The pollution level of silicone rubber insulators critically impacts power transmission reliability. This study proposes an automated recognition system combining Laser-Induced Breakdown Spectroscopy (LIBS) and machine learning for efficient, non-destructive assessment of insulator contamination, overcoming limitations of conventional laboratory-based methods. Spectral data from insulator sheds were collected via LIBS. Five machine learning models—Backpropagation Neural Network (BPNN), Genetic Algorithm-optimized BPNN (GA-BPNN), Convolutional Neural Network (CNN), Radial Basis Function Neural Network (RBFNN), and Random Forest (RF)—were trained for pollution classification. Results demonstrate that spectral data from the second and third laser pulses yield optimal detection accuracy, overcoming surface heterogeneity limitations of the first pulse. Among the models, the RF algorithm achieves the best balance of high accuracy and computational efficiency, attaining overall accuracies of 97.00% (combined 2nd+3rd pulses), 96.60% (2nd pulse alone), and 96.20% (3rd pulse alone) with processing times of 4.72 s, 4.41 s, and 4.40 s, respectively. While CNN achieved the highest accuracy (98.30% for 2nd+3rd pulses), its computational time (73.75 s) was significantly longer. This integrated LIBS-machine learning framework provides a real-time, field-deployable solution for precise pollution classification on transmission line insulators.

This paper describes semi-analytical calculation methods for the positive-sequence and zero-sequence impedances of 3-phase armored cables with solid sheaths and a single armor layer. Separate equivalent circuits are defined for these two impedances. Subsequently, circuit component values are calculated based on 2D field equations, with some 3D effects modeled. The existing approach for calculating zero-sequence impedance is corrected and expanded by inclusion of additional inductances due to sole no idal fields and magnetic armor effects. 3D finite-element models of cables with magnetic and non magnetic armor are used to assess the accuracy of our circuit models; discrepancies are less than 3% at frequencies between 50 Hz and 5 kHz. Python code with the implementation of the method is provided for accessibility.

In the previous part, simulation works were conducted to analyze the field grading effect of nonlinear materials in the application of bushing, and the aim of this paper is to present the experimental investigations. The adaptive field grading (AFG) bushing samples are prepared by a novel method based on wet winding technology, and the assembly quality is verified by scanning electron microscope, thermal gravimetric analyzer, and conductivity test of slice samples with multi-layered structure. The initial partial discharge voltage and phase-resolved partial discharge pattern characteristics of the AFG bushing samples are obtained. Taking S<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> without a field grading layer as the control group, it exhibited noticeable internal partial discharge at 36 kV. At the same time, S<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> and S<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> samples with nonlinear materials had higher initial partial discharge voltages, which are 66.67% and 38.89% higher than those of the former samples, respectively. The insulation strengths under the lightning impulse voltage of the AFG bushing samples are also tested. The S<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> sample was broken down under severe high-voltage impulses, while the S<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> sample remained well-insulated after 15 positive and negative lightning impulses. It is hoped that this work will provide an innovative technical roadmap for bushing design and manufacturing.

This paper presents a novel expression for the partial mutual inductance between the core and screen of a coaxial or pipe-type cable. This partial mutual inductance describes the electromotive force in the screen when there is current in the core and not the screen. This differs from the conventional expression for the inductance of a coaxial cable whose derivation supposes balanced opposing currents flowing in the core and screen. The magnetic field generated by a time-varying current in the core with no balancing current in the screen is an expanding field of infinite extent. This contrasts with the loop inductance of a coaxial cable, which exists only within the cable confines. The expanding nature of the field through a surface limits the forms of Faraday's Law that can be applied to describe the inductance being defined. The partial mutual inductance is determined by applying Maxwell-Faraday's Law to a surface with two closed but independent boundaries. The result aligns with field tests conducted on coaxial power cables, which indicate that the electromotive force generated by the current within the coaxial screen depends solely on the cable length and the frequency of excitation. Part II of this paper presents experimental support for this theory.

This paper describes a Matlab toolbox (MFT-NNLS) for passive macromodeling of large-scale problems defined by admittance, impedance or scattering data in the frequency domain. Vector Fitting (VF) is used for calculating an initial rational model having a common pole set. Passivity can afterwards be enforced by residue perturbation of the associated pole-residue model. Highly efficient implementations are utilized, including the so-called fast implementation of VF, and passivity enforcement with problem compacting via the QR-NNLS method with usage of a fast LU-based non-negative least squares solver. The final result is a rational model satisfying the physical conditions of realness, stability, passivity and reciprocity. The rational model is returned on alternative forms, including a pole-residue model and a state-space model with real-valued or complex-valued parameters. Optional parameters are available for tuning the behavior of the software to the given application, e.g., least squares weighting schemes and alternative methods for passivity assessment. Model export to leading electromagnetic transient programs is included. Several applications are shown, including large-scale problems from high-voltage power systems (high orders, many terminals) and high-speed electronics (many ports). The software consists of a collection of Matlab functions.

Gas relays serve as vital non-electrical protective elements for transformers but occasionally experience maloperations in their heavy gas action during actual operation. To investigate the underlying causes, this study explores the response of heavy-gas protection in BF-type double-float gas relays to pipeline vibration. The constructed experimental platform, coupled with a fluid-solid coupling dynamic model, characterises heavy gas action behaviors under transient oil flow and pipeline vibration excitations. Simulation and experimental results show a downward trend in the heavy gas action flow velocity of the BF-type double-float gas relay with increasing pipeline vibration. Specifically, numerical findings reveal that pipeline vibrations of 32–40 g reduce the heavy gas setting flow velocity by 12.34%–14.29%. Meanwhile, fluid-solid coupling simulations confirm that vibration enhances the turbulence intensity of the flow field and unsteady pressure fluctuations, which disrupt the force balance by superimposing vibration torque, leading to premature triggering of the action. The proposed vibration-oil flow coupling model reveals how pipeline vibration triggers relay maloperation. These findings provide a theoretical basis for optimizing setting values and developing anti-maloperation strategies.

Modifying the inverter of line-commutated converter- based high-voltage direct current (LCC-HVDC) to hybrid commutated converter (HCC) is currently a highly effective solution for mitigating the risk of commutation failure (CF). This paper investigates the transient performance of HCC from five perspectives while retaining the LCC-HVDC control system. Since constant extinction angle (CEA) controller leads to significant reduction in power transmission, constant voltage (CV) controller is considered a more appropriate choice as the main controller during AC faults. In addition, the extreme case of insufficient forced commutation capacity leading to CF is considered. It is revealed for the first time that the voltage clamping effect of the arrester will cause a new CF phenomenon. There will be no short circuit on the DC side and the fault current is greatly reduced. This leads to insufficient regulation capability of the CV controller, resulting in continuous CFs. Given the limited fault ride-through capability of the selected CV controller, this paper proposes an adaptive order regulation control (AORC) based on expected reactive power. It can adjust the DC current and voltage in a targeted manner according to AC faults, and determine the maximum allowable upper limit of the firing angle. This improves the reliability of HCC in resisting CF. Under extreme cases, it also effectively prevents the continuous CFs. The simulation results in PSCAD/EMTDC verify the theoretical analyses and the proposed control method.