TL;DR
A two-stage WT-PSO-RBFNN framework in which adaptive db4 level-3 wavelet thresholding performs fault detection and a PSO-optimized radial basis function neural network performs phase/ground classification using normalized wavelet-domain features is proposed.
The proposed framework employs ANN as a nonlinear feature embedding and a Radial Basis Function SVM subsequently classifies using an Error-Correcting Output Codes (ECOC) strategy, validating the effectiveness of the proposed hybridisation strategy for intelligent transmission system protection.
The results indicate that phase-angle information provides supplementary and class dependent discriminative value, but does not consistently improve all fault classes, whereas conventional voltage and current measurements alone represent a simpler and more stable alternative, whereas phase-angle measurements may be incorporated when synchronized phasor information is already available.
Large-scale renewables are weak-feed and their output current is controlled. That doesn't play nice with traditional pilot protection, which only looks at power-frequency components. When a fault happens, the current from these sources has a "low fundamental, high transient" signature—so the protection often ends up tripping when it shouldn't. This paper proposes a novel adaptive protection scheme utilizing a hierarchical weighted Euclidean distance. It constructs a duallayer feature vector from low-frequency (50Hz) and high-frequency (1kHz) current components measured at both line ends. A composite fault indicator is calculated by adaptively weighting the Euclidean distances within each layer based on real-time signal-to-noise ratio and line attenuation. This approach leverages complementary fault information across frequency bands. Simulation results in PSCAD/EMTDC demonstrate that the proposed method significantly outperforms traditional differential protection in sensitivity and reliability, effectively mitigating maloperation risks under highimpedance faults and weak-infeed conditions, offering a viable software upgrade path for existing infrastructures.
Power transmission lines, as the core of the power system, are prone to failures due to long-term exposure to outdoor environments. Traditional manual inspection and data analysis methods are unable to meet the requirements of intelligent operation and maintenance. This paper uses 1829 sets of transmission line fault data collected by drones, covering multiple features such as electrical and environmental aspects, and classifies them into five operational states and conducts variable correlation analysis. To address the problem of insufficient feature extraction and poor generalization of a single algorithm in fault identification, a TCN-Transformer algorithm that integrates local temporal feature extraction and global attention mechanism is proposed, and an end-to-end classification prediction framework is constructed. The data set is divided in a 7:3 ratio and core parameters are set. The results show that the model has an accuracy rate of 0.823 and an AUC of 0.972. It outperforms traditional machine learning and ensemble learning models such as decision trees, random forests, and XGBoost in multiple key indicators, effectively improving the accuracy and stability of transmission line fault identification in complex scenarios, and providing a feasible technical solution for intelligent fault discrimination in drone inspections.
Currently, insulation faults in the DC system of photovoltaic (PV) power stations are handled by a full shutdown strategy of inverters, and fault branch localization relies on manual inspection, resulting in low efficiency and poor accuracy, leading to prolonged unplanned outages and substantial power generation losses. This paper proposes an integrated solution combining high-precision insulation monitoring and intelligent fault line selection, which ensures the reliability of line selection criteria through improved measurement accuracy and achieves automatic fault isolation via optimized line selection strategies. The paper analyzes the mathematical essence of the ill-conditioned measurement equations of the traditional bridge method under severe single-pole grounding faults, establishes a dual-channel heteroscedastic noise model, and utilizes the inherent physical constraint that the sum of the positive and negative pole-to-ground voltages always equals the bus voltage to transform the ill-posed inverse problem into an equality-constrained optimal estimation problem, deriving an analytical solution in the sense of constrained least squares. A collaborative monitoring strategy of “balanced bridge monitoring first, unbalanced bridge precision measurement afterward” is proposed. An automatic fault line selection and isolation algorithm based on sequential branch switching is designed, which leverages the operational characteristic that PV systems allow short-term branch interruption, enabling automatic identification and isolation of faulty branches and automatic restoration of non-faulty branches without installing any leakage current sensors. Experimental results show that under severe fault conditions with a single-pole insulation resistance as low as 22 kΩ, the proposed method limits the error to within 5%; the proposed line selection strategy can complete identification and isolation of all faulty branches within at most two rounds of switching.
Power transmission lines are essential components of power systems, and their operating conditions directly affect the safety, stability, and reliability of power supply. In practical operation, transmission lines are susceptible to various fault conditions, such as short circuits, grounding faults, conductor breakage, and insulator abnormalities, due to lightning strikes, strong winds, pollution, equipment aging, and external disturbances. Conventional fault identification methods mainly depend on relay protection signals, manual inspection, and model-based analysis. Although these methods have been widely applied, they often suffer from limited adaptability, insufficient feature representation capability, and reduced identification accuracy under complex operating conditions. To improve the accuracy and intelligence level of transmission line fault diagnosis, this paper proposes a deep learning-based intelligent fault identification method for power transmission lines. The proposed method employs a deep neural network to learn fault characteristics directly from transmission line monitoring data, thereby reducing dependence on manual feature extraction. Through hierarchical feature representation and nonlinear mapping, the method can effectively distinguish different fault patterns and improve fault classification performance. In addition, preprocessing and optimization strategies are introduced to suppress noise interference and enhance model robustness under imbalanced and complex data conditions. Experimental results show that the proposed method achieves better identification accuracy, stability, and generalization performance than conventional methods. The proposed method can provide effective technical support for online fault monitoring, rapid fault diagnosis, and intelligent operation and maintenance of transmission lines.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
