Health monitor of bogie-bearing on the train can ensure constant operation of the rail transit system. Since the metro or other rail transit have high safety requirements, it is hard to acquire numerous fault samples. Besides, diagnosing train bogie-bearings under variable working conditions is challenging due to wheel-rail coupling, speed variation, and load fluctuation. An intelligent approach for bogie-bearing fault diagnosis is proposed to deal with the above problems. A third-order tensor model is established to be suitable for variable working conditions. Furthermore, a density-based affinity propagation tensor (DAP-Tensor) clustering algorithm is presented to identify different failures with unlabeled. Train bogie and public data sets were employed to simulate three probable conditions of train operation: high-frequency impact, speed variation, and load change. Compared with existing clustering methods in three cases, the proposed DAP-Tensor performs better in identifying bearing faults under variable working conditions. Moreover, The DAP-tensor has a comparable recognition rate to some deep learning methods, which unsupervised characteristics show it has potential for applications on rail transit trains.

When an existing photovoltaic (PV) system is upgraded to a residential PV/battery system, the single-phase PV inverter under both input conditions of battery and PV should be properly controlled to restrain the input current ripple and grid-current harmonics. To do this, equivalent circuits of PV array and Li-ion battery pack are first constructed and respectively analyzed. The analysis results show that the input current under the battery pack may contain serious ripple component due to the low internal impedance of the battery pack, which cannot suppress the ripple current caused by the inherent power coupling problem of the grid-connected single-phase inverter. Then, based on the small signal model of the boost dc-dc convertor, a novel active control method is proposed for mitigating the input current ripple, which adopts double-channel current feedbacks including an additional ripple current feedback channel and the normal one. To provide the feedback signal, a third-order general integrator is introduced to extract the current ripple. Besides, a proportional-resonant controller is used to restrain the grid-current harmonics. Finally, the control parameters are obtained using MATLAB toolbox, and the proposed control strategies are validated by the experimental results on a 5 kW prototype.

In this paper, a combined reactive power modulation and grid current distortion improvement approach is proposed for an H6 transformer-less full-bridge single-phase photovoltaic (PV) grid-connected inverter. H6 transformer-less inverters with traditional modulation and control strategies may not satisfy the requirement of reactive power compensation or may result in more severe zero-crossing current distortion. Therefore, contrary to the traditional modulation, a novel reactive power injection space vector pulse width modulation (SVPWM) is proposed, which consists of two operation stages-inverter modulation and reactive power modulation. The implementation of SVPWM for reactive power modulation using a digital signal processor is also investigated. Furthermore, to suppress the current zero-crossing distortion in the reactive power injection mode, a global sliding mode function based on the proportion-integration-resonance current controller is designed, and the control law of the global sliding mode is derived. Using the segment modulation and grid current distortion improvement approach, the function of reactive current injection is implemented in commercial PV inverters, and the total harmonic distortion of the grid current is decreased significantly by more than 5% in the low-power segment, under the operating conditions of a lagging or leading power factor of 0.95. The effectiveness and feasibility of the proposed approach are verified through simulation and experiment using a 5-kVA prototype.

Synchronous reference frame phase-locked loop (SRF-PLL) is widely used for grid voltage synchronization in single-phase grid-connected power converters. However, in the actual situation, dc offset component may be introduced in the input of the PLL, due to the transient fault of the power grid, sampling and measurement error, or A/D signal processing. A simple yet effective approach with additional all-pass filter based dc rejecter is presented for SRF-PLL, in this paper. Thereby, correct estimation and undesirable periodic ripple free can be achieved in SRF-PLL, when the input signal contains dc offset. The second order generalized integrator based PLL (SOGI-PLL) is first introduced, followed by the analysis on influence of the input dc offsets in SRF-PLL. The structure of enhanced-SOGI (ESOGI) with its analysis of dc offset rejection effects and performance have been then formulated in detail. Finally, experimental results are presented to demonstrate the effectiveness of the proposed ESOGI based PLL.

With the continuous development of rail transit fully automatic operation, the urgent need to improve train operation safety makes obstacle detection become the research focus. In this work, a flexible and efficient multiscale one-stage object detector FE-YOLO was proposed for image obstacle detection. The feature extraction network is composed of attention module, downsampling module, residual block, spatial pyramid pooling (SPP) module, and so on. A repeatable bidirectional cross-scale path aggregation module was designed as the core of the feature fusion network. The dataset RT2021 of rail transit obstacles was constructed based on the real scene. The mean of average precision (mAP), detection time, iteration time, parameters, and anti-interference ability were used to compare FE-YOLO with other classic object detectors. The results showed that FE-YOLO has the best comprehensive performance. The mAP can reach 92.57%, and the single-frame detection time on the onboard embedded device is up to 0.0989 s. The ablation experiment verified the effectiveness of each module of FE-YOLO and has the best generalization performance on the PASCAL VOC dataset. Finally, a rail transit obstacle detection system was developed, and multiple sensors were used to improve the detection accuracy. Experiments showed that the detection system can work normally in different environments.