In order to solve the problems of low accuracy and slow speed of the current solar panel defect detection method, a defect detection algorithm based on improved YOLOv11 was proposed. Firstly, the SimSPPF module is introduced into the backbone network to optimize the feature extraction process. In addition, the Slide Loss function is used to improve the attention of the model to difficult samples. At the same time, the LSKA attention mechanism is introduced into C2PSA, the split convolutional kernel is used to enhance the feature extraction ability, and the Mish activation function is used to enhance the network nonlinearity. Finally, the Strip Pooling strategy was introduced to improve the adaptability of the model to the changes of target shape and distribution. The experimental results show that the improved algorithm Persion reaches 86.8%, which is 3.3% higher than the original algorithm, mAP@0.5 reached 90.1%, an increase of 2.6% compared with the original algorithm, The detection speed reaches 149.254 fps, which meets the requirements of high precision and high efficiency of solar panel defect detection in industrial production.