Series Arc Fault Detection Based on Dual Filtering Feature Selection and Improved Hierarchical Clustering Sensitive Component Selection

Zou, Guofeng; Fu, Guixia; Han, Bangzheng; Wang, Wei; Liu, Chuanyang

doi:10.1109/jsen.2023.3242067

Cited by 15 publications

(14 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Embedded methods 11 – 13 such as LASSO and Decision Trees, incorporate feature selection within the algorithm itself for different scenarios. These methods usually provide a good trade-off between performance and speed but are limited by the biases of the algorithms they are integrated with for different use cases 14 – 16 .…”

Section: In-depth Review Of Existing Machine Learning Models Used For...mentioning

confidence: 99%

Feature importance feedback with Deep Q process in ensemble-based metaheuristic feature selection algorithms

Potharlanka,

2024

Sci Rep

View full text Add to dashboard Cite

Feature selection is an indispensable aspect of modern machine learning, especially for high-dimensional datasets where overfitting and computational inefficiencies are common concerns. Traditional methods often employ either filter, wrapper, or embedded approaches, which have limitations in terms of robustness, computational load, or capability to capture complex interactions among features. Despite the utility of metaheuristic algorithms like Particle Swarm Optimization (PSO), Firefly Algorithm (FA), and Whale Optimization (WOA) in feature selection, there still exists a gap in efficiently incorporating feature importance feedback into these processes. This paper presents a novel approach that integrates the strengths of PSO, FA, and WOA algorithms into an ensemble model and further enhances its performance by incorporating a Deep Q-Learning framework for relevance feedbacks. The Deep Q-Learning module intelligently updates feature importance based on model performance, thereby fine-tuning the selection process iteratively. Our ensemble model demonstrates substantial gains in effectiveness over traditional and individual metaheuristic approaches. Specifically, the proposed model achieved a 9.5% higher precision, an 8.5% higher accuracy, an 8.3% higher recall, a 4.9% higher AUC, and a 5.9% higher specificity across multiple software bug prediction datasets and samples. By resolving some of the key issues in existing feature selection methods and achieving superior performance metrics, this work paves the way for more robust and efficient machine learning models in various applications, from healthcare to natural language processing scenarios. This research provides an innovative framework for feature selection that promises not only superior performance but also offers a flexible architecture that can be adapted for a variety of machine learning challenges.

show abstract

Section: In-depth Review Of Existing Machine Learning Models Used For...mentioning

confidence: 99%

Feature importance feedback with Deep Q process in ensemble-based metaheuristic feature selection algorithms

Potharlanka,

2024

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Many efforts have been done to measure the current variation with time in the range from DC to kHz to detect the slow decrease of the arc current by means of a current shunt [20], a tunneling effect magnetic sensor [22], a lowfrequency current transformer [24], Hall effect probes [25] or commercial current transformers that can reach the range of MHz, [21], [26]. However, these devices do not comply with the constraints of sensitivity, or weight and size imposed by the aeronautic sector.…”

Section: Introductionmentioning

confidence: 99%

“…However, these devices do not comply with the constraints of sensitivity, or weight and size imposed by the aeronautic sector. Other types of sensors require a power supply [22], [24], [25], [27], [28] which could be a problem for this industrial application and should be avoided even when they have good sensitivity. Other authors, [29], use a Rogowski coil but their solution is hardly valid for aircraft systems because requires a shunt capacitor with the load on each monitored circuit and the sensitivity is well below than the proposed sensor in this paper.…”

Section: Introductionmentioning

confidence: 99%

“…Once the type of sensor has been selected, the analysis of the arc current can be done with different signal processing techniques [20], [22], [24], [30], [27], but to obtain the highest possible accuracy in the detection of the arc, it is of paramount importance to start with a sensor with high sensitivity in the appropriate bandwidth. This work shows that the signals to measure will be current pulses in the range of a few mA propagating along a transmission line.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

High-Frequency Current Transformer Design to Detect Serial Arcs in More Electric Aircraft

Robles,

Martínez-Tarifa,

Barroso-de-María

et al. 2024

IEEE Trans. Transp. Electrific.

View full text Add to dashboard Cite

Serial arcs in DC circuits is a well-known hazard in the field of photovoltaic systems and it is specially sensitive and critical in aerospace and aeronautic equipment. Unlike parallel arcs, the detection of this particular ionisation phenomenon cannot be done with the current available protection devices so the industry is interested in finding a solution. The first step is to choose a sensor capable of reading the physical consequences of an active arc and the appearance of high frequency current pulses derived from the arc has been proved to be a reliable method of detection. Inductive sensors are allegedly one of the most promising detectors in terms of sensitivity, cost and adaptability to the circuit wiring. For aircraft applications, size and weight are additional constraints which limit the use of certain heavy inductive sensors so an air-cored inductive sensor was proposed by the authors in a previous work. This paper is a step forward in the design of a solution using small and weightless toroidal ferrites to detect serial arcs. The geometric parameters, number of turns, cut-off frequency, bandwidth and sensitivity connected to a 50 Ω impedance of the proposed sensor are parameterised using a model based on a transformer equivalent circuit. The model is successfully validated with experimental measurements and a candidate is chosen according to the constraints imposed by the application in airborne electric systems. Finally, a complete set of measurements of arcs are conducted to test its performance compared with other two sensors. The aim is to quantify the sensitivity of the high frequency current pulses created by the arc activity when compared to the background noise.

show abstract

“…Previous arc detection research has developed from the time domain characteristics of arcs to high-order cumulants [26] and singular values [27], which reflects the increasing emphasis on arc feature selection. Arc feature extraction needs to relies on continuous trial and error or adopting an optimization method independent of the arc recognition model [16], [28]. Because the deep neural network model is a "black box", its internal behavior characteristics are unknown, which may lead to a mismatch between the arc features extracted by this method and those extracted by the machine learning model.…”

Section: Introductionmentioning

confidence: 99%

Explainability Approach-Based Series Arc Fault Detection Method for Photovoltaic Systems

Wang,

Zhou,

Paul

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Arc fault detection devices are mandatory worldwide for mitigating DC series arc faults in photovoltaic systems. However, they are prone to nuisance tripping. Artificial intelligence-based approaches can be a solution, but they are "black boxes" and challenging to modify. This paper proposes an explainability and attention-based method to investigate the intensive details of such an algorithm. The contributions of an arc feature to the proposed model can be visualized by the proposed interpretable methodology so that insensitive arc features can be removed to reduce the quantity of input data. Additionally, the structure of the proposed model can be optimized by cutting the redundant layers. Thus, an accuracy of 99.63% is achieved with only 48.48% of the parameters compared to the original model. Finally, the optimized model is implemented by a Cortex M7-based microprocessor with a runtime of only 7.8 ms, making it ready for industrial application.

show abstract

Series Arc Fault Detection Based on Dual Filtering Feature Selection and Improved Hierarchical Clustering Sensitive Component Selection

Cited by 15 publications

References 10 publications

Feature importance feedback with Deep Q process in ensemble-based metaheuristic feature selection algorithms

Feature importance feedback with Deep Q process in ensemble-based metaheuristic feature selection algorithms

High-Frequency Current Transformer Design to Detect Serial Arcs in More Electric Aircraft

Explainability Approach-Based Series Arc Fault Detection Method for Photovoltaic Systems

Contact Info

Product

Resources

About