Atmospheric deposition as a pathway for microplastic transport to the marine environment: Temporal variation and environmental factors.

ABSTRACT This study critically explores spatial and temporal variations of water quality in Melaka River Basin from 2019 to 2023 with the application of advanced multivariate statistical techniques: HCA, DA and PCA. HCA has efficiently stratified the monitoring sites as high medium low pollution clusters that has revealed seasonally driven pollutant dispersion patterns particularly intensified during monsoon periods. DA validated these classes by having strong discrimination ability with R2 = 0.694 wherein as coliform, NH3N, COD and turbidity were found as critical indicators. PCA further identified the main pollution drivers as sediment load and microbial input under wet conditions, with organic enrichment and salinity stress during drier periods. More importantly, forward stepwise DA revealed that even fewer variables comprising only salinity, temperature, COD, and NH3N could still maintain high predictive power for practical use in simplified versions of cost-effective water monitoring strategies. Based on study findings, this multivariate demonstrates powerful tools towards breaking down complex environmental data sets to identify specific remediation activities to reveal seasonal and spatial pollution dynamics within a river system of such historical and ecological importance that advocate evidence-based water governance and targeted policy intervention.

Purpose This study aims to enhance the DAB-LLC bidirectional isolated converter’s adaptability to wide input voltage ranges in marine hybrid power systems. It addresses efficiency losses and voltage stress caused by conventional topologies, proposing dynamic full/half-bridge switching in LLC circuits and extended phase-shift modulation in DAB circuits to achieve soft switching, reduce leakage currents and optimize power distribution. The goal is to ensure reliable energy routing, voltage regulation and electrical isolation for marine applications while minimizing losses. Experimental validation confirms the strategy’s effectiveness in maintaining stable output and high efficiency across dynamic voltage/load conditions. Design/methodology/approach This study proposes an integrated modulation strategy for a DAB-LLC bidirectional isolated converter. The LLC circuit uses dynamic full/half-bridge mode switching to extend voltage gain range and reduce DAB input voltage stress, maintaining DCX operation. The DAB circuit adopts extended phase-shift modulation to minimize leakage current RMS values and achieve full-load soft switching. Mode-switching voltage thresholds and transformer parameters are optimized to ensure efficient power distribution across wide input voltage ranges. Experimental validation confirms enhanced efficiency, stable output and ZVS operation under dynamic load/voltage conditions, demonstrating improved adaptability for marine hybrid systems. Findings Experimental tests confirm that the integrated modulation strategy can achieve a stable 80 V output within the input voltage range of 60–160 V. All switches achieve full-load zero-voltage switching, reducing losses. The dynamic mode switching has a high efficiency and reaches its peak value at the maximum input voltage. Input voltage transients and load steps show minimal output fluctuations. This strategy ensures the reliable operation of the marine hybrid system over a wide range. Originality/value This study proposes a hybrid modulation strategy integrating dynamic LLC topology switching and DAB extended phase-shift control, uniquely tailored for marine hybrid systems. To the best of the authors’ knowledge, this study is the first to optimize mode-switching thresholds for wide-voltage efficiency and ZVS-enabled reliability.

To address the limited dynamic voltage regulation performance of LLC resonant converters under wide input voltage and load variations, a reinforcement learning-based voltage control strategy is proposed in this paper. The twin delayed deep deterministic policy gradient (TD3) algorithm is adopted to learn the nonlinear mapping between system states and control actions, enabling adaptive adjustment of the converter operating parameters. Based on the established LLC resonant converter simulation model, the state space, action space, and reward function of the agent are designed to ensure rapid control response to abrupt changes in input voltage and load. Compared with the conventional PI control strategy, the proposed TD3-based strategy provides faster control actions during operating condition transitions, effectively suppressing output voltage overshoot and undershoot, and shortening the settling time. Simulation results verify that the proposed method achieves improved dynamic response performance under various operating conditions, demonstrating its effectiveness and superiority in LLC resonant converter voltage regulation.

