The flat open-circuit voltage versus state of charge (SOC) curve of LiFePO₄ batteries leads to difficulties in achieving an accurate estimation of SOC using only electrical signals. In addition, there are limitations in SOC estimation methods for single electrical, thermal, and acoustic data sources. In view of this, a multi-source data feature extraction method for SOC estimation of LiFePO₄ battery was proposed. A comprehensive feature extraction was carried out on the electro-thermal-acoustic multi-source data obtained from different angles. Considering the advantages of different feature selection methods, a new feature selection method integrating Spearman correlation coefficient, mutual information, category boosting and least absolute shrinkage and selection operator regression was proposed. The joint selection of electro-thermal-acoustic key features was realized to improve the accuracy of SOC estimation.

Firstly, an experimental platform for LiFePO₄ batteries was built. Electro-thermal-acoustic multi-source data were acquired. The transient features and short-term variation features of electrical and thermal signals, as well as the time-domain, frequency-domain, and time-frequency-domain features of ultrasonic signals were extracted, respectively. Secondly, in order to select the key features more accurately, a new method of feature selection incorporating Spearman correlation coefficient, mutual information, category boosting, and least absolute shrinkage and selection operator regression was proposed. In order to verify the performance of the proposed method, the proposed method was compared with SOC estimation results using all features and SOC estimation results under different feature selection methods. The effect of SOC estimation using single data source features versus multi-source data features was compared. The feasibility of the proposed method was verified at different magnifications and under different operating conditions. Finally, Gaussian white noise with different signal-to-noise ratios was added to the raw ultrasound signals acquired under dynamic stress test (DST) conditions and new european driving cycle (NEPC) conditions, respectively, to verify the applicability of the proposed method under high-intensity noise.

The results show that using the new method of feature selection proposed can effectively select the features that are important for SOC estimation with higher accuracy than SOC estimation using all features. With the same number of features, the SOC estimation accuracy of this method is improved compared with that of a single feature selection method. The model constructed using electric-thermal-acoustic multi-source key features has higher SOC estimation accuracy compared to single data source features. When using the BiGRU model, the mean absolute error and root-mean-square error of SOC estimates are 0.58% and 0.72%, respectively. The method performs well under a single operating condition. The method also shows good applicability at different discharge multipliers and under multiple operating conditions. Under DST conditions and NEDC conditions, the mean absolute error of SOC estimation is 0.91% and 0.98%, and the root mean square error is 1.03% and 1.13%, respectively, which verifies the validity and accuracy of the method. After adding noise with different signal-to-noise ratios to the original signals of different working conditions, the wavelet noise reduction can resist the noise interference in the actual environment to a certain extent and maintain the accuracy of SOC estimation.

With the rapid development of the global economy, offshore wind power generation technology has been advancing towards field group scale and industrialization, becoming a research hotspot in international renewable energy. However, to reduce the economic costs associated with deep-sea wind power technology and enhance the efficient of wind energy capture and utilization, the capacity of wind turbines has been gradually upgraded to 10 MW and above. This trend towards large capacity has consequently led to increased weight and volume of wind turbines, complicating offshore transportation, lifting, operation and maintenance, which limits further development of offshore wind power technology. Moreover, the significant volatility and intermittency of offshore wind power contribute to increased grid penetration issues, difficulties in large-scale grid connections, and a notable phenomenon of wind curtailment. Furthermore, the non-stationary wind power can cause grid voltage fluctuations, flicker, frequency fluctuations, harmonics and other power quality problems, affecting the stable operation of the grid.

To address these problems, Hunan University's wind power generation team proposed an innovative integrated technology for hydrogen production through offshore superconducting wind power generation. This innovative system utilizes water electrolysis to locally consume offshore wind energy, with the produced liquid hydrogen being transported to land via ships or pipelines for comprehensive utilization. Additionally, a liquid hydrogen circulation refrigeration system provides a stable low-temperature environment for superconducting wind turbines, significantly reducing platform volume and weight and ensuring the reliable operation of the integrated system.

The article provides an overview of recent development in HTS wind turbine technology and offshore wind power hydrogen production technology, both domestically and internationally. It analyzes the key structures and feasibility of the proposed innovative integrated system, highlighting how it compares to traditional technologies. Additionally, the article explores recent advancements in offshore wind power generation and transmission technologies. The discussion then shifts to the benefits of the proposed innovative technology in comparison to other existing technologies and schemes. It summarizes the advantages of integrating hydrogen production and offshore superconducting wind power generation, analyzes the variability of superconducting wind turbines output power and the limitations of current converter topology control strategies, and proposes the key technologies of designing superconducting wind turbines converter topology with efficient energy transfer capability and designing a superconducting wind power system friendly control strategy.

