Electrohydraulic shock waves (EHSWs) from pulsed discharges in liquids are used in diverse applications, where the peak pressure is particularly critical for inducing mechanical impacts. However, accurate numerical prediction of shock wave profiles typically requires high-performance computing, which limits routine use for end users. This paper presents a two-stage algorithm for the prediction of EHSWs generated by high-current pulsed discharges in weakly compressible liquids. Stage 1 computes the temporal evolution of the discharge channel and the pressure using energy-based equations with a coarse time step. Stage 2 reconstructs the spatial pressure distribution under an acoustic approximation and a cylindrical piston representation, enabling rapid evaluation with minimal computational cost within the intended operating regime. The approach is validated against two experiments: a high-current pulsed discharge and a current-interception configuration, while maintaining <2.5 % error in peak pressure across parameter sets. Parametric analysis further highlights that effective energy unifies the scaling of peak pressure across circuits, providing actionable guidance for circuit design under typical high-current discharge conditions.

Advanced pulsed-energy systems require dielectric materials that can dissipate transient electrical energy rapidly while resisting localized failure. Carbon–ceramic composites represent a typical heterogeneous ceramic dielectric in which conductive carbon networks, insulating ceramic skeletons and pores coexist across different length scales and jointly govern energy-dissipation behavior. However, carbon aggregation and nonuniform conductive-network distributions readily form locally percolated low-resistance pathways, causing current crowding, localized Joule heating and premature thermomechanical degradation. Such multiscale heterogeneity makes conventional trial-and-error optimization inefficient and also challenges generic data-driven models, which often fail to preserve phase occupancy, conductive connectivity and physical-response consistency under limited characterization data. Here, we propose a generative AI-guided framework for conductive carbon-network regulation in heterogeneous carbon–ceramic composites. A multiphase pseudo-RGB tensor encoding strategy is first developed to represent the carbon phase, ceramic skeleton and pore phase within a unified physically interpretable feature space. A WGAN-GP model constrained by physical-statistical descriptors is then used to expand the microstructure design space while preserving carbon fraction, spatial uniformity and agglomeration characteristics. A phase-field-informed optimization model further redistributes highly agglomerated carbon regions toward more uniform and multibranched conductive networks under mass-conservation, interfacial-regularization and spatial-smoothness constraints. Electrothermal–mechanical simulations and pulse-impact experiments demonstrate that the regulated carbon network suppresses current crowding and localized thermal accumulation, increasing the maximum permissible single-pulse absorbed energy density by approximately 78%. This work establishes a generative-AI route that links image-based microstructure characterization with conductive-network regulation for improved transient energy dissipation in heterogeneous ceramic composites.

Aiming to address the lack of a quantitative analysis in the design of the system for rock fragmentation by highvoltage pulsed discharge (RHPD), an optimization design method for the RHPD system based on medium breakdown characteristics and stress impulse constraints is proposed, and a multi-physical parameter coupling mathematical model based on numerical calculation and circuit simulation is constructed. Considering the drilling condition with a 3 1/2 in. wellbore diameter, the optimal energy parameter of the RHPD system with an 88.9 mm claw-shaped discharge electrode is obtained based on the optimization design model. The optimal energy parameter of the system and the corresponding equivalent capacitance, equivalent charging voltage of the system are 58.75 J, 2.966 nF, and 199.03 kV, respectively. Using these optimal parameters, a RHPD system with a coaxial Marx generator as the topological structure is developed. The rock fragmentation performance of the system is tested in a transformer oil-granite medium. After 180 min, a drilling effect is observed on the granite surface, with a hole diameter of 92.66 mm and a depth of 35.70 mm. The energy consumption per unit volume for fragmentation of the developed system is approximately 967 J/cm3, and the rate of penetration at a 10 Hz pulse discharge frequency is approximately 0.3 m/h. This research provides theoretical guidance for the parameter optimization and structural design of the RHPD system.

