Original Current Signal Research Articles

An intelligent fault diagnosis method for rolling bearing using motor stator current signals

Abstract In the diagnosis of rolling bearing faults, the Motor Current Signature Analysis (MCSA) method offers advantages such as low cost, simplicity, and convenience compared to using vibration signals, temperature information, and other diagnostic objects. However, owing to the interference of high-frequency noise, power frequency, and its harmonics in current signals, which can severely affect the accuracy of bearing fault diagnosis, it is extremely challenging to use the original current signals during bearing faults directly for diagnostic purposes. Therefore, this paper proposes an intelligent fault diagnosis method based on the feature reconstruction (FR) method and convolutional neural networks (CNN). This method can achieve high-precision fault diagnosis using single-phase stator current signals from motors as the diagnostic objects. First, the FR method effectively removes the impact of high-frequency noise, supply frequency, and its harmonics from the current signals, while also highlighting subtle fault feature signals to a certain extent. Second, a CNN suitable for learning the characteristics of the current signals was constructed. Through feature extraction, learning, and classification of the current signal samples processed by the FR method, a diagnostic method with a high classification accuracy was obtained. Visualization techniques were used to present the final diagnosis results intuitively. The experimental results demonstrated the highest diagnostic accuracy and average diagnostic accuracy of the proposed method in diagnosing rolling bearing fault types, with an average diagnostic accuracy of approximately 99% for actual faulty bearing samples.

Fault diagnosis of drone motors driven by current signal data with few samples

Multi rotor unmanned aerial vehicles (UAVs) are extensively utilized across various domains, and the motor constitutes a pivotal element in the UAV power system. The majority of UAV failures and crashes stem from motor malfunctions, underscoring the imperative need for comprehensive research on fault diagnosis in UAV motors to ensure the stable and reliable execution of flight tasks. This study focuses on quadrotor UAVs as the research subject and devises targeted fault simulation experiments based on the structural features and operational characteristics of the DC brushless motor used in quadrotor UAVs, specifically examining the stator, rotor, and bearings. To address challenges related to the UAV’s own loads, limited space for redundant parts, and the high cost and difficulty associated with installing sensors for traditional fault diagnostic signals such as vibration and temperature, this study opts to use current signals as a substitute. This approach resolves the issue of challenging data collection for UAVs and investigates a current signal based fault diagnosis method for UAV motors. Lastly, in response to the limited training samples available for fault data due to the UAV’s highly sensitive characteristics regarding the health status of its components and flight stability, traditional machine learning and deep learning methods encounter difficulties in identifying representative features with a small number of training samples, leading to the risk of overfitting and reduced model accuracy in fault diagnosis. To overcome this challenge, we propose a hybrid neural network fault diagnosis model that incorporates a width learning system and a convolutional neural network (CNN). The width learning system eliminates temporal characteristics from the original current signal, capturing more comprehensive and representative sample features in the width feature space. Subsequently, the CNN is employed for feature extraction and classification tasks. In empirical small sample fault diagnosis experiments using current signal data for UAV motors, our proposed model outperforms other models used for comparison.

Motor current signal analysis using hypergraph neural networks for fault diagnosis of electromechanical system

Graph based networks are becoming an emerging trend in the field of fault diagnosis because of their powerful ability to mine the interrelationships between nodes. However, the existing graph-based networks are limited to mining the association relationship between adjacent nodes, which cannot reflect the strong association relationship between multiple nodes and thus affect the graph data quality. To solve these problems, a time-shifting based hypergraph neural network (TS-HGNN) is proposed for the accurate classification of fault types in electromechanical coupled systems. First, the time shifting method is applied to pre-process the original current signal to remove the power-line interference. Then, a hypergraph structure applicable to current signal is established to form complex interrelationships and a hyperedge convolution operation is designed to obtain the interrelationships of higher-order data for representation learning. Finally, several datasets are designed to verify the superiority and robustness of TS-HGNN in current signal fault classification.

