Fault detection of electrolyzer plate based on improved Mask R-CNN and infrared images

Abstract

Non-ferrous metals are very important strategic resources, electrolysis is an essential step in refining non-ferrous metals. In the electrolysis process, plate short circuit is the most common fault, which seriously affects output and energy consumption. The rapid and accurate detection of faulty plates is of great significance to the metal refining process. Given the weak generalization ability and complex feature rule design of traditional object detection algorithms, and the poor detection effect of existing deep learning models in infrared images with many interference factors, an improved Mask R-CNN based fault detection algorithm is--proposed. Improve the generation strategy and the NMS algorithm of proposals to reduce the missed detection; propose the Globally Generalized IoU loss function to characterize better the position and scale relationship between the predicted box and the target box, which is beneficial to the bounding box regression. The experimental results show that the improved model has an accuracy rate of 10.4% higher than the original model, reaching 86.8%. Compared with the common one-stage and two-stage object detection models, the improved model has a stronger detection ability. This algorithm has some reference value for the accurate detection and location of electrolytic cell faults.

Recognition of Piano Play Gesture Based on Infrared Sensor Detection Rod

“…The positionbased method can provide some position information for proportional control, and the phase-based method can be used to detect most gestures. The combination of these two methods can provide a powerful solution for gesture sensing [7]. Characteristics based on spatial characteristics: In the process of piano playing, the posture angles between the upper and lower joints of the fingers and between the fingers and the back of the hand will be quite different, so the information of the fingers and the back of the hand at the key pressing moment is extracted.…”

A lightweight high-resolution algorithm based on deep learning for layer-wise defect detection in laser powder bed fusion

“…Deep learning-based object detection algorithms, capable of recognizing and locating multiple targets in images, have found widespread application across various industries [40][41][42]. Structurally, object detection can be divided into twostage algorithms, represented by Faster R-CNN [43], and onestage algorithms, represented by SSD [44] and you only look once (YOLO) [45].…”

An attention-based long short-term memory prediction model for working conditions of copper electrolytic plates

“…(3) Li et al [9] proposed an infrared thermal imaging camera to collect infrared thermal images of the electrolytic cell in real time, using a modified difference of Guassian (DoG) filter to differentiate the boom of the cathode plate, and to indirectly locate the boom of the faulty plate through the detected normal boom, thus realizing effective monitoring of the working states of the plate. (4) Zhu et al [10] proposed a method using convolutional neural networks in combination with infrared thermal images to monitor the working states of electrodes in copper electrolysis. Their method corrects the distortion of the acquired thermal images and then segments the single tank image, extracts the pixel features of the faulty plate and classifies them, and uses a focal-loss-function-based, faster-region convolutional neural network (i.e.…”

“…The method proposed by Aqueveque et al [8] requires a large amount of manpower for inspection to overcome the disadvantages of its long inspection period. The method proposed by Li et al [9] innovatively uses a computer instead of manual inspection to monitor the states of the electrolytic cells, and Zhu et al [10] utilized deep learning algorithms, which have become very popular in recent years, to investigate the problem; both have shown good results. However, the shortcoming of these methods is that by the time they detect a plate failure, the failure has already been present for a long time and has incurred a negative impact on the quality of the final output product, and therefore the economic performance of the plant.…”