Free-space optical communication terminals require compact architectures and high-precision alignment schemes. However, conventional incident-angle sensing usually relies on beam splitting and dedicated position sensors, increasing system complexity and terminal size. Moreover, power-feedback alignment typically provides a limited field of view (FOV) and slow convergence. We propose a direct beam incident-angle measurement method based on a multi-core fiber (MCF) for a common-path FSOC terminal. By exploiting the coupling-power characteristics of a seven-core fiber, the incident angle is directly inferred without any auxiliary position-sensing branch. We establish an incident-angle estimation model and further improve its accuracy through thin-plate-spline (TPS) calibration, achieving high-precision angle measurement over ±840 µrad with a mean error of 21.72 µrad and a root-mean-square error (RMSE) of 23.74 µrad. A 550 m outdoor experiment further demonstrates stable beam alignment and 10.3125 Gb/s IMDD transmission.

With the development of ultra-high voltage (UHV) transmission technology, transformer insulation systems face increasingly severe safety challenges. Natural ester insulating oils, due to their environmental friendliness and high flash point, have become an attractive alternative to mineral oils. However, their shock wave propagation characteristics under extreme electrical stress remain unclear. This study established a comprehensive high-voltage pulse discharge experimental platform. Combining a schlieren imaging system with multi-parameter simultaneous measurement technology, we captured the three stages of high-voltage pulse discharge in oil and investigated the shock wave formation mechanism and attenuation characteristics of soybean and rapeseed-based natural esters. Pin-to-pin discharges were driven by a Marx generator, and shock wave pressure attenuation was quantified using a pressure sensor array. Compared to conventional mineral oils, both natural esters exhibited superior shock wave attenuation, with attenuation coefficients ranging from 1.05 to 1.07 for soybean-based esters and 0.95 to 0.96 for rapeseed-based esters, demonstrating their engineering advantages in suppressing fault shock waves. This study provides key experimental evidence for the safety design and fault protection of natural ester-insulated transformers.

The non-uniform microstructure of Carbon Ceramic Resistor (CCR) critically limits their energy endurance, posing a significant bottleneck in high-power applications. To overcome this, this work introduces a computation-driven paradigm to quantitatively re-engineer the material’s internal architecture. Drawing inspiration from AI-based image processing, we apply convolutional operators to a digitized microstructure map, transforming the random conductive network into a highly uniform and predictable structure. This targeted optimization is validated through both simulation and experiment, demonstrating a nearly threefold enhancement in energy endurance density, from 421.63 J/cm3 to an impressive 1300.56 J/cm3. Microstructural and electro-thermal analyses reveal the underlying mechanism: the engineered uniform conductive network effectively suppresses current channeling and the formation of destructive local hotspots. This work not only presents a validated solution for CCR but also establishes an efficient, predictable, and systematic design framework to supersede traditional approaches in developing high-performance composite materials.

Magnetic-core current transformers (CTs) are susceptible to high-frequency resonance during nanosecond-scale pulsed current measurements due to distributed parasitic elements, resulting in limited bandwidth and reduced accuracy. To address this issue, an optimized design approach based on segmented damping resistors is proposed. A PCB-based coplanar interwound structure is further introduced into a conventional magnetic-core CT with self-integration to enable precise damping configuration and reduce parasitic inductance, supporting the development of dynamic damping networks. Meanwhile, a distributedparameter equivalent circuit model incorporating parasitic effects is constructed to elucidate the resonance suppression mechanism, where theoretical analysis establishes the relationship between segmentation and oscillation attenuation, and Simulink simulations are conducted to evaluate output behavior under varying segment counts. Furthermore, the experimental tests using a square-wave excitation with a 20-ns rise time are carried out to compare undamped and segmented damping configurations, showing that high-frequency oscillations are effectively suppressed, with damping performance improving as the segment number increases. In particular, a 13-segment configuration reduces overshoot to 19.7%, achieving a 72.9% decrease compared to the undamped case. These results verify the effectiveness of segmented damping in mitigating high-frequency parasitic effects and offer a theoretical and engineering basis for the precision design of fast pulsed current measurement systems.

