Search Results (4,676)

Sustainable management of biodegradable waste is a key component of the circular economy and resource recovery strategies. Composting is a complex biological process whose efficiency depends on numerous operational and physicochemical factors. Under industrial conditions, continuous laboratory monitoring of waste properties is often limited by time and cost constraints, necessitating reliable predictive tools to support process management. This study investigates the use of artificial neural networks (ANNs), decision trees (C&RT), and principal component analysis (PCA) for optimizing the composting of biodegradable waste under industrial-scale conditions. The research was conducted at a full-scale mechanical–biological treatment facility in Poland processing both the organic fraction mechanically derived from mixed municipal waste and separately collected biowaste. A dataset containing 23 records was developed from operational parameters (airflow, water addition, turning frequency, and process duration) and physicochemical properties of composted waste, including moisture content (MC), loss on ignition (LOI), total organic carbon (TOC), respiration activity (AT4), and higher heating value (HHV). The best-performing neural model achieved a predictive accuracy of 0.999 (coefficient of determination R² in the test set). For each of the neural networks, goodness of fit indices were also determined: MAE and RMSE. PCA confirmed strong relationships among key waste properties, while decision tree analysis identified airflow as the dominant operational factor affecting MC, LOI, and TOC, whereas turning frequency had the strongest influence on AT4. The results demonstrate that machine learning tools can effectively support industrial composting optimization by predicting operational parameters required to achieve desired waste stabilization characteristics, providing practical decision-support solutions for composting plant operators. It is recommended to implement single-output MLP models for dynamic, real-time process control and C&RT rules as emergency procedures. This study aligns with circular economy principles and the Sustainable Development Goals by demonstrating the potential of artificial intelligence to enhance sustainable biodegradable waste management, resource recovery, and industrial composting performance. Full article

This study proposes a surrogate model, Graph-Structured Physics-Informed Deep Operator Network (GS-PI-DeepONet), to simulate two-dimensional hydrodynamics in a tidal river network. The model was coupled with Graph Convolutional Networks (GCNs) for spatial feature extraction and Long Short-Term Memory (LSTM) network for temporal prediction, with 2D shallow-water equations (2D SWEs) embedded as physical constraints. To handle complex river network topologies, a mapping mechanism was proposed to transform discrete irregular boundaries into differentiable neural network constraints. A dynamic weighting strategy was developed to improve model training efficiency. GS-PI-DeepONet was applied to a river network within the Pearl River Basin in Zhuhai. Trained on high-fidelity Delft3D data, it achieved precise flow field reconstruction and millisecond-level extrapolation predictions, outperforming traditional data-driven models. The model can be a valuable tool for real-time hydrodynamic simulations and flood management strategies in tidal river networks. Full article

Leucine-rich repeat kinase 2 (LRRK2) has emerged as a central molecular node linking genetic risk, membrane trafficking, lysosomal homeostasis, and immune signalling in Parkinson’s disease (PD). Rather than functioning as a conventional protein kinase, LRRK2 operates as a conformationally regulated, Rab-directed signalling machine whose activity is governed by long-range interdomain communication, membrane recruitment, and cooperative interactions with small GTPases. Converging advances in cryo-electron microscopy, quantitative phosphoproteomics, and human genetics indicate that pathogenic mutations, lysosomal stress, and pharmacological inhibitors do not simply alter catalytic output, but reshape the conformational landscape of LRRK2, biasing it toward distinct structural states with divergent cellular consequences. A defining feature of this system is the selective phosphorylation of Rab GTPases at low stoichiometry—most prominently Rab8 and Rab10—yet with disproportionate functional impact on vesicle trafficking, ciliogenesis, autophagy, and organelle positioning. The identification of Rab-directed phosphatases, particularly PPM1H, further establishes that LRRK2 signalling is governed by a dynamically balanced kinase–phosphatase circuit operating in space and time. These observations, together with emerging evidence linking LRRK2 activation to lysosomal damage and immune pathways, support a unifying hypothesis: PD-associated LRRK2 dysfunction arises from maladaptive stabilization of specific conformational and spatial states within a membrane-responsive signalling network, leading to persistent misregulation of Rab-dependent trafficking and organelle homeostasis, rather than from kinase hyperactivity alone. In this review, we integrate structural, biochemical, and cellular evidence to advance this framework and discuss its implications for disease mechanisms and therapy. We highlight key unresolved challenges—including conformation-selective drug targeting, spatial control of Rab phosphorylation, and context-dependent immune–neuronal crosstalk—and propose that restoring physiological regulation of LRRK2, rather than simply inhibiting its activity, will be essential for achieving mechanism-based disease modification in Parkinson’s disease. Full article

