Search

Featured Application: When applied to online condition monitoring, the proposed method can assist wind turbine operators in quickly identifying the types of faults that trigger alarms. Therefore, it can reduce operation and maintenance costs and downtime losses. Alarms generated by a wind turbine alarm system indicate the need for emergency action by operators to protect the turbine from running into risky conditions. However, it can be challenging for operators to identify the fault types that trigger alarms, particularly with few labeled fault samples. This paper proposes a novel fault diagnosis method for wind turbines with alarms that collaboratively uses labeled and unlabeled alarms to improve diagnosis accuracy. First, the proposed method distinguishes different alarm sequences using a designed Siamese convolutional neural network with an embedding layer (S-ECNN) model. Then, the fault category of an unknown alarm sequence is diagnosed based on similarity scores. Specifically, the Skip-gram model is used to mine potential relationships among alarms in unlabeled alarm sequences, and pretrained alarm vectors are obtained. In the S-ECNN model, the pretrained alarm vectors are further optimized and trained using labeled alarm sequences. The similarity scores are calculated based on the distance between the extracted discriminative features of alarm sequences. The effectiveness of the proposed method is validated using actual alarm data from a wind farm. [ABSTRACT FROM AUTHOR]

MoS2, as a new generation of 2-D materials after graphene, shows excellent properties and has attracted extensive attention. More importantly, it can be combined with graphene devices to improve their performance. In this work, a MoS2/graphene/GaAs Schottky junction near-infrared (NIR) photodetector is investigated. A MoS2 quantum dot (QD) absorption layer is added on the traditional graphene/GaAs structure, and the QDs are used to enhance the absorption rate of graphene for NIR light, thereby improving the device performance. In subsequent tests, it is found that the photodetector has a responsivity of 19.9 mA/W under the 808-nm incident light at zero bias, which is much higher than that of ordinary graphene/GaAs photodetectors. Furthermore, the detection rate of the device can reach $4.86\times10$ 10 cm $\cdot $ Hz $^{1/2}\cdot \text{W}$ −1, and the response/recovery times are 46.8 and $557 ~\mu \text{s}$ , respectively; it is also found that the response wavelengths of the device have been extended, and it has a certain response in the 1064-nm wavelength. The excellent performance of MoS2/graphene/GaAs structure indicates that it has good application prospects in the field of NIR photodetectors. [ABSTRACT FROM AUTHOR]