Fault diagnosis of planetary gear backlash based on motor current and Fisher criterion optimized sparse autoencoder

Planetary gear reducer is widely applied in various transmission equipment, and its performance highly affects the operation of a machine. The appearance of unreasonable backlash in planetary gear reducer may lead to undesirable vibration, which may accelerate the degradation of equipment and eventually cause premature failure. In traditional condition-based monitoring (CBM), sensors such as accelerometers have been utilized to detect the fault of planetary gear. However, the complexity and integration of planetary gear limit the installation and application of vibration sensors in practice. In this case, the current signal from motor, as a convenient real-time monitoring approach, is introduced into the CBM of the planetary gear. In this paper, a fault diagnosis method based on drive motor current signal analysis (MCSA) is presented to identify the backlash fault in a planetary gear reducer. In this method, frequency domain data of the original current signal is found and used to automatically extract fault features. A deep sparse autoencoder (DSAE) extracts the required features. In particular, the Fisher criterion is introduced to evaluate the sensitivity of these features. It then selects the most effective ones for improving diagnostic accuracy as well as diagnostic efficiency. Experimental test data shows that under different load conditions, this method outperforms other typical fault diagnosis methods and exhibits the best performance.

Series arc fault identification based on complete ensemble empirical mode decomposition with adaptive noise and convolutional neural network

The effective identification of series arc faults is of considerable significance for preventing fires in residential buildings. Series arc fault currents and load currents have a similar waveform, and the fault features and nonfault features are superimposed on the current signal. Fault features are deeply hidden, making it difficult to identify them. This work proposes a method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) preprocessing and a one-dimensional convolutional neural network (1DCNN). The CEEMDAN algorithm is used to decompose the collected current signals. Then, the intrinsic mode function (IMF) components with no representational significance are eliminated by calculating the Spearman correlation coefficient before inputting it into the 1DCNN. The experimental results showed that the accuracy of the method for the measured load is 99.3%. Compared with the method that directly uses original current signals as model inputs, the recognition accuracy of the algorithm was significantly improved. Therefore, the proposed algorithm can be used for series arc fault identification in residential building power distribution systems.

A dynamic time over threshold method with fast trigger for silicon photomultiplier readout

In the readout of silicon photomultipliers (SiPMs), the original current signals are usually divided into different channels — fast timing channels, slow energy channels, position encoding channels, facilitating the formation of different trigger types. This paper proposes a modified dynamic time over threshold (DTOT) method to obtain the energy information in the slow energy channel using the fast timing trigger of a 6 mm SiPM coupled to a 10 mm cubic GAGG scintillator. The dynamic threshold generator is triggered by a fast timing signal instead of the energy signal itself. The feasibility of this method was verified using a prototype circuit, wherein the integral nonlinearity was measured to be 1.5%. This method can easily introduce external triggers to the DTOT system, such as a coincident signal with a specific trigger logic, which can extend the application of the DTOT method.

An Analytical Fast Decaying DC Mitigation Method for Digital Relaying Applications

This paper proposes a novel fast and accurate method to mitigate the decaying DC (DDC) component from the fault current signal. At the first stage, it starts to identify the DDC parameters by integrating six sub-cycle samples of current signal considering the symmetry feature of odd harmonics. By receiving the next samples, the estimation of DDC parameters is updated and improved. This procedure continues until the integral of a full cycle current signal is calculated. To improve the accuracy of phasor estimation procedure, in each stage, by subtracting the DDC component from original current signal, the discrete Fourier transform method is applied to estimate the fundamental phasor. Performance of the proposed method for different DDC parameters and fault conditions is verified through various test studies, and its robustness against harmonics, noise, off-nominal frequency, and multiple DDCs conditions is also investigated. Besides, to demonstrate the reliability of proposed method in terms of dependability and security, it is implemented on the phasor estimation block of a distance relay. Three types of data including mathematical, simulated, and real power system signals are used to evaluate the performance of proposed method. Obtained results confirm the desirable performance of proposed method in various operating conditions.

Rotate Vector (RV) Reducer Fault Detection and Diagnosis System: Towards Component Level Prognostics and Health Management (PHM).

In prognostics and health management (PHM), the majority of fault detection and diagnosis is performed by adopting segregated methodology, where electrical faults are detected using motor current signature analysis (MCSA), while mechanical faults are detected using vibration, acoustic emission, or ferrography analysis. This leads to more complicated methods for overall fault detection and diagnosis. Additionally, the involvement of several types of data makes system management difficult, thus increasing computational cost in real-time. Aiming to resolve that, this work proposes the use of the embedded electrical current signals of the control unit (MCSA) as an approach to detect and diagnose mechanical faults. The proposed fault detection and diagnosis method use the discrete wavelet transform (DWT) to analyze the electric motor current signals in the time-frequency domain. The technique decomposes current signals into wavelets, and extracts distinguishing features to perform machine learning (ML) based classification. To achieve an acceptable level of classification accuracy for ML-based classifiers, this work extends to presenting a methodology to extract, select, and infuse several types of features from the decomposed wavelets of the original current signals, based on wavelet characteristics and statistical analysis. The mechanical faults under study are related to the rotate vector (RV) reducer mechanically coupled to electric motors of the industrial robot Hyundai Robot YS080 developed by Hyundai Robotics Co. The proposed approach was implemented in real-time and showed satisfying results in fault detection and diagnosis for the RV reducer, with a classification accuracy of 96.7%.