For the future development of the integrated system, an energy island system plan that is integrated with renewable energy development is proposed. This plan is based on the operational principles of each sub-structure and aims to harness the efficient synergy of renewable energies. Research will focus on determining the appropriate ratios for various energy production and conversion devices, which will optimize the configuration of multi-energy complementarity. This approach aims to establish an integrated energy system that reduces the standby capacity required by the system’s various equipment. Furthermore, this initiative will promote the coupling of the power with renewable energy systems, facilitating the synergistic development of electric power and green hydrogen. This strategy will improve the optimized configuration of the energy supply system and establish a common technological framework for large-scale superconducting wind power hydrogen production technology.

When multiple EMUs are simultaneously in a light-load starting condition within the traction network, it can lead to low-frequency oscillations in the traction network voltage. In severe cases, this may trigger traction locking, which poses a risk to the operational safety of high-speed trains. To address the issue of low-frequency oscillation, this study proposes a method based on virtual impedance for its suppression.

First, the impedance model of the EMUs-traction network coupling system is derived, and its stability is analyzed using the impedance ratio stability criterion and the Bode diagram. Second, based on the stability criterion, virtual impedance is incorporated into the EMU control strategy to correct the impedance characteristics of the load subsystem, thereby proposing a low-frequency oscillation suppression method. Third, an adaptive control method for virtual impedance is designed to handle the complex and dynamic working conditions, enabling the system to adjust the virtual impedance parameters and enhance the effectiveness of the suppression method. Finally, the proposed control strategy is compared with the traditional approach through a low-power experimental platform, validating the effectiveness of the suppression method.

The results of the system stability analysis indicate that the logarithmic amplitude-frequency characteristic curve of the system impedance ratio is significantly lower than 0 dB when only one unit (m=1) is connected to the traction network, and the absolute value of the system impedance ratio |T_sn(s)| is much less than 1, suggesting system stability. As the value of m increases, the amplitude-frequency characteristic curve approaches 0 dB, resulting in a decline in system stability. When m=6, the amplitude-frequency characteristic curve crosses 0 dB at 7.032 Hz, and the phase angle at the crossover frequency (f_c) is 182°. The absolute value of the phase angle exceeds 180°, and the system impedance ratio |T_sn(s)| does not meet the condition of being significantly less than 1, indicating an unstable state. Simulation results with the virtual impedance control strategy indicate that when six EMUs are connected to the traction network at the same time, when the virtual impedance R_p values are 10 Ω, 5 Ω, 2 Ω and 1 Ω respectively, the system voltage and current reach a steady state at 1.6 s, 1 s, 0.3 s, and 0.2 s, respectively. The suppression effect improves as the virtual impedance decreases. The simulation results with the adaptive virtual impedance control strategy show that when six EMUs are connected to the traction network, the system reaches a steady state in approximately 0.5 s. The voltage regulation time for the DC side of the EMUs is 0.46 s, with an overshoot of 37.71% and a post-stabilization voltage fluctuation of 58 V. Subsequently, one EMU with adaptive virtual impedance control is added every 2 seconds, resulting in a total of 11 EMUs. The system remains stable, and the virtual impedance value is reduced to 4.61. The comparative experimental results indicate that, after transitioning from the traditional control strategy to the adaptive virtual impedance control strategy, the system can rapidly recover from a low-frequency oscillation at approximately 7 Hz. Furthermore, the AC-side voltage stabilizes at 20 V, while the DC-side voltage remains stable at 40 V.

Based on the analysis, the following conclusions can be drawn: (1) A higher impedance ratio between the two sides of the EMUs-traction network coupling system results in decreased system stability. When the impedance ratio does not meet the stability criterion, system instability is induced, leading to low-frequency oscillations. (2) The incorporation of parallel virtual impedance in the control strategy for the EMU's four-quadrant converter can correct the impedance characteristics of the load subsystem, enhance system stability, and suppress low-frequency oscillation. (3) The adaptive virtual impedance control method can autonomously adjust the virtual impedance value based on the intensity of voltage oscillations on the network side of the EMU, thereby suppressing low-frequency oscillations under varying load conditions, improving the adaptability of the control strategy, and ensuring the stable operation of the EMUs-traction network coupling system across different operating conditions.

