The detection method of the fault plate of the electrolytic cell

Abstract

The invention belongs to the technical field of motor manufacturing, and in particular relates to a method for detecting a faulty electrode plate of an electrolytic cell. First, a standard image is acquired and calibrated, and during the detection operation, the detection image is acquired, the electrolytic cell image is obtained based on the calibration information, and the quality of the electrolytic cell image is judged. Validity, analyze the effective electrolytic cell images, determine whether the corresponding electrolytic cell has faulty plates, locate the faulty plates, and then output the test results to end the test. Based on the calibration information, the invention can accurately and reliably obtain the image of the area in the electrolytic cell in the detection image, so that the standardized validity analysis and fault plate analysis can be performed on the image of the area in the side tank of each electrolytic cell, so as to efficiently and accurately Identify and locate faulty plates.

Description

technical field [0001] The invention belongs to the technical field of motor manufacturing, and particularly relates to a detection method for a faulty pole plate of an electrolytic cell. Background technique [0002] The copper electrolysis process is to make thick copper plates as anodes, pure copper or stainless steel sheets as cathodes, and insert them into the electrolyte alternately. Cathode diffusion, electron precipitation is obtained at the cathode, and high-purity metallic copper is obtained. During the copper electrolysis process, the plates are prone to "cold plate" and "short circuit". Due to the failure of the workers to arrange the plates, the conductive pollution of the electrolytic cell, etc., the contact of the plates is invalid, and no current or small current flows, which reduces the working efficiency of the electrolysis, which is called cold plate; due to the uneven current distribution between the plates, the anode slime sticks Additional reasons lea...

AI Technical Summary

Problems solved by technology

In the paper, it is considered that the false detection and missed detection are caused by the occlusion of the crane, the influence of the operators on the trough surface, the non-standard cover cloth and the limitations of the algorithm itself.

In order to further improve the fault detection index, an SVM-based fault cathode extraction method suitable for both the cloth area and the exposed area is constructed. However, each sample in this method corresponds to a nine-dimensional feature vector, and the application requires a lot of machine memory and calculation. time, and it is still impossible to avoid false detections and missed detections caused by truss trucks blocking, tank surface operators entering the image screen, etc., which cannot meet the needs of on-site applications

Images