Light Fidelity (LiFi) has emerged as a promising wireless communication technology for enabling indoor networks in future sixth-generation (6G) systems. It can offer very high data rates, low electromagnetic interference, elevated spatial reuse and inherent physical layer security advantages. In LiFi-based indoor deployments, however, the characteristically small coverage footprint of each AP can produce frequent handovers when users move across the indoor environment. Conventional key management approaches adopt static key lifetime policies that expire session keys after a fixed duration regardless of user movement conditions. This creates a fundamental inefficiency where static policies can generate key refresh operations that are not required. Further, an additional signaling overhead can happen when users are stationary or leave session keys active for comparatively long periods under high mobility conditions. This paper proposes a lightweight mobility-aware adaptive session key lifetime control mechanism that dynamically adjusts session key validity duration. The mechanism introduces a three-component mobility metric that aggregates user speed, handover frequency and AP residence time which is used to compute a normalized mobility risk score. The mobility score drives an adaptive lifetime formula that contracts or extends key validity according to mobility-associated exposure conditions. The proposed mechanism operates entirely within the post-authentication session phase and does not introduce new cryptographic primitives, authentication protocols or key generation algorithms. Compared with fixed session key lifetime policies of 120 s, 300 s and 600 s, the Adaptive Key Lifetime Control Engine (AKLCE) dynamically adapts the session key lifetime according to user mobility, reducing refresh operations and signaling overhead under low mobility while progressively shortening the scheduled key lifetime under higher mobility to limit key exposure. This adaptive behavior enables AKLCE to provide mobility-dependent adaptation of session key lifetime that enables a flexible trade-off between key exposure duration and signaling overhead. Full article

The transition from early non-standalone 5G deployments to 5G Standalone and, more recently, 5G-Advanced has turned mobile networks into flexible, programmable infrastructures capable of supporting private, industrial, and research-oriented deployments for the development of beyond-5G applications and architectures. Evaluating these networks’ capabilities, however, remains challenging because commercial platforms often provide limited access to internal interfaces, radio parameters, and network measurements. This paper presents an open-source private 5G SA testbed for beyond-5G application validations built using Open5GS, srsRAN, Ettus USRP N310 software-defined radio, programmable SIM cards, and commercial 5G customer-premise equipment. The platform is deployed in a semi-anechoic chamber. End-to-end operation is validated through subscriber registration, authentication, PDU session establishment, and external data connectivity. The performance of the implemented 5G network is evaluated using throughput, block error rate, modulation and coding scheme, and gNB trace logs. Unlike previous open-source 5G testbeds that primarily use RF waveguides, individual network components, or a limited set of radio configurations, the proposed platform combines COTS SIM-based UE operation with a controlled over-the-air evaluation of FDD/TDD and multiple antenna configurations and correlates application-level throughput with internal gNB radio metrics. For FDD downlink operation, the average throughput increased by approximately 74% from 1 × 1 to 2 × 2 and by a further 57% from 2 × 2 to 4 × 4, although the additional peak-throughput gain from 2 × 2 to 4 × 4 remained limited. The platform provides a reproducible environment for validating beyond-5G mechanisms, comparing network configurations, and studying the behavior of future open-source 5G SA systems under controlled conditions. Full article

