Search Results (3,253)

The reliability of the dust pump in an engine air filtration system significantly affects vehicle performance. Therefore, the extent to which the parameters of the dust pump impeller blade fixture affect its reliability is a critical consideration during blade design. This study investigated the influences of various impeller blade fixture parameters on reliability. First, a three-dimensional finite element model of the vehicle dust pump was established to analyse the reliability of the impeller blade fixture in terms of deflection and stress according to parameter value. Next, a parametric model was established, and parameter uncertainties were defined for reliability analysis. The relationships between the different parameters and the reliability of the impeller blade fixture were subsequently predicted by a BP neural network model trained and tested using 400 and 100 samples, respectively. Finally, the output of the BP neural network model was applied to analyse the principal and total importance measures of each considered impeller blade parameter to fixture reliability. This study shows that in the reliability design of the dust pump impeller blades, priority should be given to rotational speed, blade thickness, and material density, as these factors have the greatest impact on the reliability of the blade mounting system. Full article

Against the dual backdrop of the rural revitalization strategy and the pursuit of high-quality, balanced urban–rural education, optimizing rural campus spaces has emerged as an important lever for addressing educational resource disparities and improving pedagogical quality. However, conventional evaluation of campus space optimization faces two systemic dilemmas. First, top-down decision-making often neglects the authentic needs of diverse stakeholders and place-based knowledge, resulting in spatial interventions that lose regional distinctiveness. Second, routine public participation is constrained by geographical barriers, time costs, and sample-size limitations, which can amplify professional cognitive bias and impede comprehensive feedback formation. The compounded effect of these challenges contributes to a disconnect between spatial optimization outcomes and perceived needs, thereby constraining the distinctive development of rural educational spaces. To address these constraints, this study proposes a novel method that integrates regional spatial feature recognition with digital media-based public perception assessment. At the data collection and ethical governance level, the study strictly adheres to platform compliance and academic ethics. A total of 12,800 preliminary comments were scraped from major social media platforms (e.g., Douyin, Dianping, and Xiaohongshu) and processed through a three-stage screening workflow—keyword screening–rule-based filtering–manual verification—to yield 8616 valid records covering diverse public groups across China. All user-identifying information was fully anonymized to ensure lawful use and privacy protection. At the analytical modeling level, we develop a Transformer-based deep learning system that leverages multi-head attention mechanisms to capture implicit spatial-sentiment features and metaphorical expressions embedded in review texts. Evaluation on an independent test set indicates a classification accuracy of 89.2%, aligning with balanced and stable scoring performance. Robustness is further strengthened by introducing an equal-weight alternative strategy and conducting stability checks to indicate the consistency of model outputs across weighting assumptions. At the scenario interpretation level, we combine grounded-theory coding with semantic network analysis to establish a three-tier spatial analysis framework—macro (landscape pattern/hydro-topological patterns), meso (architectural interface), and micro (teaching scenes/pedagogical scenarios)—and incorporate an interpretive stakeholder typology (tourists, residents, parents, and professional groups) to systematically identify and quantify key features shaping public spatial perception. Findings show that, at the macro level, naturally integrated scenarios—such as “campus–farmland integration” and “mountain–water embeddedness”—exhibit high affective association, aligning with the “mountain-water-field-village” spatial sequence logic and suggesting broad public endorsement of ecological campus concepts, whereas vernacular settlement-pattern scenarios receive relatively low attention due to cognitive discontinuities. At the meso level, innovative corridor strategies (e.g., framed vistas and expanded corridor spaces) strengthen the building–nature interaction and suggest latent value in stimulating exploratory spatial experience. At the micro level, place-based practice-oriented teaching scenes (e.g., intangible cultural heritage handcraft and creative workshops) achieve higher scores, aligning with the compatibility of vernacular education’s “differential esthetics,” while urban convergence-oriented interdisciplinary curriculum scenes suggest an interpretive gap relative to public expectations. These results indicate an embedded relationship between public perception and regional spatial features, which is further shaped by a multi-actor governance process—characterized by “Government + Influencers + Field Study”—that mediates how rural educational spaces are produced, communicated, and interpreted in digital environments. The study’s innovative value lies in integrating sociological theories (e.g., embeddedness) with deep learning techniques to fill the regional and multi-actor perspective gap in rural campus POE and to promote a methodological shift from “experience-based induction” toward a “data-theory” dual-drive model. The findings provide inferential evidence for rural campus renewal and optimization; the methodological pipeline is transferable to small-scale rural primary schools with media exposure and salient regional ecological characteristics, and it offers a new pathway for incorporating digital media-driven public perception feedback into planning and design practice. The research methodology of this study consists of four sequential stages, which are implemented in a systematic and progressive manner: First, data collection was conducted: Python and the Octopus Collector were used to crawl online comment data related to Fuwen Township Central Primary School, strictly complying with the user agreements of the Douyin, Dianping, and Xiaohongshu platforms. Second, semantic preprocessing was performed: The evaluation content was segmented to generate word frequency statistics and semantic networks; qualitative analysis was conducted using Origin software, and quantitative translation was realized via Sankey diagrams. Third, spatial scene coding was carried out: Combined with a spatial characteristic identification system, a macro–meso–micro three-tier classification system for spatial scene characteristics was constructed to encode and quantitatively express the textual content. Finally, sentiment quantification and correlation analysis was implemented: A deep learning model based on the Transformer framework was employed to perform sentiment quantification scoring for each comment; Sankey diagrams were used to quantitatively correlate spatial scenes with sentiment tendencies, thereby exploring the public’s perceptual associations with the architectural spatial environment of rural campuses. Full article