Research on Fault Line Selection Method Based on Convolutional Neural Networks

In order to improve the reliability and accuracy of Single-Phase-to-Ground Fault line selection for small current grounding systems, this paper proposes a method named Deep Convolutional Neural Networks with Wide First-layer Kernels (WDCNN). The proposed method uses the original current signal as input and uses a wide kernel in the first convolutional layer to extract features and suppress high-frequency noise. The following convolutional layers with small convolution kernels are used for multilayer nonlinear mapping. Batch normalization layers and dropout layers are carried out for improving the generalization ability of the model. The simulation results show that the recognition rate of the proposed WDCNN can reach 99%, and it is not affected by factors such as grounding impedance, fault location and fault time.

A NEW IDENTIFICATION METHOD OF THE TRANSFORMER INRUSH CURRENT BASED ON IMPROVED HILBERT-HUANG TRANSFORM ALGORITHM

As the main form of protection for transformers, differential protection requires correct operation because it directly influences the safe operation of transformers. Meanwhile, transformer inrush current is an important factor in the tendency of transformers to malfunction. To accurately and reliably discriminate inrush current from internal fault current, this study proposed a novel method based on the Hilbert–Huang transform (HHT) algorithm. We integrated the empirical mode decomposition (EMD) algorithm with the permutation entropy (PE) algorithm and put forward a new algorithm named modified ensemble EMD. First, the original current signal was added to an opposite white noise signal with several repetitions for the EMD. Next, PE was employed to detect the decomposition results, where the noise, intermittent, and high-frequency signals with relatively large entropies were removed from the original signal. The remainder of the signal was decomposed by the second EMD to obtain a group of approximately stationary intrinsic mode munction (IMF) signals. Finally, the first IMF signal was extracted for HHT to obtain an instantaneous frequency, and the transformer inrush current and internal fault current were accurately identified by the time intervals between two adjacent catastrophe points in the instantaneous frequency. Results demonstrated that the time intervals between two adjacent catastrophe points in the instantaneous frequency of the transformer inrush current were less than 14 ms and that of various internal faults (i.e., turn-to-turn, single-phase, phase-to-phase, and three-phase faults) were approximately 20 ms. Results indicate that the inrush current and internal fault current are correctly, reliably, and quickly identified by the time intervals between two adjacent catastrophe points in the instantaneous frequency. The time intervals of the catastrophe points in the instantaneous frequency of the inrush current under different closing initial phase angles (0°, 30°, …, 330°) are less than 12.5 ms, indicating that the method is unaffected by the dead angle of the inrush current, closing initial phase angle, and non-periodic components. The proposed method provides references for identifying inrush current with regard to transformer differential protection. Keywords: Hilbert–Huang transform, Inrush current, Internal fault current, Instantaneous frequency

A Load Identification Method Based on Active Deep Learning and Discrete Wavelet Transform

Non-Intrusive Load Monitoring (NILM) makes it possible for users and energy providers to track the fine-grained energy consumption information of residential and commercial buildings. The load identification methods in NILM usually require labeling many samples for training and evaluation, which is always expensive and time-consuming. In order to reduce the labeling cost, this paper proposed a load identification method based on Active Deep Learning (ADL). In this method, Discrete Wavelet Transform (DWT) was applied to extract high-dimensional appliance features from original current signals. Then a pool-based or stream-based active deep learning model was built to learn the features and select high-value samples that worthy of labeling. A mixed dataset based on three public datasets was formed to evaluate the proposed method and three sampling approaches of active learning. The results showed that the proposed method could significantly reduce labeling cost on large datasets, and the number of samples required is 33% lower than the state-of-the-art method when the F1 score is equal. Compared with pool-based sampling approaches, the stream-based approach's benefits are that the classifier improved and the query frequency decreased with continuous input of samples.