Emerging industrial agglomeration drives green transformation through knowledge spillovers and economies of scale. However, its effects exhibit pronounced nonlinearity and heterogeneity, shaped by spatial externalities and development stages. This paper investigates 19 Chinese urban agglomerations over the period 2014 to 2023. Kernel density estimation based on enterprise-level Point-of-Interest (POI) data is used to characterize spatial agglomeration patterns across eight emerging sectors. A two-stage dynamic network super-efficiency SBM model decomposes Green Transition Efficiency (GTE) into resource utilization and pollution control sub-stages. An XGBoost-SHAP-GEO analytical framework, combined with partial dependence analysis, then identifies nonlinear driving mechanisms. The main findings are as follows: First, emerging industrial agglomeration intensifies and polarizes toward the eastern coast, whereas GTE displays a “high-west, low-east” pattern. This produces a significant spatial mismatch, rooted in the near-saturation of environmental carrying capacity in eastern regions, where congestion effects exceed knowledge spillover dividends. Second, geographic characteristics constitute the primary factor shaping GTE and operate through nonlinear interactions with industrial agglomeration and R&D investment. Notably, their moderation direction is reversible, suggesting that geographic endowments should be understood as “conditional assets” rather than fixed advantages. Third, nonlinear patterns across sectors are highly heterogeneous. The bio-industry is the only sector to achieve a J-shaped positive breakthrough. Information technology and new materials exhibit persistent inhibition, while related services display an extremely narrow threshold window with the deepest negative reversal. Thus, “moderate agglomeration” is a multidimensional concept that shifts dynamically with industry type and regional endowment. Fourth, driving mechanisms display stage-dependent evolution. The incubation stage relies on natural endowments and basic industrial pull, with the green bottleneck residing in resource utilization efficiency. The growth stage faces multiple tensions from coexisting positive and negative effects. The optimization stage shifts toward R&D innovation and industrial greening, marking a qualitative transformation from MAR externalities to Jacobs externalities. In addition, the non-significant linear coefficient in the 2SLS instrumental variable test is consistent with the inverted U-shaped nonlinear finding, further validating the necessity of a nonlinear analytical framework. These findings provide differentiated governance evidence for balancing industrial agglomeration with green sustainable development. Full article

DC bias will cause the magnetization working point of the transformer core to shift and cause local saturation, and generate abnormal vibration through the magnetostrictive effect, which threatens the safe operation of the transformer. Aiming at the problem that the time–frequency characteristics of transformer vibration signals under DC bias are complex and the adjacent bias levels are difficult to distinguish, this paper proposes a 2DWT-CNN-Transformer diagnostic method that combines two-dimensional discrete wavelet transform, a convolutional neural network, and Transformer Encoder. Firstly, the multi-physical-field finite element model of three-phase three-column transformer is established, and the L0–L5 six-class DC bias dataset is constructed. Secondly, the one-dimensional vibration signal is reconstructed into a two-dimensional matrix, and the multi-subband time–frequency features of LL, LH, HL, and HH are extracted by two-dimensional discrete wavelet transform. The local texture features are extracted by the CNN, and the multi-head self-attention mechanism of Transformer Encoder is introduced to establish the global dependence and enhance the discrimination ability of adjacent bias levels. Compared with the traditional time–frequency-feature deep learning model, the proposed method achieves higher accuracy, especially in the high-noise environment of 15 dB, where it can still maintain accuracy of 96.23%. The visualization results further show that the model can form a more compact intra-class aggregation and a clearer inter-class boundary. This also provides an effective solution for the identification and evaluation of transformer DC bias states based on vibration signals in complex environments in the future. Full article

Engineering schematics, such as electrical, mechanical, piping and instrumentation diagrams, record how industrial plants are built and operated, but most of them survive only as images or scanned sheets that software cannot read. Digitizing them by hand is slow and error-prone: an expert must find and classify hundreds of symbols, read dense technical text, and work out which label belongs to which component. Progress with learning-based methods has been held back on two fronts at once. There are almost no annotations that connect a text label to its symbol, and the drawings themselves are usually confidential, so even unlabeled sheets rarely reach the public domain. We address this with a system that turns a drawing into a structured, queryable graph: it detects and classifies the graphical components with an object detector, recovers the technical text, and then resolves which label belongs to which component. Our contributions are threefold: (i) the first at-scale dataset of manually annotated text-to-symbol links for industrial schematics; (ii) a complete, deployable digitization system combining tiled detection with sliced inference, off-the-shelf OCR, and a text-to-symbol association stage; and (iii) a rigorous, leakage-free benchmark of association methods. Under an observable-only candidate protocol, we find that on logic circuits association is dominated by geometry: a simple pairwise model reaches about 99% top-1 and a graph neural network matches but does not exceed it, whereas the denser P&IDs still benefit from a geometric rule-based chain. Detection reaches an mAP@50 of 0.995 on logic circuits and about 0.91 across the 107-class P&ID taxonomy. The system produces a partial semantic graph; connecting lines and flow direction are not extracted. Full article

