrodermal activity (EDA), and skin temperature. All measures were recorded using wearable sensors and underwent processing, such as normalization, noise filtering, and feature extraction, to ensure the signal quality was fit for analysis and interpretability. While the LSTM network can accurately represent the temporal dynamics present in the physiological sequences, the XGBoost model is critical in obtaining high accuracy through the classification of features' non-linear interactions and decision boundary optimization. The experimental validation through the technique of fivefold cross-validation shows that the hybrid model performs with high accuracy of 0.98 on average, F1-score of 0.98, and consistently low false-positive and false-negative rates when compared to SVM, Random Forest, and single deep learning model methods that serve as baseline methods. The results assure the framework's reliability, consistency, and clarity in reasoning over different data conditions. This novel method provides a strong platform for the real-time, data-driven monitoring and early detection of psychological distress, thus allowing educators, mental-health professionals, and caregivers to make timely interventions and improve the overall well-being of students.

High energy densities are vital to satisfy the increasing demand for battery storage systems for electric vehicles. One innovative battery type of the next generation is the solid-state battery, which is characterized by the high expected energy density. The polymer-based solid-state battery is notable for its high machinability in production and, therefore, offers great potential for industrial scale. One component of the polymer-based solid-state battery is the composite cathode, which faces particular challenges in the individual production processes. The calendering process is essential, as it can increase the ionic conductivity through a reduction of the composite cathode porosity. For this reason, the calendering process for polymer-based composite cathodes with different compositions of active material and solid electrolyte has been analyzed in depth in this work. This enabled extensive analysis of the calendering process with different material compositions of polymer-based composite cathodes to provide a profound understanding of the causal-effect relationships.