With the advancement of wind power technology towards larger-capacity and higher-power turbines, their main shaft bearings face significant lubrication challenges under extreme temperatures. In this study, seven modified greases were prepared by adding 0.5 wt.% of tungsten disulfide (WS2), zinc sulfide (ZnS), and sulfurized isobutylene (T321). The concentration of all additives is given in weight percent (wt.%). Using a combined approach of friction and wear testing along with rheological analysis, this study systematically evaluated the tribological performance of the greases at high temperature (80 °C)-with the friction coefficient and wear scar diameter as key parameters-and their rheological properties across a wide temperature range (-20 °C to 80 °C), focusing primarily on shear stress and viscosity. All critical input parameters, including temperature, load, and shear rate, were precisely controlled and monitored using calibrated instruments. Results indicate that the WS2 and T321 compounding system demonstrated optimal performance, achieving a low average coefficient of friction of 0.024 and an average wear scar diameter of only 0.367 mm. At the same time, the WS2/T321 composite formulation exhibits excellent shear stability at high temperatures and good flow properties at low temperatures, demonstrating optimal viscosity-temperature characteristics. This study develops a promising grease formulation through multidimensional performance evaluation, offering key experimental support for designing high-performance wind turbine spindle bearing greases under high-temperature conditions.

Purpose The purpose of this study is to develop an intelligent control strategy for operating mode selection in an LLC resonant DC–DC converter with an adaptive transformer turns ratio. By replacing conventional threshold-based mode switching with a machine learning–driven approach, the aim is to enhance efficiency, stability and adaptability under dynamic and uncertain operating conditions. Design/methodology/approach A two-mode LLC converter topology using magnetic flux manipulation for transformer turns-ratio modulation is investigated, enabling operation in common mode and differential mode. A supervised machine learning framework – using support vector machine (SVM) and random forest (RF) classifiers – is trained on combined simulation and experimental data sets. Input features include real-time parameters such as input voltage, load power, switching frequency and historical converter states. The trained classifiers replace the fixed hysteresis-based logic to enable adaptive, data-driven operating mode decisions. A 300-W experimental prototype is built to validate the proposed method. Findings Compared with conventional hysteresis-based mode selection, the proposed ML-driven controller achieves up to 1.6% higher average efficiency and more than 20% reduction in switching-frequency variation across the tested operating range. The two classifiers – SVM and RF – consistently maintain soft switching and stable regulation under dynamic load conditions. These results confirm that data-driven mode selection enhances both performance and robustness relative to traditional threshold-based methods. Originality/value To the best of the authors’ knowledge, this work is among the first to apply supervised machine learning for real-time mode selection in an LLC resonant converter with adaptive transformer turns ratio. The approach eliminates the need for manually tuned voltage thresholds and hysteresis windows, enabling robust performance under variable and uncertain conditions. The results contribute to the development of intelligent, self-optimizing power electronics systems and open new avenues for integrating data-driven control into high-frequency converter design.

This paper presents a real-time hybrid intelligent control framework for a Positive Output Super-Lift Re-Lift Luo Converter (POSLRLC), developed to enhance voltage regulation, efficiency and power quality in electric vehicle (EV) charging networks. The proposed architecture integrates a Neuro-Fuzzy C-Means (Neuro-FCM) clustering mechanism with an Adaptive Neuro-Fuzzy Inference System (ANFIS) for dynamic control optimization. The Neuro-FCM module continuously performs online fuzzy clustering on the error [E(k)] and change of error [[Formula: see text](k)], generating adaptive Gaussian membership functions, while the ANFIS layer refines rule consequents using hybrid learning combining least-squares estimation and gradient descent. This dual-layer adaptation provides self-tuning control action without manual gain adjustment, overcoming the limitations of static fuzzy and PI-type controllers. The proposed model was implemented and verified on a Xilinx Zynq UltraScale [Formula: see text] MPSoC platform under a bare-metal C environment, with real-time fixed-point inference executed on the FPGA fabric. Simulation and hardware validation were conducted under varying PV input (10–20[Formula: see text]V) and dynamic load (10–50 [Formula: see text]) conditions. Results demonstrate significant performance improvements compared to conventional PID, MPC, SMC and AGA-FLC controllers. The proposed system achieves a rise time of 6.7[Formula: see text]ms, settling time of 7.2[Formula: see text]ms and steady-state error of 0.12%, with converter efficiency maintained at 96.7–97.2% and total harmonic distortion (THD) reduced to 2.48%. The FPGA implementation yields sub-microsecond computation latency with [Formula: see text]50[Formula: see text]ns jitter, enabling deterministic high-speed control. Overall, the hybrid Neuro-FCM [Formula: see text] ANFIS control system demonstrates superior dynamic response, efficiency, stability and harmonic suppression, offering a scalable and intelligent solution for next-generation EV power conversion and renewable energy integration.

