Search

Arc detection is crucial for ensuring the safe operation of power systems, where timely and accurate detection of arcs can prevent potential hazards such as fires, equipment damage, or system failures. Traditional arc detection methods, while functional, often suffer from low detection accuracy and high computational complexity, especially in complex operational environments. This limitation is particularly problematic in real-time monitoring and the efficient operation of power systems. In order to solve these problems, this paper proposes a method of arc detection based on deep learning, called arc multi-scene detection (ArcMSD), which leverages deep learning techniques to address these challenges. The primary distinction of this method lies in its enhancement of the Inception V3 model. This paper has redesigned the original Inception module by incorporating a guided anchor mechanism, an attention mechanism, and upsampling techniques to optimize detection performance. The improved Inception V3 network uses an attention mechanism to allow the model to focus on arc features in complex backgrounds, which can also prevent the model from overfitting. It performs upsampling and fusion with low-level features in the model. The fused features have better arc discrimination capabilities than the original input features, which better improves the accuracy of the model. In order to adapt to arcs with large size differences and improve detection efficiency, the guided anchor is selected to adjust the anchor generation algorithm. In terms of dataset, continuous frame images are intercepted from the video of Integrated Supervision and Control System (ISCS), and image preprocessing operations are performed to improve the model's detection accuracy of pantograph arcs. Experimental results show that the mean Average Precision (mAP) of the deep learning model proposed in this article is 95.4%, which is far better than other models, thus verify the method's efficacy. [ABSTRACT FROM AUTHOR]

Distributed photovoltaic systems have encountered unprecedented opportunities for development given their environmentally friendly nature and flexible power generation characteristics. However, numerous connecting lines and taps within the distributed photovoltaic system can be subject to insulation issues, which will consequently cause direct current (DC) arc faults and severe electrical fire hazards. Moreover, the power semiconductor devices in the photovoltaic inverter can introduce common-mode noises to the DC current, resulting in unwanted tripping of the DC arc fault detector. The study proposes an arc fault detection method utilizing a deep residual shrinkage network (DRSN) to address this issue, thereby precisely detecting DC arc faults. A test platform for series arc faults in photovoltaic systems is built. The arc current data are collected for characteristic analysis in time and frequency domains to determine which bandwidth is preferred for the algorithm. The model's depth is increased by introducing residual connections, enhancing its feature extraction, and improving noise reduction capabilities. The residual shrinkage network has been enhanced to prevent a computation increase from increased network depth. Introducing a convolutional auto-encoder for data dimension reduction has decreased neural network parameters, thereby improving training speed. A prototype for detecting photovoltaic DC arc faults was constructed using Raspberry Pi 4B, validating the practical application value of the proposed method. Experimental results demonstrate that the prototype for detecting photovoltaic DC arc faults successfully fulfills the real-time detection standard of the conduction test. [ABSTRACT FROM AUTHOR]

Various studies on arc detection methods are described. Series AC arc is detected based on the characteristics extracted from arc voltage, frequency, and time domain of the current. Methods of arc detection using artificial intelligence have been studied previously. In the present study, the performance of multiple methods is analyzed by comparing different input parameters and artificial neural networks. In addition to the input parameters presented in the literature, the performance is compared and analyzed using the following parameters: zero-crossing period, frequency average, instantaneous frequency, entropy, combination of fast Fourier transform (FFT) and maximum slip difference, and combination of FFT and frequency average. These parameters and different neural networks are studied in the bounded and unbounded case, and the performance is compared. For different combinations of neural networks and input parameters, another research question is to identify the input parameters to be used if the number of training data is limited. Moreover, this study investigates the change in detection rate depending on the number of training samples. As a result, the minimum dataset size required to obtain the final detection rate is identified. [ABSTRACT FROM AUTHOR]

pantograph-catenary system is one of the critical components used in electrical trains. It ensures the transmission of the electrical energy to the train taken from the substation that is required for electrical trains. The condition monitoring and early diagnosis for pantograph-catenary systems are very important in terms of rail transport disruption. In this study, a new method is proposed for arc detection in the pantograph-catenary system based signal processing and S-transform. Arc detection and condition monitoring were achieved by using current signals received from a real pantograph-catenary system. Firstly, model based current data for pantograph-catenary system is obtained from Mayr arc model. The method with S-transform is developed by using this current data. Noises on the current signal are eliminated by applying a low pass filter to the current signal. The peak values of the noiseless signals are determined by taking absolute values of these signals in a certain frequency range. After the data of the peak points has been normalized, a new signal will be obtained by combining these points via a linear interpolation method. The frequency-time analysis was realized by applying S-transform on the signal obtained from peak values. Feature extraction that obtained by S-matrix was used in the fuzzy system. The current signal is detected the contdition as healthy or faulty by using the outputs of the fuzzy system. Furthermore the real-time processing of the proposed method is examined by applying to the current signal received from a locomotive. [ABSTRACT FROM AUTHOR]