Akerlof’s classical theory of the “Market for Lemons,” which conceptualizes adverse selection driven by information asymmetry, established the foundation of information economics. While the traditional model assumes static equilibria among a limited number of agents, analyzing its behavioral dynamics within large-scale, complex network environments remains a highly relevant task in computational social science. This study extends the classical lemon market model into a nonlinear dynamical system on adaptive networks. We mathematically elucidate macro-level social phase transitions—specifically structural exclusion such as scapegoating and collective ostracism—induced by computational cognitive limits, and evaluate optimal intervention strategies to mitigate these systemic failures. Multi-agent simulations utilizing large-scale tensor operations demonstrate that autonomous edge rewiring under incomplete information does not merely result in the uniform displacement of high-quality goods as predicted by static theory. Instead, the network self-organizes into an irreversible structural division: a core group of influential agents monopolizes high-quality information, while marginalized agents are isolated into a peripheral “lemon echo chamber” where only low-quality information circulates. To address this structural pathology under a resource constraint limiting intervention to 10% of the total agents, we evaluated two distinct approaches. The results indicate that providing informational support to influential hubs functions as a trap that exacerbates systemic inequality, superficially elevating the overall market evaluation but permanently fixing the exclusion gap. Conversely, the forced maintenance and protection of “weak ties” bridging disconnected clusters constitutes the mathematically optimal solution to dissolve fragmentation, effectively eliminating the price gap and facilitating social inclusion. Furthermore, this study demonstrates that the mechanism of social exclusion exhibits strong hysteresis effects. A distinct tipping point governs the progression toward a fragmented lemon echo chamber. Interventions implemented after crossing this critical threshold fail to restore the system to its baseline state despite identical resource expenditure, confirming the presence of an irreversible phase transition. These findings establish that the collapse dynamics outlined in the classical lemon market serve as a generalized model for explaining contemporary collective ostracism driven by information cascades. Consequently, the analysis highlights the necessity of early intervention prior to critical thresholds and the systemic preservation of structural bypasses rather than post-hoc remediation. Full article

Regenerative aluminum melting serves as a core process in recycled aluminum production. In the regenerative aluminum melting process, the furnace temperature is a key variable which affects product performance and energy costs. The extreme in-furnace temperature necessitates sensors equipped with protective jackets, which increases measurement costs and severely compromises real-time monitoring capability. Accordingly, accurate furnace temperature prediction is highly valuable for regenerative aluminum melting. In regenerative aluminum melting furnaces, periodic burner nozzle commutation and frequent material charging and discharging lead to complex and time-varying operating conditions, posing considerable challenges to high-precision furnace temperature prediction. To address these issues, a condition-adaptive multi-scale convolution-enhanced Transformer (CA-MC-Transformer) model is proposed for furnace temperature prediction. Firstly, an agglomerative hierarchical clustering algorithm based on the weighted dynamic time warping (WDTW) distance is designed to perform unsupervised clustering on historical process data, thereby extracting physically interpretable prior labels for macroscopic operating conditions. Secondly, multi-scale dilated causal convolutions are utilized to capture local dynamic features at diverse temporal resolutions. A soft attention mechanism is further introduced to dynamically assign fusion weights to condition embeddings and local features, enabling condition-adaptive feature reconstruction. Finally, the fused adaptive features are fed into an encoder-only Transformer network to capture the global long-range temporal dependencies and achieve accurate furnace temperature prediction. Comparative experiments conducted on real operational datasets from an aluminum plant verify that the proposed method effectively eliminates the inherent tracking lag of conventional deep learning models, and substantially improves prediction accuracy and anti-noise robustness under complex and variable operating conditions. Full article

