Aug 2026· Scientific Reports· Vol 16· 0 citations· 70 references
Medicine
TL;DR
The findings demonstrate that the proposed intelligent protection framework provides accurate and reliable fault detection, classification, and location through optimized DWT-based feature extraction and SVM-based decision making under the investigated simulation scenarios.
Abstract
This paper presents an intelligent protection framework for fault detection, classification, and location in power distribution networks by combining Discrete Wavelet Transform (DWT)-based feature extraction, Support Vector Machine (SVM)-based decision making, and Internet of Things (IoT)-enabled cloud monitoring. An IEEE 16-bus distribution system is modeled in MATLAB/Simulink, where transient current signals are processed using DWT to extract discriminative time–frequency features. A comparative evaluation of different mother wavelets and decomposition levels is performed to identify the most effective feature extraction configuration in terms of accuracy and computational efficiency. The extracted features are processed locally by SVM-based models for fault detection, classification, and location, while selected fault-related features are simultaneously transmitted to the ThingSpeak cloud platform for cloud-assisted monitoring and remote accessibility. The proposed framework is evaluated under a wide range of operating conditions, including different fault types, overload events, load switching, capacitor switching, and scenarios with integrated photovoltaic and wind generation. The results demonstrate 100% fault classification accuracy and fault-location accuracies ranging from 97.95% to 99.88% within the investigated simulation scenarios. Furthermore, the proposed approach effectively distinguishes faults from non-fault disturbances, thereby reducing the likelihood of false fault indications. The findings demonstrate that the proposed intelligent protection framework provides accurate and reliable fault detection, classification, and location through optimized DWT-based feature extraction and SVM-based decision making under the investigated simulation scenarios. Nevertheless, additional validation using noisy measurements and hardware-based experimental platforms is required to further assess the robustness and practical applicability of the proposed protection methodology under real operating conditions.
To ensure reliable operation of the distribution system, feeder faults, transformer stress and abnormal operating conditions need to be detected in a timely fashion. This paper proposes an edge-intelligent monitoring and real-time communication system for a three-phase 11 kV/415/230 V power distribution system, which is based on IoT. The proposed system combines the sensing of the voltage, current, active power and the LT-side and transformer-side temperature sensors with local decision logic based on thresholds and remote communication through a dedicated mobile phone and control-room interface. A mathematical model was derived to explain the following: phase voltage, phase current, power utilization, voltage deviation, overcurrent, imbalance and temperature state. The system was tested using scenarios in MATLAB/Simulink representing various operating conditions such as overloading, undervoltage, overvoltage, overheating, line-fault and normal operating conditions. The simulation results showed that the monitoring logic maintained a normal state correctly without triggering false alarms, generated alarm at warning stage when progressive overload occurred, triggered alarm at undervoltage and over-voltage at 8.0 s, and generated the alarm in the line-fault state when the voltage collapsed in combination with current surge, and triggered the alarm at warning state during the overheating simulation. Under the line fault condition, the maximum current was increased to 203.43A as compared with 94.48A under normal condition, and the overheating condition reached 98.34 °C. The results show that the integration of electrical and thermal monitoring with edge-level decision logic leads to a better situation awareness and quicker operator response. The proposed framework provides a scalable smart distribution monitoring platform and a future enhancement path towards implementation with hardware, event logging, fault-location techniques and machine-learning based fault classification.
Unknown authors· Journal of Global Social Tra...· 0 citations
With the increasing integration of distributed photovoltaics, energy storage systems, electric vehicle chargers, and intelligent terminals, accurate identification of heterogeneous edge-end devices in distribution networks has become challenging due to their diverse operating characteristics and similar current signatures. This paper proposes an identification method based on current time–frequency features and Random Forest. Equivalent grid-connected current models are developed for five types of edge-end devices, considering different capacity levels, operating states, ripple characteristics, and transient behaviors. A 13-dimensional feature set is extracted from time-domain, frequency-domain, and time–frequency characteristics, covering 11 device subclasses. Feature analysis is conducted to evaluate the separability of the extracted features, and Random Forest is employed for multi-class device identification. The results show that the proposed method achieves an overall accuracy above 98% on independent test samples and 96.91% in the IEEE 33-bus validation case, demonstrating its effectiveness for distribution-network edge-end device identification.