Shared energy storage provides a promising solution for the operation of microgrid clusters. This paper explores a hybrid electric–hydrogen shared energy storage model within microgrid clusters, aiming for clean energy generation and economical energy supply despite renewable energy’s unpredictability and complex stakeholder interactions. First, the proposed method features a shared energy storage operator that hosts electric storage and power-to-gas, enabling multi-microgrids energy sharing. To address market dynamics, a hybrid game theory approach using Nash bargaining and Stackelberg games is employed to manage interactions among the shared energy storage operator, microgrid operators, and internal end-users, while accounting for their differing interests. Second, to address uncertainty in renewable energy output, a distributionally robust optimization model is implemented with conditional value at risk, focusing on risk in extreme scenarios. The Adaptive Alternating Direction Method of Multipliers algorithm and Karush–Kuhn–Tucker conditions are used to solve the optimal decision scheme for each entity. Finally, a case study is used to verify the model’s effectiveness. Simulation results show that hybrid electric–hydrogen energy sharing improves resource utilization, leading to significant revenue increases for microgrids and higher profitability for shared energy storage operator. The game-theory-based approach ensures equitable revenue distribution and a 9.86% increase in coalition revenue. It provides a flexible approach to balance economic efficiency and system robustness by allowing decision-makers to adjust risk preference parameters and use historical sample data for informed decision-making. Full article

Sequencing technologies have reshaped the study of the subgingival microbiome, but selecting the appropriate method remains challenging because of differences in resolution, cost, host DNA contamination, and computational complexity. This review compares 16S rRNA sequencing, full-length 16S, shotgun metagenomics, and metatranscriptomics with respect to taxonomic resolution, functional output, sample requirements, and analytical limitations. Key practical issues, including low microbial biomass, contamination control, and the choice of appropriate bioinformatic tools, are emphasized to help researchers avoid common pitfalls. A decision-making framework is provided to link study goals to suitable sequencing methods while outlining realistic budget and sample-handling constraints. The review concludes with recommendations for integrating sequencing with complementary techniques to improve the accuracy, reproducibility, and clinical relevance of periodontal microbiome studies. Full article

Achieving sustainable rice production is essential for food security, rural livelihoods, and the resilience of rice-based food systems that supply raw materials to the food processing industry. Improving technical efficiency (TE) at the farm level is a key pathway to reduce resource use and environmental pressures per unit of output while ensuring a stable supply of high-quality rice for downstream processing and value-added products. Drawing on micro-survey data collected in 2021–2022 from 455 rice farmers selected through a multi-stage sampling strategy in Jiangsu Province, China, this study investigates how agricultural productive services (APSs) affect stage-specific technical efficiency along the production process and discusses the implications for sustainable rice production and the rice-based food industry. We apply a stochastic frontier production function to estimate overall and stage-specific TE and examine the effects of different APS combinations for land preparation, sowing, fertilization, pest control, and harvesting. The results show that overall participation in APSs significantly improves rice farmers’ TE. Stage-specific analysis reveals that APSs in land preparation, sowing, and harvesting are associated with higher TE, supporting more sustainable use of machinery and labor, while APSs for fertilization and pesticide application do not consistently improve TE and may reflect potential overuse of chemical inputs. Multi-stage service combinations that include both production and pest-control operations can further enhance TE. These findings suggest that well-designed APSs can contribute to sustainable intensification and low-carbon transformation of rice production, thereby strengthening the sustainability of rice-based food systems. Policy interventions should guide APS providers and farmers toward integrated, precision-oriented, and environmentally friendly service packages that support both farm-level efficiency and the sustainability goals of the broader food industry. Full article

