Iet Science Measurement & Technology最新文献

Accurate evaluation of electromagnetic shielding effectiveness (SE) is crucial for protecting modern electronic systems against electromagnetic interference (EMI) and transient disturbances such as electromagnetic pulses (EMP). This study investigates both time-domain shielding effectiveness (TDSE) and frequency-domain shielding effectiveness (FDSE) of metallic grid structures on dielectric substrates. TDSE metrics, including Peak SE and Derivative SE for electric and magnetic fields, quantify the attenuation of both field amplitude and its temporal rate of change under transient EM exposure. Full-wave simulations using the Finite Integration Technique (FIT) in computer simulation technology microwave studio (CST-MWS) were performed to generate datasets for training multilayer perceptron (MLP) neural networks. The MLP models map five structural and material parameters—aperture width, metal thickness, substrate thickness, relative permittivity, and loss tangent—to TDSE and FDSE responses. For TDSE prediction, the trained network achieves a root mean square error (RMSE) of 0.02215 and R² of 0.9846 on test data, demonstrating high predictive accuracy. For FDSE prediction across 1–4 GHz, the network provides close agreement with simulated spectra. Furthermore, a neural network-based surrogate model is employed for rapid optimisation of metallic grid design under target shielding criteria. Comparisons with the Trust Region Framework in CST show that the surrogate-based approach maintains high accuracy while significantly reducing computational time and optimisation cost. The proposed methodology enables efficient evaluation, prediction, and optimisation of metallic grid configurations for electromagnetic shielding applications under transient conditions.

High-voltage direct-current gas-insulated transmission lines are employed due to their advantages considering safety, voltage capacity, self-healing of the gas and reliability. However, under DC condition, high electric field stress and charge accumulation may occur, which can lead to partial discharges and therefore to insulation failures. The control of the electric field within the system is critical for ensuring operational reliability. Field control techniques are applied to regulate the electric field distribution. This work investigates the optimisation of the field control technique functional grading materials, where the spatial distribution of conductive filler particles within the spacer is adjusted to obtain a minimised and homogenised electric field distribution. Therefore, a deep neural network is utilised as a surrogate model to find the optimised distribution to keep the computational costs lower compared to conventional FEM simulations, since several simulations need to be performed for the optimisation. The results show that the DNN-based surrogate model enables a computationally efficient optimisation of the HVDC GIL spacer.

Accurate and efficient diagnosis of motor eccentricity faults is crucial for the reliable operation of electric machines. This requires large volumes of high-quality data across various operating conditions and fault types. While experimental data collection is costly and time-consuming, simulations offer a more practical alternative. However, full-order models are often computationally prohibitive due to magnetic saturation nonlinearities and repeated simulations across different conditions. This challenge motivates the development of efficient model order reduction (MOR) techniques. To alleviate the computational cost associated with nonlinear problems, look-up table (LUT) interpolation can be used to approximate nonlinear operators, thus avoiding convergence issues commonly encountered in hyper-reduction methods. However, conventional LUT methods face significant memory demands, especially in multi-parameter settings. To address these limitations, a hybrid MOR framework is proposed, based on a two-stage approach leveraging tensor decomposition, tailored for nonlinear, multi-parametric electromagnetic problems in motor diagnostics. The two-stage strategy is applied to both the LUT interpolation part and the MOR process. The framework is validated on a permanent magnet synchronous motor with static eccentricity, demonstrating superior accuracy and efficiency compared to traditional MOR techniques, and showing great potential for surrogate modeling in motor fault diagnosis.

As a critical component of unmanned aerial vehicle (UAV) data link systems, networking radios used to create wireless ad-hoc networks demonstrate considerable susceptibility to ultra-wideband ultra-wideband electromagnetic pulse (UWB-EMP) irradiation, resulting in severe degradation of communication performance. This paper presents a systematic investigation based on irradiation experiments, which clearly reveal front-door interference effects in networking radios exposed to UWB-EMP. To elucidate the underlying interference mechanisms, a QPSK communication system model and an RF module circuit model were developed. These models indicate that the observed blocking interference is primarily attributable to non-linear effects within the RF front-end. Throughout the study, the BER was employed as the evaluation metric for both irradiation tests and QPSK system simulations, while the output waveform of the low-noise amplifier served as the key indicator in circuit-level simulations. Experimental results indicate that UWB-EMP irradiation induces communication blackouts, with a data transmission success rate of only 66.7% within the specified duration. The antenna effectively couples UWB-EMP energy into the communication system, causing nonlinear distortion in the RF module. This distortion significantly increases the BER, reaching up to 50% under the given simulation conditions and may even lead to complete system failure. Detailed analysis shows that: (1) at 5 kV irradiation, the LNA enters nonlinear operation, losing amplification capability for 3–7 ns before recovery; (2) at 10 kV, transistor damage becomes more severe, extending the LNA's dysfunction period to 3–30 ns; and (3) at 30 kV and 60 kV, complete transistor breakdown occurs, with recovery times prolonged to 70 ns and 105 ns, respectively. These findings provide essential theoretical and experimental foundations for enhancing the electromagnetic compatibility and protection of networking radios in ultra-wideband electromagnetic pulse (EMP) environments.