As the penetration rate of renewable energy sources increases, the stability mechanisms of the power system are constantly changing. The double-fed induction generator (DFIG), as a mainstream renewable energy equipment, its stability is of great significance to the safe operation of the power system. The phase-locked loop (PLL) plays an important role in synchronization, but there has been less research on simultaneously considering the dynamics of the phase-locked loop and the power balance loop. Moreover, the small-signal synchronization mechanism of DFIG within rotor speed timescale needs to be further analyzed.

Firstly, the transient model of single-DFIG infinite-bus system is constructed within the rotor speed scale. The simplified model is compared with the full-order model by Matlab/Simulink, and the results show that they match very well. Then, through bifurcation analysis, it is found that the system would experience small disturbance instability under weak grid condition. And it manifests itself in the form of low-frequency oscillatory instability. Furthermore, through the dominant modal analysis, it is found that the dominant unstable loop is the power balance loop.

In order to analyze the small-signal synchronous instability mechanism of the system, the model is linearized around the operating point. The linearized model and the full-order model are compared using Matlab under small disturbance, and the results validate the rationality of linearization. The power balance loop dominates the instability, making it considered as the core loop. Therefore, the Heffron-Philips model of the system is established for analyzing the small-signal synchronous stability mechanism. Based on the complex torque coefficient method, the terminal voltage control loop plays a dominant role by introducing negative damping. And by studying the transfer function of the PLL, it is found that the PLL with typical parameters has a negligible impact on the system in the rotor speed scale, and can be approximately regarded as a constant.

Finally, the parameters of the active outer loop and the reactive outer loop are analyzed. With the changing of the grid strength, the damping torque and synchronizing torque of each branch are quantitatively calculated. It is found that increasing the proportional coefficient of active outer loop and decreasing the integral coefficient will improve the stability of the system, and increasing the proportional/integral coefficient of the terminal voltage control loop will benefit the stability of the system. These analyses have been verified through simulations and experiments.

The conclusions of this paper are as follows: (1) In the rotor speed scale, the small-signal synchronous instability of the single-DFIG infinite-bus system is dominated by the power balance loop (active outer loop and rotor dynamic), rather than the PLL. (2) By constructing the Heffron-Philips model, it is found that the synchronous phase ∆θ_pllis approximately represented by an algebraic expression of the state variable ∆ω_r/∆θ_rof the rotor. The essence of synchronous instability lies in the instability caused by the state variable of the energy storage element.3) Using complex torque coefficient method, it is found that the terminal voltage control loop is the main factor that introduces negative damping. Through the analysis of the influence of parameters, it is found that increasing the proportional coefficient of the active power outer loop and decreasing the integral coefficient will improve system stability, and increasing the proportional/integral coefficient of the terminal voltage control loop will be beneficial to system stability.

PV systems are typically equipped with reactive power compensation devices when connected to the grid, and static synchronous compensator (STATCOM) devices are widely employed due to their flexible control capabilities. The increasing utilization of power electronic devices in the power grid has resulted in a shift from physical synchronization to control synchronization as the dominant mode of system operation. Analyzing static synchronization stability problem is more challenging for these systems compared to conventional power systems, as converter output characteristics are influenced by control strategies. Therefore, it is imperative to urgently address the problem of static synchronization stabilization under control strategy dominance.

First, this paper establishes the static synchronous stability analysis model of the grid-connected converter based on the control loop and circuit structure of each converter in a parallel system under the respective dq reference frame. The dq reference frame of the converter is determined by the phase information provided by the control loop in a multiple converter parallel system, thus enabling a unified coordinate system for static synchronization stability analysis. Subsequently, an equivalent small signal model is developed for analyzing multiple grid-connected converters. In comparison with existing coordinate conversion methods, Kirchhoff's current law is incorporated to enhance accuracy and reduce errors.

Then, the stability criterion for impedance analysis is enhance, and the static grid-synchronization performance indices are created. The small perturbation oscillation characteristics are measured using overshooting and regulation time, while the participation factor is employed to analyze the impact of each pole of the system. Finally, the attenuation coefficient is utilized to assess static synchronization stability performance.