Sparse Bayesian learning achieves great success in damage identification by employing a sparsity-inducing prior distribution on the sparse coefficients. However, if the input force is unknown, a sparse prior cannot be applied to both the damage and force parameters. In order to construct a sparse Bayesian learning model for damage and unknown load input identification, a sensitivity-based sparse Bayesian method for the damage detection with unknown input is proposed. An optimization equation is constructed on the basis of the dynamic response sensitivity to convert the complex non-linear relationships into linear equations. The prior information for the force and damage parameters is established using uniform and Gaussian priors, respectively, depending on the specific characteristics of each parameter. The Bayesian learning framework based on the sensitivity-based model is derived to compensate the linear truncation errors and measurement noise. The validation of the proposed approach is conducted using both a numerical frame structure and an experimental structure. Results indicate that this method can simultaneously identify both damage and forces, even when faced with significant measurement noise and limited sensor data.

The primary challenge in utilizing Fuel Cell (FC) systems lies in their inherently low and fluctuating output voltage, which contrasts with the requirements of a direct current (DC) bus network that demands a stable and relatively high voltage level. Ensuring consistent voltage regulation in the DC bus network is essential for reliable system performance. To overcome this issue, an interface converter is required to elevate and stabilize the voltage output under dynamic operating conditions. This paper introduces a high step-up DC–DC converter integrated with an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based control scheme for enhancing the performance of fuel cell (FC) power systems. The proposed work encompasses the modeling, analytical design, and structural development of the converter and its intelligent control mechanism, supported by comprehensive simulation results. The converter structure incorporates a clamp unit, a VMC (Voltage Multiplier Cell), and cascaded QBC (Quadratic Boost Converter) stages for achieving ultra-high voltage gain, enabling a substantial voltage gain of up to 9.65 times, effectively boosting the voltage from 45 V to 400 V. The system's performance was evaluated under three distinct scenarios: (1) varying input voltage with constant load power, (2) constant input voltage and load power, and (3) simultaneous variation in both input voltage and load demand. The ANFIS controller effectively maintains a stable output voltage of 400 V with a maximum deviation of only ±3.5%. In addition, the proposed converter achieves a peak efficiency of 87% under varying load conditions, demonstrating its suitability for fuel cell-based energy systems

Abstract Estuaries are threatened by eutrophication due to anthropogenic nutrient loading. The Great Bay Estuary of New Hampshire and Maine has been designated as nitrogen impaired primarily due to a 65% loss in seagrass coverage between 1996 and 2023. This region has also experienced multiple consecutive years of low annual precipitation and more recent record precipitation years. The loss of seagrass, increased precipitation variability, and continued anthropogenic influence have biogeochemical consequences for the estuarine filter. Water quality budgets for a subregion of Great Bay Estuary were developed at annual timescales for nitrogen, orthophosphate (PO 4 3− ), dissolved organic carbon (DOC), and total suspended solids (TSS). Results show annual total nitrogen (TN) input loads, including tidal flux into Great Bay, (3,900 kg ha −1 year −1 ) are less than output loads whereas dissolved inorganic nitrogen (DIN) inputs (1,410 kg ha −1 year −1 ) exceed outputs, indicating net TN export and net DIN retention. PO 4 3− loads were nearly balanced, with a mean input and output of 257 kg ha −1 year −1 and 238 kg ha −1 year −1 , respectively. Seagrass acreage negatively correlated with DOC inputs (33,300 kg ha −1 year −1 ). High TSS output loads (227,000 kg ha −1 year −1 ) resulted in net TSS export. Despite critical seagrass loss and recent point source nitrogen load reductions, the system continues to retain bioavailable forms of nutrients, suggesting the importance of other primary producers to estuarine biogeochemical cycles. Water quality budgets for Great Bay provide useful insight into the biogeochemical capacity of an estuary during a time of habitat degradation and subsequent coastal management efforts.