Deep neural networks (DNNs) represent critical intellectual property that model owners urgently need to protect. With the increasing value of models, malicious attackers increasingly attempt to extract model functionality through techniques such as fine-tuning, distillation, and pruning. Model fingerprinting has emerged as a mainstream protection strategy. However, existing fingerprinting methods either exhibit vulnerability to model modifications due to reliance on decision boundary features or require prohibitively large query budgets for accurate verification. This paper proposes ProbeSpec, which captures model fingerprints through dynamic behavioral analysis rather than static output matching. We discover that a model’s response patterns under multi-level perturbations form a unique “behavioral spectrum”, originating from implicit decision mechanisms learned during training and preserved even after various attacks. ProbeSpec employs three complementary probe types to elicit this characteristic and leverages DCT frequency-domain transformation for efficient fingerprint extraction. Extensive experiments show that ProbeSpec achieves 100% detection rate in the majority of attack scenarios, with an overall accuracy exceeding 95% across all tested architectures. Meanwhile, it effectively distinguishes independently trained models and requires only 80 probe samples for fingerprint extraction. Full article

Background/Objectives: Synthetic artificial intelligence (AI) is increasingly used in cardiovascular medicine to generate realistic clinical data from limited samples while preserving patient privacy. Despite its promise, concerns remain regarding the clinical reliability of synthetic datasets, which hampers their integration into routine practice. This article introduces the TIMA method (Team-Implementation Multidisciplinary Approach), designed to involve clinicians directly in every phase of synthetic data development. The objective of this work is to describe the TIMA framework and to illustrate how structured clinician–data scientist collaboration can enhance the clinical robustness and plausibility of synthetic AI outputs. Methods: The TIMA approach structures the synthetic data generation workflow around continuous interaction between clinicians and data scientists. Cardiologists define clinical constraints, verify inter-variable relationships, and assess the coherence and plausibility of generated records. The framework is illustrated through multiple cardiology use cases, including atrial fibrillation risk prediction and surgical mortality estimation in infective endocarditis, to demonstrate its adaptability across different clinical contexts. Each phase includes iterative validation steps aimed at ensuring alignment with established clinical knowledge rather than reporting quantitative performance outcomes. Results: Application of the TIMA framework supported the development of synthetic datasets that adhered more closely to clinical logic and domain-specific constraints. Clinician–data scientist collaboration enabled early detection of implausible variable interactions, improved interpretability of synthetic data patterns, and enhanced internal consistency across different cardiology-oriented scenarios. Conclusions: TIMA represents a scalable and replicable methodological model for integrating synthetic AI into cardiology by embedding clinical expertise throughout the data generation process. Its structured, multidisciplinary workflow supports the production of synthetic data that is not only statistically coherent but also clinically meaningful, thereby strengthening trust and reliability in AI-assisted cardiovascular research. Full article

Dry gas reservoirs play a pivotal transitional role in meeting the net-zero target worldwide. Accurate modelling and simulation of this energy source require fast and reliable prediction of the gas compressibility factor (Z-factor). The experimental measurements of Z-factor are the most reliable source; however, they are expensive and time-consuming. This makes developing accurate predictive models essential. Traditional methods, such as empirical correlations and Equations of States (EoSs), often lack accuracy and computational efficiency. This study aims to address these limitations by leveraging the predictive power of machine learning (ML) techniques. Hence in this study three ML models of Artificial Neural Network (ANN), Group Method of Data Handling (GMDH), and Genetic Programming (GP) were developed. These models were trained on a comprehensive dataset comprising 1079 samples where pseudo-reduced pressure (Ppr) and pseudo-reduced temperature (Tpr) served as input and experimentally measured Z-factors as output. The performance of the developed ML models was benchmarked against two cubic EoSs of Peng–Robinson (PR) and van der Waals (vdW), and two semi-empirical correlations of Dranchuk-Abou-Kassem (DAK) and Hall and Yarborough (HY), and recent developed ML based models, using statistical metrics of Mean Squared Error (MSE), coefficient of determination (R²), and Average Absolute Relative Deviation Percentage (AARD%). The proposed ANN model reduces average prediction error by approximately 70% relative to the PR equation of state and by over 35% compared with the DAK correlation, while maintaining robust performance across the full Ppr and Tpr of dry gas systems. Additionally paired t-tests and Wilcoxon signed-rank tests performed on the ML results confirmed that the ANN model achieved statistically significant improvements over the other models. Moreover, two physical equations using the white-box models of GMDH and GP were proposed as a function of Ppr and Tpr for prediction of the dry gas Z-factor. The sensitivity analysis of the data shows that the Ppr has the highest positive effect of 88% on Z-factor while Tpr has a moderate effect of 12%. This study presents the first unified, statistically validated comparison of ANN, GMDH, and GP models for accurate and interpretable Z-factor prediction. The developed models can be used as an alternative tool to bridge the limitation of cubic EoSs and limited accuracy and applicability of empirical models. Full article

