Abstract:Blade icing frequently occurs on wind turbines operating in cold weather conditions, leading to reduced power output, unstable equipment operation, and even severe mechanical failures. Therefore, developing effective early warning methods for wind turbine icing is of great practical significance. In this study, Supervisory Control and Data Acquisition (SCADA) operational data are analyzed, and key features are constructed based on wind speed, power output, and ambient temperature. An early warning model for blade icing events is established using a random forest algorithm. In addition, real-time monitoring of ice thickness is achieved through a rotating cylindrical array device, based on which a real-time icing early warning model and a dynamic warning mechanism are developed. A 3.2 MW wind turbine at the Wanbao Wind Farm in Chongqing is used as a case study to validate the proposed approach. The results show that the icing occurrence warning model achieves a classification accuracy exceeding 95%, and warning signals are issued multiple times within 1 h prior to blade icing events. Furthermore, the real-time warning model continues to generate alerts after icing occurs, demonstrating its capability to continuously track the evolution of the turbine icing environment. Overall, the proposed dynamic early warning model provides effective decision support for the safe operation and efficient management of wind turbines.

Abstract:Power capacitors are widely used for reactive power compensation, harmonic filtering, carrier transmission, and high-frequency protection in power systems. In practical operating conditions, the voltage applied to power capacitors contains significant harmonic components, resulting in complex loss characteristics. To accurately characterize the active power losses of capacitors under harmonic voltages, a wideband distributed equivalent model based on the extended Debye model is proposed, taking into account the actual physical structure of the capacitor. Analytical expressions for the voltage and current distributions on the capacitor plates are derived, and their spatial characteristics are systematically analyzed. By considering both electrode (plate) losses and dielectric losses, the distribution of active power loss within the capacitor is investigated, and the corresponding rules are summarized. The results indicate that harmonic voltages increase the amplitude of the current flowing on the capacitor plates, leading to a nonlinear current distribution. Moreover, as the voltage frequency increases, the active power loss of the capacitor rises significantly, and in the high-frequency range, the plate losses become the dominant contributors to the total active power loss. Based on these findings, a loss reduction strategy is proposed, in which the number of voltage input points is increased to promote a more uniform electrode current distribution, thereby effectively reducing the active power loss of the capacitor.

Abstract:Fault detection of railroad signal cables is of great significance for ensuring the safe operation of railroad systems. To address the most common high-resistance fault problem in railroad signal cables, the inhomogeneous characteristics of cable structures are analyzed by measuring the electrical parameters of multiple 1 m cable segments. Based on these measurements, a normal model of inhomogeneous cable parameters is constructed, followed by the establishment of an inhomogeneous high-resistance fault simulation model for railroad signal cables. The feasibility of the frequency-domain reflectometry (FDR) method for locating high-resistance faults in railroad signal cables is then verified. Furthermore, a dedicated hardware system based on the FDR method is independently designed. This system transmits a 0.1 MHz to 5 MHz swept-frequency signal into the cable and collects the reflected signals required for FDR analysis through an intermediate-frequency detector. The collected data are subsequently processed on a host computer to realize the localization of high-resistance faults with resistance values of 0.1 MΩ and 0.5 MΩ within a cable length of 1 000 m. Experimental results confirm the effectiveness and accuracy of the FDR method for detecting and locating high-resistance faults in railroad signal cables.

Abstract:In large hub city power grids, load centers often lack flexible and continuously adjustable reactive power sources across multiple voltage levels, leading to a “hollowing-out” of power supply. This paper proposes a coordinated two-level voltage control optimization method tailored for such “hollowed-out” power grids. In this method, the high-voltage busbar of all controllable power plants within a region are generalized as voltage-dominant nodes, and regional voltage regulation is achieved by minimizing the sum of squared deviations between the busbar voltages and their reference values. Meanwhile, the coordination relationships among upper-level and lower-level regions are explicitly distinguished, and threshold-based control objectives between hierarchical regions are introduced to improve the reactive power support capability of lower-level areas. The effectiveness of the proposed method for regional reactive voltage control in hollowed-out power networks is verified through simulations using a standard test system combined with actual grid data from China.

Abstract:The equal-cycle preventive maintenance strategy is widely used due to its simplicity and ease of implementation. However, existing optimization models typically adopt a coarse modeling granularity and rarely account for fault differences, particularly the varying effects of preventive maintenance on different failure types. To address these limitations, this study takes the meta-action unit as the basic research object and classifies unit failures into damage failures and intrinsic fatigue failures, while explicitly considering the differing impacts of preventive maintenance on these two failure types. A hybrid failure-rate model is used to characterize imperfect maintenance effects, and an optimization model for an equal-cycle imperfect preventive maintenance strategy is developed. Numerical analyses are performed to verify the effectiveness of the proposed model and to investigate the influence of different maintenance cost structures on the optimal maintenance cost rate and maintenance strategy. The results indicate that neglecting differences in preventive maintenance effects leads to an underestimation of the maintenance cost rate of the meta-action unit, and that both maintenance cost types and the proportion of failure modes have a significant impact on the determination of the optimal maintenance strategy.

