Abstract: Disclosed is a method for recognizing distribution network equipment based on Raspberry Pi multi-scale feature fusion. The method includes obtaining an initial sample data set; constructing an object detection network composed of EfficientNe-B0 backbone network, multi-scale feature fusion module and a regression classification prediction head; training the object detection network by taking the initial sample data set as a training sample; finally, detecting inspection pictures by using a the trained object detection network. A light-weight EfficientNet-B0 backbone network feature extraction method obtains more features of objects. Meanwhile, an introduction of multi-scale feature fusion better adapts to small object detection, and a light-weight y_pred regression classification detection head is effectively deployed and realized in Raspberry Pi embedded equipment with tight resources and limited computing power.

Abstract: Disclosed is a condition assessment method for an outdoor post-mounted vacuum switch. The method includes: establishing a condition assessment mathematical model for the post-mounted vacuum switch and acquiring scores of condition indicators of the post-mounted vacuum switch; and acquiring a condition assessment score based on the condition assessment mathematical model of the post-mounted vacuum switch and according to the scores of the condition indicators of the post-mounted vacuum switch so as to realize on-line monitoring of a health condition of the post-mounted vacuum switch and assess the health condition of the post-mounted vacuum switch. Further disclosed is a condition assessment device for the outdoor post-mounted vacuum switch.