Tightening technology online fault detector based on process data analysis and method of detector

Abstract

The invention discloses a tightening technology online fault detector based on process data analysis and a method of the detector. The tightening technology online fault detector comprises a tightening process data acquisition module, a processor and a fault classifier, the tightening process data acquisition module is used for acquiring bolt tightening process data in real time and transmitting the data to the processor, the processor is used for preprocessing the received bolt tightening process data, retaining tightening process data with pure rising moments, transmitting the preprocessed tightening data to the fault classifier acquired through offline training to perform fault classification and outputting fault classification results, particular processes of the fault classifier acquired through offline training include that according to historical data of tightening processes and corresponding faults of the tightening processes, the historical data of the tightening processes areclustered to form a plurality of data subsets, tags are set to form a tightening technology fault training sample set, and learning training of process curves of tightening technology is performed bythe aid of a machine learning classification algorithm based on the tightening technology fault training sample set to obtain the fault classifier.

Description

technical field [0001] The invention belongs to the field of on-line fault detection, and in particular relates to an on-line fault detection instrument for a tightening process based on process data analysis and a method thereof. Background technique [0002] The bolt tightening process involves the assembly quality of the product, especially in the engine, motor vehicle, ship and other industries, which is the root of the three leaks and other problems. Common tightening equipment usually judges the quality of tightening by whether the final torque reaches the set value. Combined with simple The over-limit alarm method is used to light the green or red light to indicate the qualified or unqualified state of the tightening. It is difficult to find the tightening defects caused by quality problems such as bolts and nuts. For example, impurities in the threads will cause the bolts to fail to maintain ideal clamping for a long time power and other potential failures. [0003]...

AI Technical Summary

Problems solved by technology

[0002] The bolt tightening process involves the assembly quality of the product, especially in the engine, motor vehicle, ship and other industries, which is the root of the three leaks and other problems. Common tightening equipment usually judges the quality of tightening by whether the final torque reaches the set value. Combined with simple The over-limit alarm method is used to light the green or red light to indicate the qualified or unqualified state of the tightening. It is difficult to find the tightening defects caused by quality problems such as bolts and nuts. For example, impurities in the threads will cause the bolts to fail to maintain ideal clamping for a long time Potential failures such as force

[0003] Moreover, there are various types of bolt faults, including bolt slippage, bolt jamming, and poor thread quality. Only through the existing fault detection device can it be impossible to accurately obtain the fault type of the bolt during the tightening process, and thus cannot timely correct The bolts are overhauled, which affects the efficiency of the bolts

Images