Structural hyperparameter selection (HPS) in neural network (NN) regression faces two critical, computationally expensive barriers: the mandatory splitting of datasets for validation, which significantly impairs sample efficiency, and the inability of conventional metrics (like Data MSE) to decouple true modeling error from detrimental output noise, leading to suboptimal architectural complexity and overfitting. To resolve these systemic limitations, we propose the Reconstruction Error Minimization for Hyperparameter Selection (REM-HPS) framework, a novel, non-Bayesian approach grounded in statistical learning theory. REM-HPS fundamentally shifts the optimization objective by minimizing the Reconstruction Mean Squared Error (MSE), which precisely isolates and measures the model’s intrinsic ability to recover the underlying noise-free function. Since this target error is typically inaccessible, the framework employs the observable Data MSE (validation error) to construct a reliable, probabilistic estimate, yielding a deterministic and noise-aware selection criterion. REM-HPS utilizes a deterministic structural hyperparameter selection criterion that removes randomness due to validation data splitting, while remaining compatible with standard stochastic training procedures. This strategy allows for the use of the entire dataset for training, eliminating the need for explicit data splitting or the introduction of tuning-intensive regularization hyperparameters. Rigorous empirical validation demonstrates that REM-HPS consistently selects significantly more compact architectures (minimal complexity) while achieving superior generalizability and estimation accuracy, particularly across varied Signal-to-Noise Ratios and data regimes. By providing an efficient and optimal selection metric, REM-HPS offers a transformative, resource-efficient alternative to structural HPS in modern data-driven systems. Full article

The accurate interpretation of land cover changes in multi-temporal satellite imagery is critical for Earth observation. However, existing methods typically yield static outputs—such as binary masks or fixed captions—lacking interactivity and user guidance. To address this limitation, we introduce remote sensing image change analysis (RSICA), a novel paradigm that enables the instruction-guided, multi-turn exploration of temporal differences in bi-temporal images through visual question answering. To realize RSICA, we propose DeltaVLM, a vision language model specifically designed for interactive change understanding. DeltaVLM comprises three key components: (1) a fine-tuned bi-temporal vision encoder that independently extracts semantic features from each image in the input pair; (2) a visual difference perception module with a cross-semantic relation measuring (CSRM) mechanism to interpret changes; and (3) an instruction-guided Q-former that selects query-relevant change features and aligns them with a frozen large language model to generate context-aware responses. We also present ChangeChat-105k, a large-scale instruction-following dataset containing over 105k diverse samples. Extensive experiments show that DeltaVLM achieves state-of-the-art performance in both single-turn captioning and multi-turn interactive change analysis, surpassing both general multimodal models and specialized remote sensing vision language models. Full article

Underwater vehicles performing sampling tasks often encounter significant buoyancy variations due to payload adjustments and environmental changes, which severely challenge the stability and accuracy of controllers. To address this issue, this paper proposes a hybrid control framework that integrates Proximal Policy Optimization (PPO) with adaptive PID tuning. The framework employs PPO to dynamically adjust PID parameters online while incorporating output saturation, stepwise quantization, and dead zone filtering to ensure control safety and actuator longevity. A dual-error state representation—combining instantaneous error and its derivative—along with actuator command buffering is introduced to compensate for system lag and inertia. Comparative simulations and experimental tests demonstrate that the proposed method achieves faster convergence, lower steady-state error, and smoother control signals compared to both conventional PID and pure PPO-based control. The framework is validated through pool tests and field trials, confirming its robustness under realistic hydrodynamic disturbances. This work provides a practical and safe solution for adaptive depth control of sampling-capable AUVs operating in dynamic underwater environments. Full article

