Intelligence Defect Detection of the Water Wall in Boiler System Based on CNN

Abstract

As an important part to ensure the normal operation of the boiler, the defect detection of water walls based on the internal surface information of water walls has been paid more and more attention. But the traditional manual detection method is time-consuming and laborious with the low efficiency. Therefore, an intelligence detection method for detecting water wall defects in boiler systems based on the convolutional neural network (CNN) is proposed. The CNN has the characteristics of local connection, weight sharing, pooling and multi-layer structure, which can effectively extract local features of the water wall to build the intelligence defect detection model. Finally, the proposed method is used in the defect detection of the water wall of an actual thermal power plant. The experimental result has proved the effectiveness of the proposed method, And the recall rate and identification speed of the proposed method can meet the requirements of intelligent detection of the water wall defects.