Numerous studies show that power semiconductors are significant contributors to the overall failure rate of PWM voltage source rectifier (VSR) systems. Aimed at this problem, this paper presents an improved diagnosis system for opencircuit faults of a three-phase PWM VSR based on AC current distortion characteristics, which consists of a diagnosis module, a register module and a decision module. The diagnostic module employs a hysteresis comparator to judge the change trend of the currents, and thereby diagnoses the single OC faults. The register module is employed to achieve multi-switch OC fault diagnosis, and the decision module is used to output the final result. The proposed method extends the previous work to multi-switch OC fault conditions as well as improves the diagnosis accuracy. The experimental results show that the proposed method can correctly achieve fault diagnosis, including 6 kinds of single switch faults and 15 kinds of different double switches faults. Additionally, the proposed method also shows superior antiinterference performance and high robustness on abrupt load transient conditions, different power factor conditions, unbalanced and/or distorted grid voltage conditions and different AC filter structure conditions. Furthermore, the average diagnostic time of this method is only 3.68 ms.

Ultrasonic sensor detection has been widely used in partial discharge (PD) location. However, this PD localization method based on an ultrasonic sensor is computationally expensive, inaccurate, and sensitive to the time differences. A localization method for PD based on time difference of arrival (TDOA) and truncated singular value decomposition (TSVD) regularization is proposed in this paper, which can effectively ensure the accuracy of PD location in a transformer. First, the PD signal is received by the ultrasonic sensor, and an oscilloscope is used to receive the ultrasonic signal from the operational amplifier. Second, a spherical linear transformation method is used to transform the general nonlinear localization equations. Third, a TSVD regularization method is used to solve the linear equations. Then, to eliminate the influence of coordinate selection on the matrix b, the matrix b is centred. By dealing with the balance of matrix A, the ill-conditioned degree of linear equations is weakened. Finally, the optimal solution serves as a good approximation of the original problem. The proposed method is applied to the PD location in a transformer, and the localization results are compared with direct Newton iteration and direct TSVD methods. A comparison of the experimental results shows that the method in this paper can reduce the ill-conditioned degree of a linear equation, and effectively weaken the influence of time difference error on the solution. The localization results of this method are better than the other two methods.

This paper proposes an external breaking vibration identification method of transmission line tower based on a radio frequency identification (RFID) sensor and deep learning. The RFID sensor is designed to obtain the vibration signal of the transmission line tower. In order to achieve long-time monitoring and longer working distance, the proposed RFID sensor tag employs a photovoltaic cell combined with a super capacitor as the power management module. convolution neural network (CNN) is adopted to extract the characteristics of vibration signals and relevance vector machine (RVM) is then employed to achieve vibration pattern identification. Furthermore, the Softmax classifier and gradient descent method are used to adjust the weights and thresholds of CNN, so as to obtain a high-precision identification structure. The experiment results show that the minimum sensitivity of the proposed solar-powered RFID sensor tag is −29 dBm and the discharge duration of the super capacitor is 63.35 h when the query frequencies are 5/min. The optimum batch size of CNN is 5, and the optimum number of convolution cores in the first layer and the second layer are 2 and 4, respectively. The maximum number of iterations is 10 times. The vibration identification accuracy of the proposed method is over 99% under three different conditions.

Lithium-ion batteries are the primary power source for new energy vehicles, making accurate estimation of their state of health (SOH) essential for ensuring the safe operation of battery systems. This paper proposes a Capacity–Power Analysis (CPA) method that incorporates temperature features to enhance feature extraction across a broader range. Additionally, we introduce an SOH estimation method for lithium batteries based on a Convolutional Neural Network (CNN) and a Kolmogorov–Arnold Network (KAN). By extracting the capacity–power curve and average temperature features during constant-current and constant-voltage charging, the CNN-KAN model establishes a nonlinear mapping relationship between the extracted features and SOH, enabling high-precision SOH estimation for lithium-ion batteries. Four 18650 batteries were tested under various charging and discharging conditions in a laboratory setting. The coefficient of determination (R²) exceeded 96.4%, the root mean square error (RMSE) was below 0.86%, and the mean absolute error (MAE) was under 0.7%, confirming that the proposed method demonstrates excellent estimation performance.

The rapid development of the Internet of Things (IoTs) with newly proposed wireless communication technology has compactly generated connections between humans and the physical world. Beginning with research on wireless sensor networks, environmental monitoring research has currently developed into wider applications aiming to detect the world with multitudinous IoT technologies. Recently, low power wide area technologies have provided such benefits as longer communication range, larger network capacity, and low unit-price, providing more choices for managers to set up more detailed monitoring. However, the operational lives of battery-powered end-devices still hinder long-term monitoring. In this paper, we propose a decentralized framework for energy-efficient environmental monitoring considering both the operation cost and current computational capabilities of the end-devices. The core function is a low-complexity scheduling approach that can balance the monitoring performance and energy consumption for different environmental states. Driven by the prediction accuracy of the surrogate model, monitoring nodes are selected. Meanwhile, the energy consumption of the other nodes is saved. Simulation results demonstrate that the proposed monitoring system has high energy efficiency with acceptable performance. The battery life of the whole system can be prolonged by up to 136.22%.

This paper presents a novel scheme for short-circuit diagnosis with high diagnosis accuracy, robust fault tolerance, and shorter time delays. The scheme is operated on a fog computing-based framework with the current power system infrastructure. Wide-area backup protection information is processed into targeted knowledge with smaller package sizes and transmitted among the relative fog nodes to establish incidence matrices in each node. Each fog node can calculate its diagnostic value using the matrices for the fault location. The resources of computing and communication are both economized during the process. The performance is validated with the IEEE 118 bus system. The results demonstrate that the scheme can maintain 100% accuracy with any three-error scenarios and provide a shorter latency with high network utilization.