This paper proposes a mismatch-index-driven coordinated flexible-step terminal-free distributed model predictive control (DMPC) scheme with an adaptive prediction horizon for asynchronous multi-agent systems (MASs) subject to bounded disturbances. The proposed approach explicitly exploits the inherent symmetry of the undirected communication topology among the agents, which ensures reciprocal information exchange, balanced cooperative interactions, and facilitates the rigorous analysis of consensus under asynchrony. By extending the generalized discrete-time control Lyapunov function (g-dclf) framework to the perturbed setting, we introduce a robust g-dclf together with a robust average decrease constraint that explicitly accounts for the worst-case effect of disturbances. A coordinated self-triggering mechanism, built upon the cost prediction mismatch index and the flexible-step execution strategy, is developed to simultaneously determine the inter-execution times and the number of control steps to be applied in each iteration. In addition, an adaptive shrinking prediction horizon strategy is incorporated to further reduce the computational complexity of the local optimization control problems (OCPs) as the agents approach consensus. The resulting robust flexible-step terminal-free DMPC (RFSTDMPC) algorithm is fully distributed, handles asynchronous communication, and operates without any stability-related terminal constraint. Recursive feasibility of each local OCP and input-to-state stability (ISS) of the overall closed-loop MAS are rigorously established under the symmetric network structure. Simulation results on the consensus problem of three perturbed nonholonomic vehicles demonstrate the effectiveness of the proposed scheme in achieving practical full-state stabilization while significantly alleviating the online computational burden. Full article

Flood disaster risk assessment serves as an important foundation for formulating regional sustainable development strategies. This study establishes a risk assessment model for flood disasters in small and medium-sized rivers based on a theoretical framework integrating Geographic Information Systems (GIS), the Delphi method, and Bayesian networks (Delphi–BNs). An indicator system for the assessment was developed from three dimensions: hazard, vulnerability, and exposure. Hazard is represented by flood inundation area and depth; vulnerability is indicated by population distribution and economic layout; and exposure is reflected by road accessibility. By constructing a GIS-based Bayesian network and employing the Delphi method to create a probabilistic and spatially explicit model, this approach quantifies various sources of uncertainty in the assessment process, enabling a probabilistic expression of risk. Based on the risk assessment results, a stratified, phased flood emergency rescue and personnel transfer plan was established, designating extremely high-risk areas as the core zones for the first phase of personnel transfer, high-risk areas as the second-phase rescue zones, and medium-risk areas as the third-phase rescue zones, thereby providing clear operational guidance for flood emergency response in the basin. The Delphi–BN assessment framework developed in this study focuses on the core elements of flood disaster risk formation, organically integrates expert experience with spatial big data, and effectively overcomes the limitations of traditional assessment methods, such as strong subjectivity, insufficient accuracy, and poor quantification. It achieves a refined and quantitative assessment of flood risk in small and medium-sized river basins in semi-arid regions. The outcomes of this research contribute to a clearer understanding of both the driving mechanisms and the spatial patterns of regional flood risk. Furthermore, they establish a scientifically credible and operationally relevant foundation for key disaster-response decisions, encompassing timely emergency actions, phased population transfers, and the optimized deployment of limited emergency resources. Full article

Multiphase swirling flows in confined spaces induce highly destructive, nonlinear fluid–structure interaction (FSI) vibrations. Understanding the underlying physical mechanisms is critical for ensuring the safety of industrial operations. This study proposes a mesoscopic multiscale framework coupling the Multi-Relaxation Time Lattice Boltzmann Method with Large Eddy Simulation (MRT-LBM-LES) and the Flügge thin-walled cylindrical shell equations to analyze two-way FSI responses. Variational Mode Decomposition (VMD) and the Hilbert–Huang Transform (HHT) are employed to decouple non-stationary broadband excitation signals. The macroscopic topological evolution of the swirling air core—from initial depression to critical breakthrough—is accurately captured. Dynamic mapping reveals a strict time-domain phase-locking mechanism between macroscopic flow instability and microscopic high-frequency structural excitation caused by cavitation bubble collapse. Furthermore, a dimensionless cross-scale energy cascade index is defined to quantify energy transfer. Results indicate that while higher discharge flow rates delay the critical breakthrough, they trigger a delayed, high-amplitude step mutation in the energy cascade, amplifying the global cumulative excitation energy by nearly 75%. Notably, the dominant high-frequency excitation consistently converges within a narrow band of 760 Hz to 790 Hz, independent of flow rate variations. These findings provide a theoretical foundation for unsteady excitation source localization and targeted vibration reduction in complex industrial pipeline networks. Full article