Diagnosis of Rotor Asymmetries Faults in Induction Machines Using the Rectified Stator Current

Fault diagnosis of induction motors through the analysis of the stator current is increasingly being used in maintenance systems, because it is non-invasive and has low requirements of hardware and software. Nevertheless, its industrial application faces some practical limitations. In particular, the detection of fault harmonics that are very close to the fundamental component is challenging, as in large induction motors working at very low slip, because the leakage of the fundamental can hide the fault components until the damage is severe. Several methods have been proposed to alleviate this problem, although all of them increase noticeably the complexity of the diagnostic system. In this paper, a novel method is proposed, based on the analysis of the rectified motor current. It is shown that its spectrum contains the same fault harmonics as the spectrum of the original current signal, but with a much lower frequency, and free from the fundamental component leakage. Besides, the proposed method is very easy to implement, either by software, using the absolute value of the current samples, or by hardware, using a simple rectifier. The proposed approach is presented theoretically and validated experimentally with the detection of a broken bars fault of a large induction motor.

Fault location estimation for series-compensated double-circuit transmission line using parameter optimized variational mode decomposition and weighted P-norm random vector functional link network

In this paper, both the proposed parameter optimized variational mode decomposition (POVMD) and weighted P-norm random vector functional link network (WPRVFLN) are integrated for fault detection, location estimation, and classification in a series capacitor compensated double circuit transmission line (SCCDCTL). The required number of decomposition is determined from the magnitude spectrum of full-cycle current signals from the point of fault inception. An entropy index is used to obtain the optimum value of data-fidelity factor for extracting the most suitable band-limited intrinsic mode functions (BLIMFs). The four efficacious features namely standard deviation of the magnitude, energy, Renyi entropy and crest factor are computed from the Hilbert transformed array of the BLIMFs to construct the feature vector. A diagonal matrix W is computed based on zero sequence current of the original current signals as a weighting factor to categorize the ground fault accurately. Numerous faults are generated with a wide variation of the system conditions in MATLAB/Simulink simulation environment. An efficient WPRVFLN computational intelligence technique is proposed to recognize the fault by taking the extracted feature vector with weight factor, and its performances are compared with the recently developed classifiers in the MATLAB interface. The lesser computational complexity, faster learning speed, superior fault location estimation accuracy, and short event detection time prove that the proposed POVMD–WPRVFLN method can be implemented in the real power system for online fault recognition. Finally, the feasibility of the proposed method is tested and validated by using the fast FPGA digital circuitry in a loop.

Recognition and Classification of Incipient Cable Failures Based on Variational Mode Decomposition and a Convolutional Neural Network

To avoid power supply hazards caused by cable failures, this paper presents an approach of incipient cable failure recognition and classification based on variational mode decomposition (VMD) and a convolutional neural network (CNN). By using VMD, the original current signal is decomposed into seven modes with different center frequencies. Then, 42 features are extracted for the seven modes and used to construct a feature vector as input of the CNN to classify incipient cable failure through deep learning. Compared with using the original signals directly as the CNN input, the proposed approach is more efficient and robust. Experiments on different classifiers, namely, the decision tree (DT), K-nearest neighbor (KNN), BP neural network (BP) and support vector machine (SVM), and show that the CNN outperforms the other classifiers in terms of accuracy.

Investigation of film formation properties during electrochemical oxidation of serotonin (5-HT) at polycrystalline boron doped diamond

The change in surface morphology of oxygen-terminated polycrystalline boron doped diamond (pBDD) during electrochemical oxidation of the neurotransmitter serotonin (5-HT), resulting in a corresponding deterioration of the current signal, is investigated for the first time using both high resolution ex situ and in situ microscopy under a range of different electrochemical conditions. In situ electrochemical-atomic force microscopy (EC-AFM) reveals the formation of a granular film over the surface, which grows faster at higher-doped regions of the electrode surface and increases in thickness with repetitive potential cycles. The film properties were investigated using both cyclic voltammetry, with a range of redox species varying in charge, and conducting-AFM. These studies reveal the film to be positively charged and electrically insulating. The extent to which the film forms during 5-HT oxidation could be significantly minimised using different electrochemical procedures, as verified by voltammetry and in situ EC-AFM. Finally, even after extensive film formation, the original current signal could be recovered simply by leaving the electrode at open circuit potential for a short period of time, highlighting the suitability of BDD electrodes for neurotransmitter detection.