The accuracy of online parameter identification for permanent magnet synchronous motors (PMSMs) is constrained by discrete model errors, rank deficiency in the steady-state identification matrix, and voltage deviations resulting from inverter nonlinearities. This paper proposes a multi-parameter identification method acting as a high-precision virtual sensor, based on Zero-Order Hold (ZOH) discretization and an inverter nonlinear voltage compensation scheme utilizing a dual-sampling strategy. First, a discrete model of the PMSM, accounting for rotor position variations within the control period, is established using the ZOH discretization method. Compared with the forward Euler discretization method, this approach effectively minimizes discretization model errors, especially under high-speed operating conditions where rotor position variations are significant. Second, the rank deficiency problem of the steady-state identification matrix is overcome by combining d-axis small-signal injection with a dual-sampling strategy. Furthermore, the Forgetting Factor Recursive Least Squares (FFRLS) algorithm is introduced to achieve online multi-parameter identification. Finally, the influence mechanisms of the dead-time effect, power switch voltage drop, and turn-on delay on the output voltage are analyzed. Consequently, an inverter nonlinear voltage compensation strategy tailored for the dual-sampling mode is proposed. Experimental results demonstrate that the proposed method significantly enhances parameter identification accuracy across the entire speed range. Specifically, under high-speed conditions, the identification errors for resistance, inductance, and flux linkage are maintained within 5.47%, 4.05%, and 2.46%, respectively. Full article

Accurately assessing battery health across mixed datasets remains a challenge due to differences in chemistry, format, and usage history. This study presents a reproducible framework for preparing battery cycling data using incremental capacity analysis (ICA), with the aim of supporting machine learning (ML) workflows across both first-life and second-life battery datasets. The methodology includes IC curve generation, feature extraction, encoding and scaling, feature reduction, and unsupervised learning exploration. A two-tiered outlier detection system was introduced during preprocessing to flag edge-case samples. Two clustering algorithms, K-means and HDBSCAN, were applied to the engineered feature space to explore patterns in the IC feature space. K-means revealed broad health-related groupings with overlapping boundaries, while HDBSCAN identified finer clusters and flagged additional ambiguous samples as noise. To support interpretation, PCA and t-SNE were used to visualise the feature space in reduced dimensions. Rather than using clustering as a classification tool, the resulting cluster and noise labels are proposed as structure-aware meta-features for supervised learning. The framework accommodates heterogeneous battery datasets and addresses the challenges of integrating data from mixed sources with varying histories and characteristics. These outputs provide a structured foundation for future supervised classification of battery state of health. Full article

This paper presents a novel design for a reconfigurable CDC as a multiplexed sensor fusion that converts three analog signals into digital output bits with different resolutions. The proposed reconfigurable CDC design uses the SAR technique that introduces a small chip area and low power consumption. The proposed novel CDC introduces reconfigurability by using a switching capacitive DAC that solves the problem of converting more than one analog signal with a single converter to a different number of output bits, giving better performance than previous designs. In this paper, three analog signals are used (as a case study) in a weather station to be converted. These signals are temperature, pressure, and humidity that are sensed using the BME-280 Bosch sensor. All CDC specifications are measured for each reconfigured number of output bits. The used supply voltage is 1.0 V, and the sampling frequency is 100 kHz. The 12-bit resolution consumes 2.54 µW, ENOB is 11.47 bits, and SNR equals 73.4 dB. The 8-bit resolution consumes 1.7 µW, ENOB is 7.39 bits, and SNR equals 46.24 dB. The 4-bit resolution consumes 0.68 µW, ENOB is 3.58 bits, and SNR equals 23.45 dB. The total chip area is 0.18 mm². Full article

Although innovation is widely recognized as an important driving force for enterprise development, there may not be a simple linear relationship between innovation and enterprise value. An in-depth investigation of the relationship between enterprise innovation and enterprise value is of great significance for the development of new energy vehicle (NEV) enterprises and the industry. Utilizing the sample of Chinese NEV listed companies from 2012 to 2022, this study empirically examines the effects of enterprise innovation on enterprise value from two perspectives: innovation input and innovation output. The results indicate that enterprise innovation does not necessarily promote the growth of enterprise value in all cases. Innovation input exhibits a U-shaped effect on enterprise value, while innovation output has a linear positive impact. And the operational efficiency of enterprises plays a partial mediating role. Furthermore, we explore the effects of internal and external environments on the relationship. High internal control costs weaken the U-shaped relationship of innovation input and reduce the positive impact of innovation output on value. In contrast, greater market competition attenuates the U-shaped effect of innovation input but strengthens the positive effect of innovation output on enterprise value. These findings highlight the contingent nature of innovation value creation in complex industrial systems and provide insights for strategy and policy in the NEV industry. Full article