The dynamic positioning of the mass-switched unmanned vessel is remotely controlled via networks from the land-based control station. However, the modal and location information transmitted over networks is susceptible to tampering by deception attacks. Existing attack detection methods have significant limitations in accuracy, as they cannot effectively distinguish between modal tampering caused by deception attacks and modal switching resulting from mass variation. To address this issue, this paper proposes a bumpless detection approach for deception attack based on dynamic residual. By suppressing amplitude bumps caused by switching and data updates, it effectively distinguishes genuine attack behaviors. Simultaneously, the dynamic residual not only more accurately reflects the actual observed state but also effectively suppresses the fluctuating disturbances introduced by static residuals. Finally, this paper designs a resilient learning observer. This observer's resilient storage mechanism and resilient coefficient can proactively discard historical data before the current switching instant and incorporate attack detection results as the basis for data selection, while also enabling dynamic adjustment. The effectiveness of the proposed bumpless deception attack detection method is validated through simulation experiments
The energy consumption from vessel waiting and handling at ports needs to be addressed, giving rise to policies such as Shore Power Connection (SPC) and Just In Time (JIT). However, converting these policies into operational decision metrics is challenging. This study fills that gap by embedding key indicators, such as Berth Occupancy Rate (BOR), Shore Power Utilization Ratio (SPUR), Quay Crane Utilization Ratio (QCUR), and Average Vessel Completion Rate (AVCR), into a novel optimization model for the integrated Berth Allocation and Quay Crane Scheduling Problem (BAQCSP). The model incorporates a dynamic prediction mechanism to estimate quay crane handling times, accounting for real-world variables such as tidal conditions and container stowage positions. Experimental results demonstrate that the proposed model demonstrates superior performance in resource utilization compared to sequential scheduling, single-objective optimization, and fixed quay crane handling time strategies. Additional ablation studies quantify the individual impact of each sustainability indicator on scheduling outcomes. The findings provide port authorities with actionable evidence to operationalize sustainability goals into enforceable strategies.
In the selective catalytic oxidation of ammonia (NH3-SCO), the construction of a binary synergy is considered a highly promising strategy for balancing low-temperature activity and N2 selectivity. However, effectively suppressing the decline in N2 selectivity under high-temperature conditions remains a significant challenge. In this study, we designed and synthesized a Ru catalyst supported on Ti-doped CeO2 for the efficient selective oxidation of NH3 to N2. Ru/Ce9-Ti1 catalyst exhibited a 22.4% higher NH3 conversion rate than Ru/CeO2 at 238 degrees C, and its N2 selectivity increased by 39.8% at 400 degrees C. Furthermore, it maintained over 84.5% N2 selectivity across a broad temperature range of 250-400 degrees C. The characterization results indicated that the Ru/Ce9-Ti1 catalyst exhibited enhanced surface acidity and superior thermal stability relative to Ru/CeO2, which facilitated NH3 adsorption and activation. In situ DRIFTS results revealed that both Ru/CeO2 and Ru/Ce9-Ti1 catalysts followed both the internal selective catalytic reduction (i-SCR) mechanism and the imide (-NH) mechanism in the NH3-SCO reaction. Notably, the introduction of Ti shifted the reaction toward the i-SCR pathway, effectively promoting the reduction of surface nitrate species and significantly improving N2 selectivity. This work provided valuable insights for the design of high-performance heterogeneous catalysts.
Background Bio-oil is a renewable resource derived from biomass. It is widely used as a fuel for the chemical industry, but its application potential in the field of green energy remains insufficiently exploited. Here, we present a sustainable strategy to upcycle pine-derived bio-oil into functional carbon nanodots for advanced photocatalysis and clean energy production. Methods Nitrogen-doped carbon nanodots (PNCDs) were synthesized from pine bio-oil via a DMF-assisted hydrothermal process. After calcination, PNCDs were integrated with g-C₃N₄ nanosheets to construct a PNCDs/g-C₃N₄ photocatalyst for hydrogen evolution via water splitting. Significant findings Experimental data indicate that a hydrogen production rate of 2.43 mmol/g/h can be achieved by adding PNCDs solution. And the catalyst notably attains a high apparent quantum efficiency (AQE) of 24.4%. By introducing PNCDs, the bandgap structure of g-C₃N₄ was successfully modulated, thereby significantly enhancing its light absorption capacity. Furthermore, PNCDs act as an electron reservoir, capturing electrons released from the conduction band of g-C₃N₄ to achieve spatial separation of photo-generated electrons and holes. This study aims to explore the vast potential of bio-oil in converting into high-quality products, while enhancing the photocatalytic hydrogen production efficiency of carbon nitride under visible light.
We study ex-ante fleet deployment and slot allocation on fixed weekly liner networks. Before O–D demand, freight-rate coefficients, and deliverable capacity are realized, the carrier chooses vessel deployment, cadence, and baseline slot allocation; the fixed plan is evaluated for profit and reliability, with no scenario-dependent re-optimization. The central idea is to let one deployment-allocation plan carry three linked roles: it protects accepted O–D demand, prices accepted flows through offline conservative revenue coefficients, and enforces correlated deliverable-capacity safeguards over shared disruption regions. We formulate a mixed-integer conic model that links these three uncertainty blocks through common decisions, not one estimated joint ambiguity set. Demand and deliverable capacity receive moment-based distributionally robust safeguards: Cantelli lower-confidence caps and a Boole-based second-order cone approximation for protected capacity reliability. Freight rates enter only as fixed offline conservative revenue coefficients, not dynamic prices, recourse decisions, or an in-model pricing-optimization block. In this formulation, the fleet, cadence, and voyage decisions directly set how much correlated capacity loss each shared disruption region is exposed to. To keep the conic model tractable at network scale, a PCA-based outer-approximation layer reduces the covariance dimension in the capacity certificates, retaining the leading directions and bounding the residual tail. On LINER-LIB-derived instances, out-of-sample diagnostics, component ablations, a 21-point PCA threshold sweep, and a public-data calibration and holdout plausibility check show how the safeguards change deployment decisions and when PCA-OA improves tractability. The evidence supports internal consistency and plausibility, not operational validation on proprietary carrier data.