The model is developed in Matlab/Simulink for simulation verification. Subsequently, an analysis is conducted on the impact of parameters such as the control loop parameters and STATCOM capacity on static synchronization stability. The main conclusions are summarized as follows:

(1) Optimizing the reactive power output of the grid-connected converter based on known control parameters can significantly enhance static synchronization stability performance, with the dominant influence of small perturbations after oscillation mode being attributed to poles generated by phase-locked loop control.

(2) The attenuation coefficient of the system exhibits a rapid increase in proximity to the critical stability region. Hence, it is imperative for the system to possess a certain margin of attenuation coefficient during operation. Based on the simulation analysis results presented in this study, static synchronous instability phenomena occur when the attenuation coefficient of the PV system exceeds 300. Conversely, when the attenuation coefficient falls below 250, the system remains in a state of static synchronous stability. These findings establish a criterion for analyzing and assessing static synchronization stability within such systems.

(3) The addition of STATCOM to the PV system primarily impacts the conductance matrix transfer function of the q-coupled channel. The phase-locked-loop coupling oscillations between the grid-connected converters do not affect the dd channel. Within the stable operating region, an increase in bandwidth for the DC voltage control loop, active current control loop, and reactive current control loop results in an amplification of both attenuation coefficient and system oscillation amplitude.

(4) When the phase-locked loop parameters of the STATCOM are the same as the PV system, the stability performance of the system is mainly affected by the grid impedance and the equivalent conductance transfer function of each grid-connected converter. And each grid-connected converter can independently connect to the grid and achieve stable operation to ensure that the system achieves static synchronous stability in this case.

(5) In cases where the active output of the PV system is low, STATCOM typically adjusts its capacitive or inductive reactive power provision to improve static synchronization stability performance. Conversely, when compensating for capacitive reactive power, utilizing STATCOM may yield superior results compared to using the PV grid-connected converter alone. Hence, allocating an optimal capacity for STATCOM can significantly enhance static synchronization stability performance.

As the penetration rate of renewable energy resources continues to increase, the traditional power system based on synchronous generators is evolving into a power system based on diversified power electronic equipment. The small disturbance stability analysis problem of multi-converter grid-connected system has attracted widespread attention. State-space model and impedance-based model are two main small disturbance stability analysis methods. Being as the white-box method, state-space model can be difficult to apply in practice because the differential equations describing the controllers of converters are not generally openly available due to commercial confidentiality. Impedance models have been popular in the field of power electronics for analysis of interactions between grid and converters. However, applying it directly to the stability analysis of multi-converter system will make the analysis process very complicated. Generally, the existing state-space and the impedance method still have room for improvement for the small disturbance stability analysis and sensitivity analysis of the oscillation mode of each converter.

Firstly, this paper proposes a single-input single-output (SISO) dq impedance stability criterion for analyzing the small disturbance stability of the multi-converter grid-connected system. Secondly, based on the formula of the stability criterion proposed, an expression for calculating the closed-loop pole of the system is derived. Because this formula is only a scalar function, the accuracy can be guaranteed for the usage of vector fitting (VF) method. Furthermore, a method for analyzing the sensitivity of oscillation modes to the impedance/admittance of each converter is proposed. This method can effectively evaluate the influence of different converters on the oscillation modes and help identify the dominant converter that causes oscillations. Finally, the accuracy of the proposed method is verified by Matlab/Simulink simulation and hardware-in-the-loop experiment.

The results are as follows: firstly, the proposed multi-converter system model can be used to represent the converter grid-connected system with any network structure and any number of grid-forming and grid-following converters. Based on the proposed method, it can be used to analyze the overall stability of the system as well as the influence of each converter on the system stability. Secondly, the proposed sensitivity analysis method can be used for evaluating which power converters are more sensitive to the close-loop poles and have a significant contribution to the harmonic instability.