Torsional vibrations in deep drilling systems, such as the nonlinear stick-slip effect, pose significant challenges to operational stability and tool longevity. Building on previously derived analytical setting formulas for Proportional-Integral (PI) state control, this paper applies them to a two-mass oscillator system, commonly used as a simplified model for deep drilling setups. The proposed approach leverages an adaptive algorithm that dynamically tracks the friction torque and adjusts the feedback vector according to the corresponding load input functions. Hereby, the torsional vibrations, induced due to the contact between the drill bit and the rock formation, have been modelled using the static friction (Stribeck) model and measured using different rock samples on a test rig in the form of a scaled-down drilling rig. By integrating these setting formulas with the developed adaptive mechanism, the controller ensures robust performance across a wide range of operating scenarios. Simulation results demonstrate the system’s ability to suppress torsional oscillations effectively, achieving significant vibration reduction without compromising the stability of the system. This research extends the theoretical foundations established in prior work and provides a robust framework for advancing vibration control technologies, contributing to safer, more reliable, and efficient operations in the energy industries. Key words. PI state control, adaptive tracking algorithm, torsional vibrations, nonlinear friction, stick-slip

Under random wave loading, the crack growth rate exhibits jump-like cycle-to-cycle variations, which limit the direct use of efficient integration schemes such as the Euler method. In addition, the crack growth life is highly sensitive to the initial crack size and aspect ratio, while the initial defects are often difficult to determine accurately in practice, leading to increased uncertainty in life assessment. To address these issues, a cycle-scaling-based crack size accumulation method for random loading is proposed. A predictor–corrector improved Euler method is then established, and a fourth-order Runge–Kutta scheme incorporating the cycle-scaling transformation is derived. Furthermore, based on spectral analysis theory, a mapping between the wave spectrum and the crack-tip stress intensity factor response spectrum is developed. A stress intensity factor range sequence is generated by concatenating short-term sea states, thereby providing a random loading input that preserves the required statistical characteristics. Finally, a 21,000-TEU container ship is analyzed as a case study to investigate crack growth evolution for different initial aspect ratios. The results show that the crack aspect ratio gradually converges to a particular trend during propagation. A convergent aspect ratio curve is fitted. And a unified life assessment curve is constructed. An equivalent transformation is used to map an arbitrary initial crack shape and size to an equivalent convergent aspect ratio crack. As a result, fatigue life can be rapidly estimated using a single “initial crack size–fatigue life” curve, providing support for crack growth life assessment and the definition of defect acceptance limits for ship hull structures.

Introduction: The plantar flexors (PFs) are crucial for ankle push-off during gait, providing forward propulsion and controlling foot movement during the stance phase. In stroke survivors, reduced activation of the PFs leads to difficulty with propulsion and toe clearance, contributing to abnormal gait patterns known as hemiplegic gait. This issue is often exacerbated by decreased weight bearing (WB) on the paretic limb and adaptive shortening on the paretic PFs, as PF activation is highly dependent on loading input for balance and gait. Purpose and Hypothesis: This study investigates whether simultaneously increasing paretic WB and PF muscle length can enhance paretic PF muscle activation during a dynamic skateboarding task. We hypothesize that this combination will significantly increase paretic PF activation. Methods: This study was a single-session, randomized controlled trial. We recruited 15 individuals with chronic stroke and measured the activations of their paretic plantar flexors using surface electromyography (EMG). Measurements were taken under six randomized conditions: three paretic limb loadings (50%, 75%, and 90% of BW) paired with two surface configurations (level and incline). During each condition, participants were instructed to roll a skateboard forward and backward for three cycles using their unaffected limbs, all while maintaining the targeted WB level on their paretic limbs. A high-speed camera system recorded foot trajectories to define the onsets of the forward and backward cycles during the skateboarding motions.PF muscle activation was calculated by integrating the EMG signal within a cycle and then normalized to 50% BW on the level condition. Results: The results showed increasing paretic BW loading resulted in significant increases in paretic PF activation. The normalized paretic PF activation was significantly higher during 75% BW and 90% BW loading compared to 50% (p =0.04 and 0.01 respectively). There was a trend of increasing PF activation on the incline surface compared to the level surface, however, this difference was not statistically different. Discussion and Conclusion: These findings suggest greater paretic PF activation is associated with increasing paretic WB. Locomotor exercises should aim to increase WB on the paretic limb to enhance paretic PF activation, which may improve gait and standing balance. Further studies with subacute stroke populations and a larger sample size are needed to generalize these results.