Underwater pulsed arc discharge plasma serves as the "tool" for generating shock waves and high-energy species. To deeply study the characteristics of plasma channels, this paper conducted temperature diagnosis of underwater microsecond high-current pulsed discharge. A spectral measurement system calibration method considering the water absorption effect was proposed. Time-resolved emission spectra of the plasma channel were obtained, and the spectral characteristics during the arc discharge phase were analyzed. A diagnostic approach based on the combined analysis of spectral-line profiles and intensities was developed to determine the excitation temperature and blackbody radiation temperature. The time-dependent evolution of the blackbody radiation temperature and the excitation temperature of the plasma was characterized. The results indicate that, for kA-level pulsed high-current underwater arc plasmas, the temperature ranges of the order of 1-2 & times; 10(4) K.

The power conditioning system is a critical component in large-scale, high-power laser facilities. Due to its high-voltage, high-current operation, the system is prone to unexpected failures that can disrupt normal facility operation. This study analyzed failure mechanisms by examining malfunction incidents in thyristor-based solid-state switches within this system. Failure replication experiments identified inconsistent reverse recovery characteristics and inadequate resistance to voltage spikes as the primary causes of thyristor overvoltage damage. Consequently, corrective measures were implemented, including reverse-recovery characteristic matching and passivation treatment processes. The effectiveness of these improvements was subsequently confirmed through comparative experiments.

With the growing application of pulsed power generators in insulation tests, the demand for high-voltage pulses with complex waveforms has surged. However, limited by simplistic modulation architectures and control algorithms, traditional generators struggle to extend beyond basic waveforms. To address this, the paper proposes an advanced modulation and control algorithm for all-solid-state pulsed power supplies to generate complex, high-voltage, arbitrary waveforms. A matrix-based parameterization enables systematic mapping between the target waveform and the timing data of each semiconductor switch. The distributed control strategies solve the problem of widely varying timing parameters by allocating a separate memory to each switch and cyclic data loading. To verify the theoretical feasibility, a prototype capable of outputting ±21kV (49 levels) bipolar waveforms was constructed and tested. The prototype reliably generates high-voltage arbitrary pulses and operates stably under both resistive and capacitive loads, providing a flexible, high-performance voltage source for advanced testing applications.

As the application scenarios of solid-state high-voltage pulse generator (SSHVPG) continue to expand, the requirements for reliability are getting higher and higher. Inside the SSHVPG, the drive circuit is the key to its reliability. In this article, a novel, highly reliable drive circuit is proposed for the SSHVPG that provides both overvoltage protection and overcurrent protection. The drive circuit uses active clamping to suppress the turn-off overvoltage, and the overcurrent protection circuit based on the thyristor is applied to ensure that the semiconductor switches can turn off quickly and stably when overcurrent appears. The simulation model was established to analyze the protection mechanism in different situations, and it was found that the overcurrent protection and overvoltage protection do not work independently, but there is a mechanism of mutual influence, and both of them work together to complete the overcurrent protection of the semiconductor switch. Prototype experimental results verify the feasibility and practicality of this drive circuit.