The following conclusions can be drawn from the above results: (1) A recursive stability evaluation method for analyzing the stability of the multi-converter grid-connected system based on SISO dq impedance ratio is achieved, and a complete stability evaluation procedure is provided. Compared with the stability analysis method based on generalized Nyquist criterion, the stability analysis problem of a MIMO system is transformed into the stability analysis of a series of SISO systems, and the stability analysis of the whole system can be realized only by the impedance ratio of d-axis and q-axis in the stability analysis process. Because the SISO impedance ratio is used for stability analysis, the solution of the eigenvalues of the high-order return rate matrix required by the traditional method can be avoided, and the complexity of Nyquist plot analysis required for MIMO system can be effectively reduced. (2) In the proposed impedance stability criterion, dq impedance is adopted to model the VCI while dq admittance is used to describe the CCI, so the complicated procedure for obtaining the RHP open loop poles can be avoided. (3) Based on the proposed stability criterion proposed, an expression for calculating the closed-loop pole of the system is derived. Since this paper adopts dq coordinate system for modeling and derivation, compared with the sequential impedance model, the transfer function matrix elements can be guaranteed to be rational fractions, so the closed-loop poles can be obtained by VF method. (4) The sensitivity formula of the system's closed-loop poles on the dq admittance/impedance of the converter in the system is derived in this paper. Combining with the residual of the closed-loop poles obtained by the VF method, it can be used to analyze the influence of each converter on the key modes in the system. Therefore, it is helpful to identify the source that causes oscillatory instability.

The increasing deployment of wind turbines in challenging environments has led to the prevalent issue of converter faults, which significantly affect the reliability and efficiency of wind power systems. Given the critical role that the converter plays in optimizing the wind power conversion process, detecting and identifying open-circuit fault in wind converter is essential for maintaining operational integrity and maximizing energy output. Current fault identification methods often suffer from limitations related to robustness and computational complexity, necessitating improved solutions. To address these shortcomings, this paper introduces an innovative fault identification method that integrates analysis of the direct current (DC) bus voltage and rotor current characteristics. It can accurately recognize the single and double tube faults of converter power tubes.

Firstly，Fault detection is facilitated by the fact that the DC bus voltage signal is easily accessible, independent of the load and control strategy. Extraction of DC bus voltage edge gradients using mathematical morphology as a feature to detect the occurrence of faults. Secondly, the Pearson correlation coefficients of the rotor side currents are calculated to analyze waveform characteristics. The coupling relationship between the three-phase currents is theoretically deduced, and it is proved that the Pearson correlation coefficients between the two-phase currents are significantly different under different fault conditions, enabling precise identification of the fault phase. Moreover, the location of the fault bridge arm is determined using the average value of the current, enhancing the accuracy of fault identification. Finally, the decision function is used to locate the faulty power tube and realize the fault classification.

Simulation results of the open-circuit fault model of doubly-fed wind power converter show that the proposed method in this paper can accurately determine the occurrence of faults and locate the position of power tubes. By comparing under large data sets, it is found that the proposed method improves the accuracy while maintaining a shorter detection time compared to other methods, which is more practical and reliable. The simulation results show that the wind speed fluctuation has a negligible effect on the DC bus voltage and rotor current, and no fault occurrence is detected, while the current characteristics are stabilized in the range of the fault-free case, which indicates that the proposed method can overcome the interference of wind speed fluctuation. By simulating voltage dips to model the load fluctuations, it is found that the fault detection module misjudges the occurrence of faults, and the current characteristics is small affected but similar in size. It is judged that no faults have occurred, so the fault identification module can be used as a verification of fault detection. A Gaussian white noise with a signal-to-noise ratio of 20 dB is also added to the acquired voltage and current data, and the results show that the proposed method is not disturbed by noise.

The following conclusions can be drawn from the simulation analysis: (1) Compared with existing methods, the method is not only simple and effective in calculation, but also has a higher accuracy rate. (2) The fault detection method based on mathematical morphology utilizes the DC bus voltage, which is easy to obtain data and rapid to detect, and is not affected by noise. (3) The Pearson correlation coefficient-based fault classification method classifies the rotor three-phase currents according to their waveform correlation, and the consistency of theoretical and simulation results shows that the method is effective and of practical significance, and the method has strong robustness.

To monitor the dynamic characteristics of the receiving AC system and assess the risk of wideband oscillation in LCC-HVDC without additional equipment, this paper presents a non-invasive impedance wideband measurement method for LCC-HVDC systems. Unlike existing invasive methods, such as single voltage/current or wideband harmonic disturbance injections, the proposed method does not inject harmonic disturbances into LCC-HVDC, thereby avoiding the potential resonance risk. Furthermore, the proposed method introduces curve similarity and a dynamically adjusted time-step sampling scheme. First, this paper discusses the interaction principles between the LCC system and the receiving AC system. Secondly, this paper establishes the harmonic state space (HSS) impedance model for LCC-HVDC, based on HSS theory, focusing on the unipolar earth loop topology. Subsequently, a non-invasive impedance wideband measurement method is proposed to observe the operating state of the system, judge the risk of wideband oscillation, and analyze the interaction and wideband oscillation characteristics between the LCC system and the receiver network through impedance analysis. Finally, an example of CIGRE standard model is used to verify the correctness and practicality of the proposed method.

