For the long-primary double-sided linear induction motor (LP-DSLIM) excited by the symmetrical current, there are double slip frequency ripple in the thrust due to the dynamic longitudinal end effect. The expression of thrust ripple fluctuation caused by the dynamic longitudinal end effect are deduced firstly. Secondly, the method of suppressing the injected harmonic current is adopted, the expression of each part of thrust after injection are deduced and its characteristics are analyzed, from which the feasibility of this method to reduce the thrust ripple is illustrated. Finally, finite element simulation and experiment results indicate that the thrust ripple is significantly reduced after the injection of the harmonic current.

In order to improve the thrust density and efficiency of linear induction motors (LIM) in large load transport situations, a cross-shaped linear induction motors (CSLIM) structure is proposed in this paper. Firstly, for analyzing the characteristics of CSLIM, the magnetic field and electromagnetic thrust expression of long primary double-sided linear induction motors (LP-DSLIM) when considering the saturation of iron core are deduced, based on which the effect of iron saturation on electromagnetic thrust is analyzed. Secondary, the equivalent circuit of CSLIM is proposed and the effect of coupling on the electromagnetic thrust is analyzed. The results show that, the maximum electromagnetic thrust is reduced and the corresponding slip frequency is increased when iron saturation and coupling. For improving the problem of iron core saturation and coupling of CSLIM, a structure with square-type iron and lap winding is further proposed, and the rationality of CSLIM structure is verified by comparing the finite element simulation results of flux density on core, eddy current on secondary and the electromagnetic thrust.

The application of deep neural networks to address automatic modulation recognition (AMR) challenges has gained increasing popularity. Despite the outstanding capability of deep learning in automatic feature extraction, predictions based on low-data regimes with imbalanced classes of modulation signals generally result in low accuracy due to an insufficient number of training examples, which hinders the wide adoption of deep learning methods in practical applications of AMR. The identification of the minority class of samples can be crucial, as they tend to be of higher value. However, in AMR tasks, there is a lack of attention and effective solutions to the problem of Imbalanced-class in a low-data regime. In this work, we present a practical automatic data augmentation method for radio signals, called SigAugment, which incorporates eight individual transformations and effectively improves the performance of AMR tasks without additional searches. It surpasses existing data augmentation methods and mainstream methods for solving low-data and imbalanced-class problems on multiple datasets. By simply embedding SigAugment into the training pipeline of an existing model, it can achieve state-of-the-art performance on benchmark datasets and dramatically improve the classification accuracy of minority classes in the low-data imbalanced-class regime. SigAugment can be trained for uniform use on different types of models and datasets and works right out of the box.

The permanent magnet linear synchronous motors have the advantages of high efficiency, high thrust density and high power factor, which is mostly attractive to the fields of electromagnetic launch. The electromagnetic field of long primary PMLSM with different structures are analyzed firstly by finite element method in this paper. Then, the thrust characteristics of the motors with different topologies are studied and compared, which fully demonstrates the advantages and feasibility of the proposed PMLSM.

In this paper, the electromagnetic and mechanical characteristics of the improved homopolar linear synchronous motor (H-LSM) have been studied. Firstly, the influence of the ratio of the width and pitch of the rotor teeth is investigated on the magnetic field distribution and vibration traits. By establishing a simplified three-dimensional magnetic circuit model, the distribution of air gap flux density is attained. Secondly, the time and space distributions of the air gap flux density is decomposed, and the ratio of tooth width to tooth pitch with larger fundamental component is selected as the key index to design the H-LSM. Thirdly, through the modal analysis of the primary core, detailed investigation on the harmonics response is carried out, and the resonance frequency domain in the working process is clarified. Finally, 1-10th order frequency spectrum on the H-LSAM resonance is decomposed, which can provide a judgment for the fault diagnosis.