Power distribution network power equipment identification method based on multi-modal data fusion

• Main Authors: LIAO XIAO, WANG YING, MENG XIANGLIANG, CAO HUI, OU BEI, XU YONGGANG, LIU KEJUN, LIU YUHAN, ZHU ZHENG, GUAN QINGSHU, CUI WEI, YANG YANG, WANG SHILIN, LI QIANG, XIE KE
• Format: Patent
• Language: Chinese; English
• Published: 25.06.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a power distribution network power equipment identification method based on multi-modal data fusion, and the method comprises the following steps: 1, text data feature extraction: converting text data into a word embedding form, employing sine and cosine position coding as embedding, and inputting the text data into a Transform model for feature extraction; 2, image data feature extraction: carrying out preprocessing such as partitioning on the image, and carrying out feature extraction by using a VIT model; and 3, multi-modal data fusion: fusing the image and text data by using a mutual attention module to form fusion features. And 4, power distribution network power equipment identification: inputting the fusion features into a least square support vector machine classifier to realize power distribution network power equipment classification and identification. The performance of the recognition model is improved, and the recognition accuracy is improved. 本发明公开了一种基于多模态数据融合的配电网电力设备识别方