Based on the HSS theory and the concept of digital modulus, this paper introduces a non-invasive impedance wideband measurement method. Firstly, the system topology and the interaction principles of the LCC-HVDC unipolar earth loop subsystems are detailed. Then, the impedance modeling of the LCC-HVDC system is elaborated, and the HSS impedance mathematical model for wideband measurement is derived based on the system's topology. Finally, the system model is constructed in MATLAB/Simulink using parameters from the CIGRE standard model, and the impedance calculation model is applied to measure the impedance of the LCC system. Combined with the measured wideband impedance curve, the interaction between the LCC system and the receiving network, as well as the wideband oscillation characteristics, were analyzed using the generalized Nyquist criterion for stability. At the same time, the Pearson correlation coefficient was introduced to capture the similarity of impedance curves with high granularity, thereby exploring the adaptability of impedance models of different orders and the factors influencing the oscillation risk. It offers theoretical support for the measurement methods used in the engineering application of impedance measurement, and appropriately expanding the permissible error range in the amplitude-frequency intersection can enhance the universality of impedance analysis methods.

The main contributions and conclusions of this paper are summarized as follows: (1) A non-invasive LCC-HVDC impedance model and impedance wideband measurement method are designed, which does not require injecting harmonic disturbances into the system. The Pearson correlation coefficient between the measured impedance curve of the 13th-order model and the fitted curve from active measurements is 0.998 3, which is very close to 1. This high value indicates that the accuracy of the measurement model is high, suggesting a close match with the actual impedance characteristics. (2) By analyzing the similarity matrix of measurement curves from models of varying orders, it is observed that the Pearson correlation coefficients for the phase measurement curves are closely matched for orders h=3, h=7, and h=11, with the highest deviation being within 0.2%. Consequently, the acceptable error margin for amplitude-frequency intersection points may be suitably broadened. Considering computational efficiency, a lower-order measurement model within the range of 3 to 13 can be selected. (3) The characteristics of the measured wideband impedance curve of the system can be observed under two conditions: adaptive adjustment of the sampling time step and variation in transmission power. Adaptive adjustment of the time step is not only conducive to accurately identifying the crossover points of the amplitude-frequency response but also reduces the computational load and minimizes storage requirements. Conversely, an increase in transmission power raises the risk of system instability.

Current electric shock detection methods are primarily designed to address faults between the live wire and the ground wire, mainly relying on monitoring changes in residual current to identify issues. However, in the case of a neutral-to-live electric shock fault, the fault circuit often does not cause a significant change in the residual current. This presents a considerable challenge for existing detection methods when it comes to identifying neutral-to-live electric shock incidents.

To address the aforementioned issues, a low-voltage neutral-to-live electric shock faults detection method based on dynamic fault characteristics and a light gradient boosting machine has been proposed. Firstly, a 1:1 prototype experimental platform for a low-voltage distribution network was established in a real system. Under various operating scenarios involving multiple household loads, experiments reproducing live neutral shock faults were conducted alongside control experiments using a sliding resistor to replace the electrically shocked body. A substantial amount of experimental samples representing both fault and normal operating states was collected, creating a comprehensive database. Secondly, the complexity of neutral-to-live electric shock faults is assessed based on the interference of load current on fault current. A fault circuit electrical equivalent model is established by considering the dynamic resistance and breakdown arcs at the dual contact points of the neutral-to-live shock, in conjunction with biological dynamic impedance. The impact of fault current on the main circuit current is analyzed. Finally, features of the main circuit current are extracted from the perspective of magnitude and high-frequency components, and the temporal changes of individual features before and after the occurrence of faults are compared. Given the difficulty in clearly distinguishing between fault and non-fault states based on individual features alone, along with the fact that these features exhibit varying sensitivity to both states, a multidimensional representation of the system state is employed. Following an ensemble computational approach, a lightweight gradient boosting machine model is developed, leveraging its uni-directional gradient sampling method and ensemble operation mechanism to accurately classify the two states.