This article introduces an integrated control scheme combining an Adaptive Extended Kalman Filter (AEKF) with a Passivity-Based Control (PBC) approach to stabilize a DC-DC boost converter feeding both constant voltage and constant power loads (CPLs) in DC microgrids. Unlike conventional observers, the AEKF adapts its covariance matrices in real time to accurately estimate both system states and the unknown load dynamics introduced by CPLs, thereby eliminating the need for additional sensors and enhancing estimation convergence. Coupled with the PBC, the estimated disturbances are compensated via a feedforward path, significantly improving the system’s resilience to input voltage fluctuations and load variations. Through a Lyapunov-based stability analysis, the combined strategy is proven to ensure large-signal stability while maintaining a rapid transient recovery profile, even under significant parametric uncertainties. The simulation of this algorithm was implemented using PLECS, thoroughly validating the effectiveness and robustness of the proposed method.

As CMOS technology scales into the nanoscale regime, ensuring the reliability of digital circuits in radiation-rich environments has become a critical challenge. Standard cell libraries, which are foundational to digital design, are typically characterized using extensive SPICE simulations to capture gate delays as functions of input transition time and load capacitance. However, these libraries do not account for total ionizing dose (TID) effects, which are caused by prolonged exposure to ionizing radiation and introduce oxide-trapped charges and interface states that degrade key transistor parameters, such as threshold voltage and leakage current. This results in significant timing inaccuracies, compromising digital timing closure in mission-critical applications such as aerospace and nuclear electronics. In this work, we propose an efficient, TID-aware standard cell characterization methodology for nanoscale CMOS technologies that generates cell characterization data in standard Liberty format, enabling accurate prediction of timing closure under TID influence without incurring any SPICE simulation overhead. Our approach leverages well-calibrated 32 nm Synopsys©Sentaurus TCAD simulations and variation-aware analytical timing models to capture TID-induced degradation. These effects are incorporated into cell netlists through adjustments to the BSIM parameters to generate both pre- and post-radiation standard cell libraries. Validated using a set of reference designs, including ISCAS benchmark circuits, the proposed methodology achieves accurate path-level timing predictions under radiation while reducing SPICE simulation effort by approximately 81.25%. By bridging device-level radiation effects with cell-level timing abstraction, this scalable framework offers a practical solution for robust and radiation-resilient digital integrated circuit design in harsh environments.

ABSTRACT Mandatory waste sorting policy in Beijing (2020) lowered mercury inputs to the representative MSWI plant by 67.7 % versus 2019 through removal of batteries and other Hg-rich articles. The front-end change propagated through the entire system, cutting stack emissions 82 % (1.35 ± 0.6 → 0.24 ± 0.05 µg m/³) and decreasing the Hg²⁺ fraction from 45.5 % to 28 %. Higher post-sorting plastic loads elevated chlorine, promoting in-furnace oxidation of Hg0; the resultant Hg²⁺ was efficiently captured by existing wet scrubbing/fabric filters. Mass-balance and speciation data show that source separation functions as a chemical pre-treatment that shifts mercury toward easily removable forms, amplifying downstream control performance without hardware retrofits. Implications: This study provides the first field-scale proof that waste-classification policy can govern mercury emission chemistry. By altering waste composition at the source, mandatory sorting oxidizes elemental mercury and enhances removal efficiency, offering decision-makers a low-cost, proactive tool to meet stringent Hg limits while advancing circular-economy goals.

