Islanding detection is one of the conditions necessary for the safe operation of the microgrid. The detection technology should provide the ability to differentiate islanded operations from power grid disturbances effectively. Given that it is difficult to set the fault threshold using the passive detection method, and because the traditional active detection method affects the output power quality, a microgrid islanding detection method based on the Sliding Window Discrete Fourier Transform (SDFT)-Empirical Mode Decomposition (EMD) and Long Short-Term Memory (LSTM) network optimized by an attention mechanism is proposed. In this paper, the inverter output current and voltage at the point of common coupling (PCC) are transformed by the SDFT. The positive sequence, zero sequence, and negative sequence components of voltage and current harmonics are calculated and reconstructed by adopting the symmetrical component method (SCM). Meanwhile, the current and voltage are decomposed into a mono intrinsic mode function (IMF). The symmetric components of voltage, current, and IMFs are used as inputs to the deep learning algorithm. An LSTM with the features extracted to classify islanding and grid disturbance is proposed. By using the attention mechanism to set the weight values of the features of hidden states obtained by the LSTM network, the proportion of important features increases, which improves the classification effect. MATLAB/Simulink simulation results indicate that the proposed method can effectively classify the islanding state under different working conditions with an accuracy level of 98.4% and a loss value of 0.0725 with a maximal detection time of 66.94 ms. It can also reduce the non-detection zone (NDZ) and detection time and has a certain level of noise resistance. Meanwhile, the problem whereby the active method affects the microgrid power quality is avoided without disturbing the current or power of the microgrid.

The acoustic-based detection is regarded as an effective way to detect the internal defects of arc magnets. Variational mode decomposition (VMD) has a significant potential to provide a favorable acoustic signal analysis for such detection. However, the performance of VMD heavily depends on the proper parameter setting. The existing optimization methods for determining the optimal VMD parameter setting still expose shortcomings, including slow convergences, excessive iterations, and local optimum traps. Therefore, a parameter-optimized VMD method using the salp swarm algorithm (SSA) is proposed. In this method, the relationship between the VMD parameters and their decomposition performance is quantified as a fitness function, the minimum value of which indicates the optimal parameter setting. SSA is used to search for such a minimum value from the parameter space. With the optimized parameters, each signal can be decomposed accurately into a series of modes representing signal components. The center frequencies are extracted from the selected modes as feature data, and their identification is performed by random forest. The experimental results demonstrated that the detection accuracy is above 98%. The proposed method has superior performance in the VMD parameter optimization as well as the acoustic-based internal defect detection of arc magnets.