Series arc fault is a common phenomenon in the power system, it will directly affect the working reliability, but there is no mature method to detect it due to its concealment and chaos. Common detection methods that build on the arc fault eigenvectors obtained by manual analysis are subjective and incomprehensive. A series arc fault diagnosis and line selection method based on recurrent neural network (RNN) for a multi-load system was proposed in this paper. Firstly, a series arc fault experiment under a multi-load system was carried out, the training set and test set were built by using the data obtained from the experiment. Then, the RNN model was built, trained, and tested through the training set and test set. Finally, the fast-continuous detection method and the probability-based classification result correction method were proposed, and the detection speed and accuracy were improved much further. The results show that the proposed method is effective for diagnosing series arc fault and line selection under a multi-load system, without analysis of arc fault characteristics.

Certain faults occurring in a three-phase asynchronous motor generate a mechanical vibration of a specific frequency. Moreover, poor electrical contact points cause series arc faults (SAFs) under mechanical vibration. To detect SAFs under different vibration frequencies and amplitude conditions, we propose an SAF detection method based on an improved recurrence quantification analysis (IRQA) and a back-propagation neural network. First, we designed an arc fault generator and then completed an SAF experiment of the motor with inverter circuits under different vibration conditions. Next, we extracted the high-dimensional characteristics of the fault phase current by using an IRQA. In the IRQA, we performed phase space reconstruction processing on the fault phase current signal and then converted the processed signal into thresholded and unthresholded recurrence plots. The nine SAF features in the plots were extracted. Training and testing sets were constructed based on these SAF features. Finally, by training and testing of the BP neural network, we established an SAF detection model. The test results show that the SAF detection model distinguished between the normal and SAF current signal with 100% accuracy, and that it judged the mechanical vibration frequency caused by some motor faults with 98.17% accuracy.

Under the conditions of a mechanical fault in a motor, mechanical vibration of a specific frequency can be generated. The electrical contact points directly connected to the motor can vibrate at the same frequency. The electrical contact points with poor contact can easily produce a series arc fault under vibration conditions, which affects the reliability of the power supply. In order to detect the series arc fault under different vibration conditions, the arc fault generator is connected between the back end of the frequency converter and the motor. An arc fault experiment under different vibration conditions was carried out and the fault phase current and arc voltage signals were collected. In this paper, the noise-assisted multivariate empirical mode decomposition and the correlation coefficient between each intrinsic mode function are used to select the fault feature signals. Then, the reconstructed signal is input into the series arc fault model combining a multi-scale convolutional neural network and a bidirectional long short-term memory network for training. The research results show that the series arc fault detection method proposed in this paper can effectively detect the series arc fault and can preliminarily identify the type of motor fault causing the mechanical vibration of the motor; the model has good noise immunity and generalization.

This paper proposes a novel heading control law for an unmanned surface vehicle subject to external disturbance. A robust heading autopilot is developed with trajectory linearization control algorithm and disturbance observer. In the past research, trajectory linearization control is often used in the field of motion control of aircraft. However, in this paper, it will be applied to the heading control strategy of the unmanned surface vehicle. The advantage of trajectory linearization control technology is its simple structure and the robustness of the controlled system can be guaranteed by dealing with external disturbance through a feedback term. Nevertheless, the ability of trajectory linearization control to deal with external disturbance is ultimately limited, so in order to further improve the robustness of heading control strategy, a disturbance observer is designed to compensate external disturbance. Finally, the numerical simulation results of keeping keeping and tracking are given to verify the feasibility and correctness of the proposed scheme.

Structural optimization plays a crucial role in enhancing the performance of magnetic actuators. Traditional design approaches, such as parametric scanning, are limited by their reliance on empirical geometries. Meanwhile, widely used topology optimization techniques—including the Solid Isotropic Material with Penalization (SIMP) method and level-set methods—often encounter difficulties such as a large number of design variables, high computational cost, and unclear structural boundaries. To overcome these limitations, this paper proposes a novel gradient-free topology optimization method based on superellipses for designing magnetic actuator yokes. The proposed approach offers three key benefits: (1) It requires very few design variables, with each superellipse described by only seven parameters, thereby reducing the dimensionality of the design space and simplifying the optimization problem. (2) It yields clear and smooth structural boundaries without the need for post-processing. (3) It operates without gradient information, employing stochastic algorithms such as genetic algorithms that rely solely on objective function evaluations. A case study on yoke optimization demonstrates that our method achieves magnetic force output comparable to or better than the SIMP method, but with significantly fewer variables and a simpler implementation. This work provides an efficient and new tool for the conceptual design of magnetic actuators and related electromagnetic devices.