Knowledge migration and data generation-based photovoltaic panel fragmentation detection method

• Main Authors: WU CHUNPENG, LIN LONG, SUN YI, SHI CUNCUN, CHEN MINGHAO, SUN SHUMIN, LI YONG, YU PENG
• Format: Patent
• Language: Chinese; English
• Published: 26.07.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention relates to the technical field of photovoltaic detection, and discloses a photovoltaic panel fragmentation detection method based on knowledge migration and data generation, and the method comprises the steps: constructing a source domain data set and a target domain data set, training a preset network model by using the two data sets to obtain a first network model containing source domain knowledge and a second network model containing target domain knowledge; training a preset network model by using the source domain data set and the target domain data set and aligning with distribution previously learned by the first network model and the second network model to obtain a third network model containing uniform distribution of source domain samples and target domain samples; and removing the generation network of the third network model, adding a classification layer at the tail end of the identification network for outputting a detection result, fixing the network parameters trained before be