Intelligent quality control method for micromotor based on noise characteristics

Abstract

The invention discloses an intelligent quality control method for a micromotor based on noise characteristics. The method comprises the steps of a, collecting a noise waveform signal of the micromotorin an operation state, and manually labeling a sample; b, directly extracting a sound feature F from the noise waveform signal of the micromotor collected in the step a, and forming a sample micromotor noise feature set T for training in combination with manual sample labeling; c, establishing a micromotor fault intelligent identification model; d, acquiring a target micromotor noise signal; e, extracting a signal feature set from the target micromotor noise signal; and f, inputting the signal feature set into the micromotor fault intelligent identification model, and automatically judging whether the target micromotor has quality defects or not by the model. The method can adapt to intelligent quality control scenes with various field fault types and extremely low micromotor operation noise, and can realize the detection effects of omission ratio of 1% and false detection rate of 0.1%.

Description

technical field [0001] The invention relates to the quality technical field of micro-motor manufacturing, in particular to an intelligent quality control method for micro-motors based on noise characteristics. Background technique [0002] Since the invention of generators and motors, motors have been widely used in all aspects of social life. Micro motors refer to motors with a diameter of less than 160mm or a rated power of less than 750mW. Due to their small size and light weight, they are used in smart home appliances and micro robots. , precision engineering, aerospace and other technology-intensive fields. There are many manufacturing processes of micro-motors, involving precision machinery, fine chemicals, micro-machining, magnetic material processing, winding manufacturing, insulation treatment and other process technologies, requiring a large number of process equipment and high precision. At the same time, the micro-motor production line has a large output and hig...

AI Technical Summary

Problems solved by technology

[0005] The defects of the existing technical solutions are as follows: In the manufacturing process, the traditional micro-motor quality inspection is a manual inspection solution, which relies on the quality inspection personnel to manually identify the difference in the sound of the motor running

This method not only requires a lot of labor costs, but is also easily affected by factors such as the status of quality inspectors, resulting in false or missed detection

In addition, the existing automatic motor fault detection scheme based on audio features is essentially a template matching method, and its fault detection effect is related to the discriminant function based on the fault characteristic frequency used in the detection scheme. The discriminant function effect depends on the sample set and the weight coefficient. Therefore, the discriminant function can only be used to detect or predict intelligent quality control scenarios where motor models and fault types are similar to the sample set, and the operating noise of the micro motor is extremely small, which is easily affected by the background noise of the acquisition hardware.

Images