Welding anomaly detection method and device, computer equipment and storage medium

Abstract

The invention relates to a welding anomaly detection method and device, computer equipment and a storage medium. The method comprises: receiving a welding anomaly detection request; feeding back various preset welding anomaly detection models; obtaining corresponding model combination information and weight setting information; inputting the welding data into a preset welding anomaly detection model, and obtaining each model detection result corresponding to the welding data; and according to each model detection result and the weight setting information, receiving a multi-model welding anomaly detection result corresponding to the welding anomaly detection request. After the welding anomaly detection request is received, the operation of selecting a proper preset welding anomaly detection model combination according to various fed-back preset welding anomaly detection models by field detection personnel is responded, and the weight ratio of the models is set by responding to the field detection personnel, therefore, under the condition of different requirements on the basis of complex and changeable working conditions, the welding anomaly detection recognition rate for various welding data is improved.

Description

technical field [0001] The present application relates to the field of welding technology, in particular to a welding abnormality detection method, device, computer equipment and storage medium. Background technique [0002] Welding is one of the most important processes in the field of machinery manufacturing. According to different welding properties, methods, and applications, welding can be divided into manual welding, semi-automatic welding, and automatic welding. Due to the occurrence of abnormal welding during the welding process, it often leads to missing welding, welding penetration, and uneven welding seam formation. Therefore, it is necessary to perform welding abnormality detection after the welding is completed, so as to ensure the quality of the welded product. [0003] At present, it is generally possible to detect welding anomalies through anomaly detection models. However, during welding inspections, customers' needs for welding products and welding conditi...

AI Technical Summary

Problems solved by technology

[0003] At present, it is generally possible to detect welding anomalies through the anomaly detection model. However, during welding inspection, the customer's demand for welding products and welding conditions often change with the order requirements. At this time, a single anomaly detection model generally cannot meet the requirements. All conditions of welding anomalies, which affect the recognition rate of welding anomalies detection

Images