Quantity traceability serves as the fundamental prerequisite for ensuring the accuracy of impulse current measurements. Existing methods primarily focus on the impulse scale factor, achieving its traceability through AC calibration at multiple frequency points and a synthetic model; however, the model accuracy is limited by preset static parameters. To address this limitation, this article proposes a dynamic and self-adaptive traceability method based on parameter identification of the impulse current's magnitude-frequency characteristics. A universal analytical expression for the output impulse voltage of the measuring system is established to accurately fit the output waveform and extract its magnitude-frequency characteristics. Combined with the AC scale factors obtained via frequency sweep, the magnitude-frequency response of the input impulse current is reconstructed. Subsequently, parameter identification is performed based on the magnitude-frequency characteristic model of the input impulse current, enabling the inversion of its time-domain waveform. Throughout the uncertainty propagation process, uncertainty evaluation is conducted synchronously, thereby enabling a systematic assessment of the traceability accuracy for both the impulse current peak and time parameters. To validate the effectiveness of the proposed method, experimental investigations were carried out using a current transformer (CT) and a shunt. The results demonstrate that the relative uncertainties of the reconstructed input impulse current's peak, front time (T-1), and time to half-value (T-2), derived from the output impulse voltages of the two measuring devices, do not exceed 0.926%, 0.719%, and 0.664% (at 95% confidence level), respectively, for the 8/20- mu s impulse waveform. Furthermore, the obtained quantities exhibit good consistency, validating the accuracy and efficacy of the proposed theoretical approach.

Currently, there is no unified metrological standard for large impulse currents. Existing traceability methods struggle to effectively correlate the impulse quantity values with established standard quantities, nor can they quantitatively evaluate the linearity of measuring devices under high-amplitude impulse current conditions. To address these issues, this paper proposes a traceability method for large impulse currents based on waveform reconstruction. By integrating the magnitude-frequency characteristics of the measuring device, the impulse voltage output by the device is reconstructed into the input impulse current. A waveform similarity evaluation index is established to assess the reconstruction quality, which indirectly reflects the linearity of the measuring device over its full range. On this basis, precise calibration of the measuring device can be achieved using the reconstructed impulse current. To verify the effectiveness of the proposed method, experimental studies were carried out using a laboratory-made Rogowski Coil (LMRC, passive type with poor low-frequency performance) and a Pearson Current Transformer (PCT). The experimental results show that, over the full range, the similarity between the reconstructed input impulse currents of the two devices increases from 0.44 before reconstruction to nearly 1, which fully confirms the effectiveness of the proposed waveform reconstruction method and the excellent linearity of the two measuring devices. Furthermore, through the data fusion of the reconstructed impulse current values, it is found that the linearity of PCT and LMRC is no more than 0.120% and 0.185%, respectively. In addition, impulse tests were performed using PCT and a laboratory-made current transformer (LMCT), and the evaluation results show that their linearity is no more than 0.115% and 0.260%, respectively. Within a 95% confidence interval, the relative expanded uncertainties of their impulse scale factors are no more than 0.483% and 0.621%, the front time corrections are -0.055 μs ± 0.089 μs and -0.020 μs ± 0.095 μs, and the time to half-value corrections are -0.053 μs ± 0.206 μs and 0.001 μs ± 0.205 μs, respectively. The research results indicate that LMCT is comparable to PCT in terms of impulse scale factor and linearity, while it outperforms PCT in time parameter measurement performance, indicating its potential for large impulse current traceability and the establishment of a reference standard.

Free-space optical communication (FSOC) terminals require rapid and accurate beam alignment. Conventional alignment schemes typically rely on beam splitting and dedicated sensing paths, which increase optical loss and system complexity. In this paper, we propose a common-path FSOC terminal enabled by a terraced multi-core fiber (MCF) and a gradient-descent laser nutation (GDLN) algorithm for fine alignment. The performance and operational field of view (FOV) of the proposed system were experimentally validated on a collimated-light tunnel platform, achieving a fine-alignment FOV of up to ±3.33 mrad. An outdoor experiment was further conducted between two buildings separated by 860 m. A six-axis motion platform applied external disturbances with an amplitude of 3° to the terminal at 4-hour intervals over a duration of 24 h. The system automatically realigned to restore the link within 4 s. Furthermore, real-time single-wavelength 10 Gbps IMDD transmission was demonstrated with a BER of 1.242×10-11.