High-penetration renewable energy and power electronic devices make high-order harmonic voltages much concerned in modern power systems. Accurate measurement of harmonic voltages facilitates harmonic source localisation, harmonic mitigation and power quality evaluation. A series resistive-capacitive divider for harmonic voltage measurement of gas-insulated switchgear was studied. It has a harmonic voltage amplification characteristic, which enables accurate measurement of high-order, low-amplitude harmonics. In this manuscript, a design scheme of the series resistive-capacitive divider was presented and an equivalent circuit was built. Using the finite element method, the partial capacitances in the equivalent circuit were calculated. Then, they were calibrated through the experiment. Finally, based on the calibrated capacitances, the effectiveness of the series resistive-capacitive divider in improving the harmonic voltage measurement accuracy was experimentally verified. It amplifies harmonic components, thereby fully utilising the resolution of the analogue-to-digital converter, increasing the significant figure for the harmonic voltage measurement and enabling the accurate measurement of harmonics.

This study established four models representing common defect types in 25-kV cross-linked polyethylene power cable joints and analysed partial discharge (PD) signals associated with such defects. An acoustic emission sensor was used to measure acoustic signals induced by the PD phenomenon in the power cable joints. A chaotic synchronisation system was applied to analyse dynamic errors associated with the signals. Motion trajectories were calculated using the master and slave systems in the Chen-Lee chaotic system and then plotted as three-dimensional error trace diagrams for each of the four models. Two chaotic gravity distances were obtained as features and a backpropagation neural network algorithm was used for cable joint pattern recognition. Furthermore, acoustic signals were measured from 160 sets of cable joints and random white noise was added to test the robustness of the feature extraction algorithm against noise. This study also evaluated the use of fractal dimensions for extracting features from 3D PD patterns. The results confirmed that the proposed method achieved high accuracy, was straightforward to implement and effectively distinguished between different cable joint models.

Current partial discharge (PD) recognition models are often constrained by limitations such as insufficient recognition accuracy, limited processing speed, and procedural complexity. To mitigate these limitations, time-frequency multi-scale residual attention network (TFMRAnet) is designed to analyse PD signal, which comprises a multi-scale residual attention-based adaptive denoising network, a frequency-domain recognition network, and a decision fusion module based on Dempster–Shafer (D–S) evidence theory. Specifically, the multi-scale residual attention adaptive denoising module is used to extract multi-scale features by dilated convolutions of different scales and accelerate training convergence by residual connections. Moreover, a GAM attention mechanism and an adaptive soft-thresholding function are used for denoising, which preserves PD information and amplifies cross-dimensional global interactions, thereby improving the performance of the network model. In frequency domain recognition networks, frequency domain features are extracted by performing Fourier transforms on PD signals. The recognition results of the time-domain and frequency-domain models are deeply combined through the D-S evidence theory to improve the multidimensional recognition capability for PD patterns. Finally, a PD experimental platform was built to create four representative PD fault models and generate PD datasets corresponding to distinct fault types. Results from comprehensive comparative and ablation experiments demonstrate that the proposed model achieves perfect recognition accuracy (100%) on the collected PD dataset and exhibits the fastest inference speed (0.643 ms). These findings underscore its significant potential for practical engineering applications.

Hysteresis modeling is widely employed in diverse disciplines such as engineering, physics, biology, chemistry, and mathematics for representing these intriguing nonlinear phenomena. More specifically in electrical engineering, classical models, facilitating progress in applications such as electric machines. Nonetheless, these models encounter limitations due to their mathematical intricacy and corresponding parameter estimation procedures. The more recent limiting loop proximity (L2P) model has emerged as a promising alternative, offering high precision in modeling magnetic hysteresis while requiring fewer parameters. However, the L2P model is better suited for symmetric and centered hysteresis, making it less effective in scenarios that require shifted or asymmetric hysteresis. To address this limitation, this paper introduces an enhanced model, termed the logistic limiting loop proximity (L3P) model, which incorporates the logistic equation to refine the characterization of non-centered hysteresis curves. By integrating the logistic function, the L3P framework significantly expands the model's applicability across diverse fields, particularly those governed by logistic dynamics. In this way, the proposed method not only improves the accuracy of non-centered hysteresis modeling but also ensures a lower-complexity formulation with a moderate number of parameters. Simulations validate the approach, demonstrating its ability to generate a wide range of shifted and asymmetric hysteresis curves.

The pulsed eddy current method is an indispensable part of non-destructive testing, capable of effectively detecting defects in ferromagnetic materials and locating the faults. This paper proposes an orthogonal pulsed eddy current detection method considering thermal effects. A 1+3 pulsed eddy current combined coil is designed. A set of directional coils is used as the excitation, and three sets of orthogonal coils are used as the receivers. By collecting the induced signals of the three-dimensional receiving coils, the relationship between the temperature-coupled coil output signal and the defect function of the metal component is established. An integrated detection hardware device for transmission-reception is designed, and conductor plate defect tests are carried out. The tests show that this method can effectively detect metal component defects, expanding the traditional one-dimensional measurement method to three-dimensional detection, increasing the available information, and proving the effectiveness and practicality of the new probe.

Microtopographic features significantly influence lightning faults in transmission line towers. However, the mechanisms by which microtopography affects the lightning trip rate are complex, and the results of existing calculation methods often deviate substantially from actual observations. Current research still lacks effective methods to accurately quantify this influence. To improve the accuracy of lightning trip rate simulations, this study automatically identified and extracted microtopographic features of transmission line towers in Yunnan using geographic information software and surface runoff simulations. A BP neural network model was developed using the actual lightning trip rate, simulated trip rate, ground flash density, tower parameters, and microtopographic indices as inputs. The corrected lightning trip rate achieved a relative error within ± 20% compared with historical statistics. This study provides valuable data support for improving lightning protection designs in power systems.