Abstract:To examine in-cabin noise characteristics under different load conditions and vehicle speeds, psychophysical objective parameters are measured and subjective ratings are obtained through paired-comparison jury tests. The effects of factors such as vehicle model, speed and load on the in-cabin sound quality of heavy-duty commercial vehicles are systematically analyzed. Furthermore, an analysis of the correlations among acoustic quality parameters under different operating conditions was conducted. Subsequently, a multiple linear regression model is constructed using ridge regression, in which psychophysical objective parameters serve as independent variables and subjective ratings as the dependent variable. The results indicate that ridge regression significantly reduces multicollinearity while maintaining a satisfactory level of predictive accuracy, thus providing valuable guidance for sound quality optimization.

Abstract:Microfocus X-ray digital radiography (DR) offers high image resolution and is therefore well suited for the nondestructive inspection of small-diameter aero-engine tube welds. To optimize the inspection process parameters for microfocus DR applied to small-diameter aero-engine tube welds, a numerical model describing the relationship between process parameters and DR image quality was developed to guide parameter selection. First, a step wedge made of the same material as the aero-engine small-diameter tubes was used to simulate different X-ray penetration thicknesses of the workpiece. A quadratic regression model was then established to characterize the relationships between key process parameters (tube voltage, tube current, magnification, and penetration thickness, etc.) and DR image quality indexes (spatial resolution and contrast-to-noise ratio). Then, three small-diameter aero-engine tubes with different wall thicknesses were selected as representative workpieces, and the optimal sequence of process parameters was determined using a non-dominated genetic algorithm. Experimental results show that the measured DR image quality indices closely match the predicted values of the model, demonstrating that the proposed quadratic regression model can effectively predict the relationship between process parameters and DR image quality. Compared with conventional methods that adjust process parameters individually, the proposed method exhibits higher efficiency and stronger guidance and can be extended to similar DR inspection applications.

Abstract:Metal inert gas (MIG) welding provides advantages such as a high deposition rate, high thermal efficiency, and suitability for automated and efficient production. Due to their high specific strength and good damping performance, magnesium alloys have broad application prospects in transportation, aerospace, and other industries. MIG welding is considered a suitable joining method for magnesium alloys, and significant research efforts and technological developments have been devoted to this field. This paper presents the current research status of MIG welding of magnesium alloys, reviews related studies on droplet transfer modes, microstructure evolution, and MIG hybrid welding technologies, and discusses future research directions for MIG welding of magnesium alloys.

Abstract:Rocking walls offer advantages in controlling deformation modes and preventing weak-story failures. Step-terrace frame structures suffer from significant structural deficiencies, including irregular lateral stiffness distribution, abrupt changes in floor load-bearing capacity, and concentrated deformation in upper stories. To improve the yielding mechanism and mitigate unfavorable failure modes, a bottom-hinged rocking wall was introduced to regulate the structural response. Furthermore, to reduce seismic responses and limit residual deformations, a step-terrace frame-energy-dissipating rocking wall structural system incorporating buckling-restrained braces (BRB) was proposed. Numerical models of ordinary step-terrace frame structures, step-terrace frame-rocking wall structures, and step-terrace frame-energy-dissipating rocking wall structures were developed for a seismic intensity 7 region (0.15g) using OpenSEES. The reliability of the numerical models and material parameters was verified by simulating low-cycle reversed loading tests of step-terrace frame structures. Incremental dynamic analysis (IDA) was then performed to systematically evaluate the seismic fragility of the step-terrace frame-energy-dissipating rocking wall structure, considering IDA curve clusters, partition curves, seismic demand probability models, fragility functions, damage-state exceedance probabilities, fragility indices, collapse resistance reserve factors, and safety margin ratios. The results indicate that, under seismic excitations with the same peak ground acceleration (PGA), rocking walls effectively limit the development of structural plasticity, reduce response dispersion, and decrease the probability of exceeding various performance levels, thereby exhibiting enhanced collapse resistance. The incorporation of BRB further improves the seismic performance and collapse resistance of the rocking wall-frame system. Among the investigated 3 kinds of structures, the step-terrace frame-energy dissipating rocking wall structure exhibits the best seismic behavior and collapse performance, while the conventional step-terrace frame structure shows the poorest performance.

Abstract:To study the dynamic mechanical behavior of concrete under three-dimensional coupled static-dynamic loading, split Hopkinson pressure bar (SHPB) experiments were carried out under combined axial compression, confining pressure, and impact loading. A stress initialization method was introduced into the LS-DYNA program, and a sequential analysis approach consisting of implicit static loading followed by explicit transient dynamic loading was used to simulate the coupled static-dynamic loading process. The influence of different static load combinations on the strength and failure characteristics of concrete was systematically analyzed. The results show that, at the same impact velocity, the dynamic compressive strength of concrete increases with increasing confining pressure, which provides a protective effect on the specimen. A critical axial compression value is observed: below this threshold, axial compression enhances specimen strength, whereas beyond it, axial compression leads to strength deterioration. The introduction of the stress initialization method enables accurate realization of constant pre-stress conditions before dynamic analysis. Comparative analysis between experimental and numerical results shows that static loading provides limited compaction enhancement, while changes in static load combination significantly alter the internal stress distribution, which is the main factor affecting the dynamic strength of concrete. Numerical simulations effectively capture the failure process of concrete under three-dimensional coupled static-dynamic loading, revealing that the dominant failure mode is compressive-shear failure. Furthermore, the damage evolution of concrete can be reliably predicted by analyzing the time-history curves of damage variables in the numerical model.