The disadvantages of infrared image are low resolution, bad stereoscopic sense, fuzzy image and low SNR, according to the application of infrared image in fault diagnosis of electronic power equipment, in this paper ,we make a comparative research on pre-processing technique of image de-noising and enhancement, and propose an infrared image enhancement algorithm based on platform histogram equalization combined with enhanced high-pass filtering, the algorithm can effectively improve the contrast by comparison, it is obvious to the noise effect, highlighting the objectives and details, and makes a good foundation for the subsequent target identification and fault diagnosis.
nfrared temperature measurement technology is an important means that early detection of equipment external overheating fault and internal insulation fault, and advanced detection methods of power equipment condition monitoring.Infrared temperature measurement technology found the hidden dangers of electrical equipment heating of the external connection points, has been widely applied in power system. With comparing the original ant colony optimization, this paper presents an improved weighted Ant Colony Optimization path planning algorithm based on Infrared image acquisition noise reduction processing and combining WCDMA network transmission mode, and constitute a visual background management system with rural power network .The system could capture power equipment infrared images and reduction noise simply, send it to the background server through the 3G network, for artificial auxiliary to predicting grid failure, scientific planning repair inspection path with improved navigation algorithm, and improve the stability of the power grid operation.
Bearings are important parts in industrial production and are related to the normal operation of mechanical equipment. For bearing fault identification， traditional method often includes feature extraction, which involves professional prior knowledge and is time-consuming. This paper proposes the deep convolution residual network (DCRN) to identify the bearing fault. Based on the end-to-end learning characteristics of deep neural networks, this method can directly use raw data for training, and does not require feature extraction. Moreover, under the effect of skip connection, DCRN can exert the powerful fitting ability of deep neural network. In this paper, by stacking residual blocks, three different architecture of DCRN are designed and all three achieve very high accuracy, respectively 99.60%, 99.71% and 99.81%. Compared with other methods, DCRN have better generalization performance.
Multilevel Threshold Segmentation Method for Infrared Image of Power Equipment with Two-dimensional Entropy Based on Bat Algorithm