Typical resonant converter controllers are based on linearised averaged models, which have significant modelling errors when there are wide fluctuations in the input voltage, load and reference voltages. In this article, a piecewise affine (PWA) switching surface with active border tuning of affine sections, called the Partition Border Tuning (PBT) controller, is proposed for DC–DC series resonant converters (SRCs). Lyapunov stability analysis is used to ensure closed‐loop stability. A new Chattering Mitigation (CM) technique is proposed to suppress unwanted oscillations between modes and output voltage overshoot under transient conditions, which are generally present in conventional switching surface controllers. This technique eliminates chattering, reduces output voltage overshoot and limits the maximum inductor current and capacitor voltage amplitude of the resonant tank under transient conditions. Simulation and experimental data are presented to demonstrate the effectiveness of the proposed approach.

This paper presents a single-phase five-level inverter topology using a modified quasi-Z-source network (Five-level Modified Quasi-Z-source H-bridge Inverter – 5L-MqZS-HBI), which combines two quasi-Z-source networks with an H-bridge inverter. 5L-MqZS-HBI is designed to reduce the number of inductors, improve overall efficiency, and enhance voltage gain while maintaining the ability to generate multilevel output voltages, making it suitable for applications requiring high-quality, high-voltage outputs with low harmonic distortion under low input voltage conditions. A small-signal analysis is performed considering parasitic elements of inductors and capacitors to establish the relationships between capacitor voltage, inductor current, and control parameter, so the system transfer function is derived to design a PI controller that ensures stable output voltage despite changes in input voltage or load. The paper also presents detailed operating states, evaluates controller performance, efficiency, and THD, and confirms the analysis through PSIM, MATLAB simulations and laboratory experiments.

This study presents an in-depth analysis of heat loss mechanisms in a rotary kiln system used for biomass torrefaction, with briquetted biomass fuel serving as the primary thermal energy source. The study evaluates five principal heat loss pathways: wall heat loss, exhaust gas heat loss, hydrogen-related heat loss, moisture evaporation, and heat loss due to incomplete combustion. Experimental tests were conducted at three torrefaction temperatures (230°C, 250°C, and 270°C), and thermal energy losses were quantified through temperature measurements, energy balance equations, and gas composition analysis. Results indicate that while absolute heat loss values increased with higher torrefaction temperatures due to elevated energy input and system load, the percentage of heat loss relative to total input decreased, improving net thermal efficiency. Wall heat loss was the dominant component across all conditions but declined in percentage terms from 80.5% to 30.9% as temperature increased. The reuse of exhaust gas for drying briquetted biomass was also investigated, demonstrating that waste heat recovery significantly reduces drying time—from 19 hours at 230°C to 13 hours at 270°C—without compromising fuel integrity. These findings confirm that integrating exhaust gas utilization into the torrefaction process enhances energy efficiency, supports continuous operation, and reduces external energy demand, offering a viable strategy for sustainable biomass fuel processing at an industrial scale. The findings provide design guidance for integrating heat recovery into industrial-scale biomass torrefaction systems.

Grippers are essential components in robotic systems, particularly for tasks involving object grasping and manipulation. A constant force gripper provides the advantage of generating a nearly constant output force over a range of input loads without relying on sensors or control, which significantly reduces cost and system complexity. In this article, a topology optimization model is proposed for the design of a constant force gripper, which can simultaneously optimize both rigid links and material distribution. The combination of rigid links and deformable material is expected to reduce the low preload stroke, enabling the gripper to reach the constant force stroke rapidly and thereby improving operational efficiency. To develop the topology optimization model, rigid links and material distribution are represented by state and density variables, respectively, with the constant force behavior as the objective function. To implement a numerically efficient gradient‐based algorithm, the sensitivity of the objective function with respect to the densities of all elements and the states of all links is derived. The experimental result shows the preload stroke of the gripper is 2.2 mm, and the constant force stroke reaches 6.4 mm. The preload stroke of the gripper is reduced by 85.33% compared with the fully compliant constant force gripper.