Hui Fan, Jie Zhao, Zhao Zhao et al.· Electronics· 0 citations
The rapid expansion of offshore wind farms has introduced significant challenges to operation and maintenance (O&M), particularly under harsh marine environments where reliable electromagnetic information transmission and constrained wireless communication resources directly affect intelligent monitoring performance. Traditional Supervisory Control and Data Acquisition (SCADA) systems relying on cloud-centric architectures often encounter excessive latency and bandwidth bottlenecks when transmitting high-frequency vibration signals, limiting real-time fault diagnosis. To address these issues, this study proposes an Internet of Things (IoT)-based intelligent monitoring and fault detection method built upon an Edge-Cloud collaborative architecture. A lightweight Adaptive One-Dimensional Convolutional Neural Network (A-1D-CNN) is developed for deployment on edge gateway devices, enabling direct extraction of fault characteristics from raw vibration signals without manual feature engineering. Combined with an “ Edge-Training, Cloud-Update” strategy, the proposed framework continuously optimizes diagnostic performance while substantially reducing communication overhead across wireless sensing and electromagnetic transmission infrastructures. Experimental evaluation on a standard bearing fault dataset demonstrates that the proposed method achieves a fault diagnosis accuracy of 99.25% with a compact model size of only 0.45 MB, providing an effective balance between diagnostic precision and deployment efficiency. The results indicate that the proposed framework offers a practical solution for real-time intelligent monitoring in bandwidth-limited offshore environments and provides technical support for reliable electromagnetic-enabled sensing networks and distributed fault diagnosis in next-generation offshore energy systems.
Y. Ouyang, W. Liang· Advanced Electromagnetics· 0 citations
A new hybrid approach to fault detection in power systems based on S-Transform feature extraction and SVM-based intelligent classification is proposed that can clearly distinguish between power swings and actual faults, including symmetrical three-phase faults that have characteristics similar to power swings.
P. Sharma, M. Silas, P. Roy et al.· African Journal Of Applied R...· 0 citations
Power quality (PQ) monitoring plays a crucial role in the operating conditions of electrical distribution networks and ensuring compliance with power quality standards. The increasing need for pervasive and distributed monitoring motivates the development of low-cost measurement instrumentation capable of operating directly at the network edge. With these aims, this paper proposes the design of a compact and distributed instrumentation device suitable for deployment in low-voltage networks and resource-constrained measurement scenarios implemented on an ESP32 microcontroller platform. The power signal is acquired through a cost-effective sensing front-end and processed using a multisinusoidal decomposition technique, which provides the feature extraction for the detection and classification of PQ events such as harmonics, voltage sags and swells, and transients. For the classification, the extracted features are used as inputs to a machine learning algorithm. In order to select the most suitable one, a further contribution of this paper is to test several supervised machine learning algorithms, which are systematically compared in terms of classification accuracy, robustness to measurement noise, and computational complexity. Particular emphasis is placed on algorithm suitability for real-time execution on embedded measurement hardware with limited memory and processing resources, such as the ESP32. Experimental results are obtained using emulated PQ signals. The results confirm that the integration of multisinusoidal signal analysis with lightweight machine learning techniques represents an effective solution for cost-effective and scalable PQ instrumentation.
D. Carní, Francesco Lamonaca· 2026 IEEE International Work...· 0 citations
This study develops a low-complexity data-driven method for real-time fault classification, and localization on a 20 kV radial feeder, while accounting for the computational and memory constraints of embedded hardware. The proposed approach is motivated by the changes introduced by distributed generation (DG), which alters fault-current behavior and reduces the reliability of conventional impedance-based protection. These limitations are mainly caused by reverse power flows and the additional infeed contribution from DG units during fault conditions. The proposed hybrid architecture processes time- and frequency-domain features, including symmetrical components and wavelet energy, through a dual-stage inference chain. The first stage uses a Random Forest classifier for fault-type identification, while the second stage uses a Multi-Layer Perceptron (MLP) distance estimator for fault localization. Evaluation on 23,778 MATLAB/Simulink fault scenarios, was performed using a grouped scenario-level holdout split to reduce overlap between training and test cases. The proposed approach reduces the localization error compared with the conventional impedance method, which produced errors exceeding 14 km under DG operation. The Random Forest classifier achieved 99.44% fault-type accuracy, and the MLP estimator reached a Mean Absolute Error (MAE) of 93.8 m, with localization accuracy of 99.51% within 1 km. Analytical resource profiling estimates less than 300 KB Flash and under 50 µs execution time, assuming a Cortex-M4F/M7-class target, pending hardware-in-the-loop validation.
M. Lemkharbech, S. Sarih, Z. Boulghasoul et al.· International Conference on...· 0 citations