Phase stability and osteoconductivity are very important for the hydroxyapatite (HA)-coated implants. The design of HA coating with c-axis orientation is linked to improved phase stability and enhanced osteoconductivity. In this study, the microstructural characteristics of oriented hydroxyapatite coatings were tailored by micro-plasma spray (MPS) and atmospheric plasma spray (APS) techniques. SEM results indicated that both coatings exhibit a columnar structure, and the XRD results correspond to a strong (002) orientation texture. To elucidate the formation mechanism of c-axis oriented HA coatings, cross-sectional samples encompassing both the coatings and individual splats were prepared for TEM observation by FIB techniques. TEM images of MPS sprayed coatings revealed fine grains at the splat-splat interfaces. Within the central regions of splats, the columnar grains exhibited widths of 200-400 nm and heights ranging from 2 to 10 mu m. While, smaller columnar grains, generally measuring 36-170 nm in width and 1-2 mu m in height, were observed within the splats of APS coatings. The detailed characteristics of in-flight particles and splats collected on the polished substrates were analyzed to discuss the formation mechanism underlying the structural differences in highly oriented hydroxyapatite coatings deposited by MPS and APS techniques. The parameters of spraying power and substrate temperature interact synergistically, emerging as critical determinants of the microstructural characteristics in oriented HA crystal domains. In addition, the biological properties of the highly oriented hydroxyapatite coatings were investigated. No significant difference was observed in the activity of MPS and APS deposited coatings to stimulate MC3T3-E1 osteoblast proliferation, the MPS deposited coatings demonstrated a superior ability to promote the formation of mineralized nodules compared to the APS deposited coatings.

Rock fragmentation by high-voltage pulsed discharge (RHPD) has gained significant attention in various engineering fields, yet the mechanism of rock fracture induced by the instantaneous high temperature of plasma remains insufficiently quantified. This study establishes a numerical model of thermomechanical damage in rocks under RHPD conditions, analyzes the time evolution of plasma temperature at different fragmentation distances, and proposes a method for calculating the time response of plasma temperature. Based on the discrete element model (DEM), the study simulates the thermal diffusion and thermally induced fracture process inside the rock under the instantaneous high temperature of plasma, revealing the multi-field coupling fracture mechanism induced by high-voltage pulses. The results indicate that plasma-induced high temperatures cause localized mineral particles inside the rock to heat up. However, due to the inherent physical properties of the rock, the range and magnitude of temperature changes remain limited. The bonding between minerals, the presence of voids, and the rock's low thermal expansion coefficient, low thermal conductivity, and high specific heat capacity collectively result in low heat transfer efficiency, making the effect of transient thermal loading on rock fracture negligible. Even under extreme thermal loading conditions (80000 K, 100 μs), although the range of temperature change increases and a small number of fractures are generated, the contribution of thermomechanical damage remains much smaller than that of shockwave-induced fracturing. The findings of this study enhance the understanding of thermomechanical coupling mechanisms in RHPD and provide a reference for the development of multi-phase medium coupling models.

High-voltage and high-current impulse waveforms, characterized by a diverse frequency spectrum, pose challenges for accurate traceability to national standards. Traditional methods for tracing impulse quantities to a single-frequency measurement standard are not universally applicable. This article introduces a novel approach that establishes a general traceability method by mapping the impulse scale factor to a multifrequency ac scale factor based on the energy spectrum, enabling indirect traceability to the power frequency standard under high-voltage or high-current conditions. A full-waveform inversion method is employed that integrates the frequency spectrum of measured impulse waveforms using fast Fourier transform (FFT) with the frequency response characteristics of the measuring system, which are determined through convolution based on square wave responses. This approach facilitates traceability of time parameters for arbitrary impulse waveforms. Experimental results demonstrate that the resistive impulse voltage divider's scale factor uncertainties at low and high voltages are 3.100x 10(-4) and 1.344x 10(-3) , respectively. Additionally, the uncertainties for the T-1 and T-2 parameters of the inverted waveform of the attenuator are within 0.45% and 0.03%, respectively. These findings validate the effectiveness of our traceability theory.