The proposed method was evaluated on a test dataset consisting of 50 666 samples, achieving an overall accuracy of 96.82%. Specifically, the identification accuracy for 35 831 normal samples was 97.50%, while the accuracy for 14 835 neutral-to-live electric shock faults was 95.17%. The test results indicated that the proposed method could accurately distinguish neutral-to-live electric shock faults from normal operating conditions, including those in the control group with the sliding rheostat added, even when the fault information was significantly obscured by high load currents. Compared to existing methods, the proposed approach shows an advantage in accurately detecting low-voltage neutral-to-live electric shock faults.

The following conclusions can be drawn from the analysis: (1) By incorporating the time-varying impedance of biological tissues, variations in contact resistance, and breakdown arcs, the dynamic characteristics of faults were examined, revealing two effects of neutral-to-live electric shock faults on the main circuit current: changes in current magnitude and variations in high-frequency components. These findings served as the basis for constructing feature vectors. (2) The contribution of individual features to distinguishing between neutral-to-live electric shock faults and normal operating conditions is limited, resulting in significant inter-class ambiguity that can easily disrupt the sample fitting performance of traditional pattern recognition models. However, if features can exhibit a certain degree of sensitivity across different classes, the combination of multidimensional features can facilitate comprehensive discrimination. (3) Due to its inherent resilience to disturbances, the ensemble model can effectively mitigate interference caused by inter-class ambiguity and demonstrate strong generalization capabilities.

Hydrogen energy system, with its inter temporal and spatial transfer characteristics, shows great potential for enhancing the resilience of distribution grids. However, few literatures have considered the inter temporal and spatial flexibility of hydrogen energy system and the inter-regional support capability of mobile emergency resources, and the post-disaster collaborative recovery mechanism of multi-region electric-hydrogen integrated energy system (MR-EH-IES) is still unclear, which makes it difficult to exploit the inter-regional support potential of mobile resilience resources. Aiming at the above problems, this paper proposes a post-disaster recovery strategy for MR-EH-IES with cross-regional resource sharing.

This paper firstly proposes a two-layer MR-EH-IES disaster recovery framework based on the idea of “intra-regional autonomy, resource integration, inter-regional sharing”. In the lower layer, the electric-hydrogen integrated energy system (EH-IES) carries out intra-zone autonomy. The potential of synergistic cooperation between mobile electric energy storage, hydrogen fuel power generation vehicles, maintenance personnel and hydrogen energy system in disaster recovery is fully considered, and the EH-IES disaster recovery model considering the synergistic scheduling of distributed power sources and maintenance personnel is established. At the upper level, the joint disaster resilience center carries out the coordinated allocation of mobile resilience resource (MRR). Starting from the disaster recovery mechanism of different types of MRR, the key factors affecting its allocation are analyzed, the MRR disaster recovery mechanism considering cross-region support is proposed, and the MRR disaster allocation model considering cross-region resource sharing is established. Then, based on the above framework and strategy, the MR-EH-IES two-layer disaster recovery model considering cross-region resource sharing is proposed.

The simulation analysis shows that the total cut-load loss of MR-EH-IES decreases by 22.4% after considering cross-region resource sharing, in which the cut-load loss of region 1 and region 3 increases slightly by ¥1.3×10³ and ¥7.3×10³, respectively, while the cut-load loss of region 2 and region 4 decreases by ¥157.8×10³ and ¥62.5×10³, respectively. Specifically, in the early stage of disaster recovery, when the mobile power supply left from region 1 and region 3 to support region 2 and region 4, the weighted load recovery rate of region 1 and region 3 showed a short drop, with the maximum drop of 0.5% and 1.1%, respectively, but both of them were higher than the weighted proportion of important loads. Meanwhile, the load-weighted recovery rates of region 2 and region 4 increased significantly, with maximum enhancements of 11.0% and 4.2%, respectively. In addition, when region 1 and region 3 were restored, idle mobile power supplies and maintenance personnel were the first to support other regions.

The following conclusions can be drawn from the simulation analysis: (1) The post-disaster recovery strategy proposed in this paper is able to rapidly restore the supply of important loads and reduce the system damage in the early stage of disaster recovery through the reasonable allocation of mobile emergency resources, and improve the utilization rate of mobile emergency resources in the later stage of disaster recovery. (2) The inter temporal and spatial flexibility of the hydrogen system and the long tube trailer can increase the energy transfer channels of MR-EH-IES in time and space scales, giving full play to the ability of hydrogen energy system to support the power grid in disaster recovery.