Multi-site water quality sequences are jointly affected by seasonal periodicity, meteorological disturbances, and inter-site differences in the Jianghuai Watershed region. Conventional quality prediction models struggle to simultaneously achieve multi-scale feature extraction, spatial heterogeneity characterization, and prediction uncertainty expression. This study used daily-scale monitoring data on dissolved oxygen (DO), pH, and ammonia nitrogen (NH3N) from 32 monitoring stations within the region in 2025 and proposed the FT-TransONet (Fourier-enhanced Temporal Transformer Operator Network) multi-site probabilistic water quality prediction model. Within a Transformer framework, the model employed a FourierTime learnable frequency-domain filtering module, a GeoBias (Geographic Bias) attention bias mechanism, and a multi-residual ensemble strategy composed of a multilayer perceptron (MLP), a gated recurrent unit (GRU), and a temporal convolutional network (TCN) combined with a mass conservation constraint, thereby achieving both point and interval prediction of key water quality indicators. The results showed that FT-TransONet achieved the lowest Macro_RMSE among all compared methods on the multi-site water quality prediction task. At a prediction horizon of three days, its Macro_RMSE reached 0.2255, which was 21.89% lower than that of the long short-term memory network and 5.57% lower than that of the strongest baseline MC-Dropout. For the three individual indicators, the model attained coefficients of determination of 0.9135, 0.9338, and 0.8660 for dissolved oxygen, pH, and ammonia nitrogen, with corresponding root-mean-square errors of 0.5145, 0.1098, and 0.0523, confirming its potential to characterize the temporal variation in the main water quality indicators. Under multi-step prediction, the error grew gently, with the Macro_RMSE rising only from 0.2255 to 0.2384 as the horizon extended from three to seven days, and the ablation experiments, together with the probabilistic prediction results, further supported the effectiveness of the proposed structural design. Validated on 32 water quality monitoring stations in the Jianghuai Watershed, the method improved multi-site prediction accuracy while accounting for stability and uncertainty quantification, providing a preliminary reference for regional water quality early warning and management. Full article

Digital twins are becoming vital decision-making infrastructures across critical infrastructure sectors such as smart city urban services, transportation, energy, and healthcare. As digital twins become autonomous and gain real-time intervention capabilities, their governance becomes increasingly essential. Yet existing governance mechanisms remain largely procedural: they emphasize compliance without operationalizing the cognitive practices required to question assumptions, detect algorithmic harms, or support legitimate multi-actor deliberation. Drawing on a systematic scoping review, this study synthesizes the literature on digital twin autonomy, algorithmic risks, epistemic foundations, and governance mechanisms. The review reveals a fundamental gap: current governance mechanisms lack institutionalized cognitive capacities for continuous validation, proactive detection of emerging harms, and structured multi-stakeholder deliberation. This gap is corroborated by a limited but growing body of empirical studies on governance in deployed DT settings. To address this gap, the paper proposes the skeptical intelligence framework, developed through design science research. The framework integrates three cognitive functions: validation, detection, and deliberation supported by operational principles, governance artifacts, and distributed accountability roles. The framework advances digital twin governance beyond compliance toward a model rooted in critical epistemology, reflexivity, transparency, and democratic legitimacy. Consistent with design science research, the framework is delivered and evaluated at design time; empirical implementation and outcome evaluation are planned across multi-actor digital twin infrastructure contexts, including smart city governance, port logistics, and energy networks, where DT-mediated decisions redistribute opportunities and risks across heterogeneous stakeholders. Empirical validation in an operational setting is planned as the next phase of this research. Full article