Transformers are susceptible to explosions under internal short-circuit faults, posing a significant threat to the safety and stability of power systems. Investigating the shock wave characteristics induced by pulsed arc discharges in transformer oil, which simulate real internal short-circuit fault scenarios, holds considerable theoretical and engineering significance for enhancing transformer damage resistance. In this study, a comprehensive experimental platform was developed to analyze pulsed discharges in transformer oil, which measures voltage, current, and shock wave signals, and visualize the arc channel and shock waves. To capture the complex coupling between electrical and fluid dynamic processes, a liquid-electric coupled shock wave model was constructed based on an improved arc impedance channel. This model incorporates the effects of channel radius, temperature, and time-varying conductivity. By leveraging the experimentally measured data, the model's parameters and initial conditions were iteratively calibrated to ensure computational accuracy. The arc channel properties and shock wave pressures predicted by the model were then compared with experimental results, demonstrating good agreement and validating the model's reliability. Overall, the proposed coupled model offers a critical engineering framework for improving the shock resistance and anti-explosion performance of transformer oil under fault conditions. (c) 2025 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International (CC BY-NC-ND) license (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Aircraft skin is composed of multilayers of heterogeneous functional coatings, and it is often necessary to selectively remove one or several layers of coatings during aircraft maintenance and repair. However, it presents a significant challenge in laser micro/nano processing because the spatial structure and physicochemical properties of the coatings are difficult to identify online. A laser adaptive cleaning technology (LACT) based on coaxial spectral monitoring is proposed in consideration of the synchronous induction of plasma spectra during laser processing, which utilizing the differences in wavelength and amplitude of characteristic spectral lines from heterogeneous materials. Combined with spectral/photoelectric online monitoring modes, element recognition and interface sensing with MHz high-speed of multilayer heterogeneous materials were completed during laser scanning processing. Meanwhile, two closed-loop feedback strategies were established by combining FPGA (Field Programmable Gate Array) control hardware and intelligent algorithms, including interface adaptive and energy adaptive. As a result, with the two intelligent processing strategies, the precision and efficiency of laser selective processing of multilayer heterogeneous materials are simultaneously improved, and the laser adaptive cleaning of multilayer heterogeneous skin coating is finished. As an innovative technology, it is expected to provide a new method and technology for high-precision selective processing of multilayer heterogeneous materials.

Polyether electrolytes have been widely recognized for their favorable compatibility with lithium-metal, yet they are hampered by intrinsically low oxidation thresholds, limiting their potential for realizing high-energy Li-metal batteries. Here, we report a general approach involving the bridge joints between non-lithium metal ions and ethereal oxygen, which significantly enhances the oxidation stability of various polyether electrolyte systems. To demonstrate the feasibility of the ion-bridging strategy, a Zn2+ ion-bridged polyether electrolyte (Zn-IBPE) with an extending electrochemical stability window of over 5 V is prepared, which enables good cyclability in 4.5 V Li||LiCoO2 batteries. Ampere-hour-level quasi-solid-state batteries of SiO-graphite||LiNi0.8Mn0.1Co0.1O2 (10 Ah, N/P ratio of 1.12, 303 Wh kg-1 at 0.1 C based on the total weight of the pouch cells) and 60 mu m-Li||LiNi0.9Mn0.05Co0.05O2 (18 Ah, N/P ratio of 2.5, 452 Wh kg-1 at 0.33 C based on the total weight of the pouch cells) pouch cells with Zn-IBPE present elevated electrochemical performance, benefiting from adequate interfacial stability. Nail penetration tests evidence high battery safety enabled by Zn-IBPE in 4 Ah graphite||LiNi0.8Mn0.1Co0.1O2 pouch cells without combustion or smoke. This work offers a pathway for designing high-voltage polymer electrolytes and a general solution for achieving high-performance quasi-solid-state batteries.