In intelligent fault diagnosis, transfer learning can reduce the requirement of sufficient labeled data and the same data distribution. However, for the diagnosis of a new machine, there are still some limitations, such as low accuracy or the demand for some labeled data with fault information in the new machine. In this article, we propose a stack autoencoder transfer learning algorithm based on the class separation and domain fusion (SAE-CSDF) to solve these problems. According to the characteristics of bearing faults, the proposed weighted domain fusion strategy can ensure the direction and balance in the transfer process. The proposed class separation degree can improve the accuracy of the target domain indirectly by extending the differences between the classes in the source domain. The effectiveness of the SAE-CSDF is verified via the mutual transfer of two public datasets and one laboratory dataset. The results show that the accuracy of the algorithm can reach 97% in the transfer between different machines, even if there is no labeled fault data in the new machine.

Online parameter identification is vital for boosting the accuracy of the battery equivalent circuit model (ECM) under dynamic profiles. However, traditional recursive least squares (RLS) method easily decays with the noise corruption from sensors or insufficient exciting signal in reality, which further limits the performance of ECM in battery modeling and states estimation. This article thus proposes a reliable online parameter identification method for battery ECM, which utilizes a well-designed information appraisal procedure based on the Fisher-information-based Cramer–Rao lower bound (CRLB). Without increasing much computing complexity, a comprehensive appraisal indicator, derived recursively from CRLB, enables a new mechanism for online parameter updating. Simulation and experimental results prove the validity of the proposed method under different driving cycles, temperatures, and aging conditions. The results show that the identification accuracy of the proposed method has been significantly improved comparing with a typical RLS and a multiple adaptive forgetting factors RLS method.

Most of the mechanical fault diagnosis methods of induction motors (IMs) are based on vibration signals. However, vibration sensors are expensive and require direct contact with IMs. This article proposes a data-driven mechanical fault diagnosis method for IMs using stator current signals, which is more convenient and low cost, since no additional sensors are required. Aiming at the weak representation of mechanical faults in current signals, an intelligent noise elimination method based on noise reconstruction model is proposed to improve the signal-to-noise ratio. Through the automatic feature extraction and classification of the residual current envelope spectrum, high diagnosis accuracy can be obtained even if some differences have existed among samples. The effectiveness of the proposed method is verified by high accuracy diagnosis results on two experimental platforms. The results show that the average diagnostic accuracy of the proposed algorithm for one bearing fault and two eccentricity faults can reach 96%. Even if the fault type and the fault degree are distinguished at the same time, an accuracy of 90% can be achieved for six kinds of bearing faults.

This paper proposes a finite control set model predictive torque control (FCS-MPTC) with reduced computation burden for permanent magnet synchronous motor (PMSM). Based on discrete space vector modulation (DSVM), the proposed DSVM-MPTC extends the control set. Therefore, compared to the traditional FCS-MPTC, it greatly reduces the torque ripples and current harmonics while maintaining a constant switching frequency. Furthermore, a new voltage vector optimization strategy is employed to decrease the number of candidate voltage vectors from 38 to 13, the proposed DSVM-MPTC significantly reduces the computation burden without compromising the control performance. The validity of the proposed DSVM-MPTC is verified by comparative experiments.

This paper presents an enhanced hierarchical control framework of microgrids, as an effective complementary of conventional hierarchical control, to improve operation efficiency and perform thermal management. The enhanced hierarchical control framework is implemented by two control levels. In primary control level, droop control strategy with active thermal management is proposed to perform identical thermal sharing among paralleled inverters. Furthermore, secondary control level is developed to improve operation efficiency and perform thermal management in system-level according to time-varying load profiles, where finite state machine model is adopted to optimize operation number of inverters. In addition, in gridconnected microgrid, resonance phenomena may be caused by time-varying grid impedance. Hence, resonance mitigation capability of microgrid is developed in the secondary control level, and small signal model of grid-connected microgrid is established to investigate the resonance phenomenon. Simulation and experimental results show that the proposed hierarchical control framework is able to perform active thermal management, improve operation efficiency, and mitigate harmonic resonance in either autonomous mode or gridconnected mode. The proposed hierarchical control framework is an enhanced complementary for conventional hierarchical control of microgrids.

Maternal sepsis is a life-threatening condition and ranks among the top five causes of maternal death in pregnancy and the postpartum period. Herein, we conducted a retrospective study on sepsis cases to explain the related risk factors by comparing them with bloodstream infection (BSI) and control maternities. In total, 76 sepsis cases were enrolled, and 31 BSI and 57 maternal cases of the same age but with neither sepsis nor BSI were set as controls. Genital tract infection (GTI) and pneumonia were the two most common infection sources in both sepsis (22 cases, 29% and 29 cases, 38%) and BSI cases (18 cases, 58% and 8 cases, 26%). Urinary tract infection (UTI)/pyelonephritis (9 cases, 12%) and digestive infection cases (11 cases, 14%) only existed in the sepsis group. Significantly different infection sources were discovered between the sepsis-death and sepsis-cure groups. A higher proportion of pneumonia and a lower proportion of GTI cases were present in the sepsis-death group (17 cases, 45% pneumonia and 9 cases, 24% GTI) than in the sepsis-cure group (12 cases, 32% pneumonia and 13 cases, 34% GTI). In addition, although gram-negative bacteria were the dominant infectious microorganisms as previously reported, lower proportion of gram-negative bacteria infectious cases in sepsis (30 cases, 50%) and even lower in sepsis-death group (14 cases, 41%) was shown in this study than previous studies. As expected, significantly greater adverse maternal and fetal outcomes, such as higher maternal mortality (26.3% vs. 0% vs. 0%), higher fetal mortality (42.2% vs. 20.8% vs. 0%), earlier gestational age at delivery (26.4 ± 9.5 vs. 32.3 ± 8.1 vs. 37.7 ± 4.0) and lower newborn weight (1,590 ± 1287.8 vs. 2859.2 ± 966.0 vs. 3214.2 ± 506.4), were observed in the sepsis group. This study offered some potential pathogenesis and mortality risk factors for sepsis, which may inspire the treatment of sepsis in the future.