Circuit board fault detection method and system based on measurement data machine learning

Abstract

The invention provides a circuit board fault detection method and system based on measurement data machine learning. In the training process, a user uploads historically-stored original defective product maintenance record data to a server in bulk; the server classifies the data of different types of circuit boards and performs feature engineering; and for each subclass, separate modeling and training is carried out by utilizing a machine learning algorithm. In the forecasting process, the user uploads corresponding feature instances of defective circuit boards; the server carries out classification according to product types, and then, extracts available characteristics by utilizing the result of the feature engineering; and a trained machine learning model is utilized to analyze characteristics of a current instance and predict fault cause. On one hand, fault detection is carried out by combining data analysis and machine learning technology, thereby greatly reducing labor cost and time for maintenance and test; and on the other hand, feature selection and incremental learning are realized, thereby improving overall system training efficiency and accuracy.

Description

technical field [0001] The invention relates to the technical field of circuit board fault detection, in particular to a circuit board fault detection method and system based on measurement data machine learning. Background technique [0002] In the production environment of electronics manufacturers, each electronic product will have various performance indicators specified. As for the employees of the testing and maintenance assembly line, their task is to judge whether a product is qualified according to various basic indicators, and to find the problem of the unqualified product. Taking circuit boards as an example, since each circuit board product has a certain degree of difference, and each product has a large number of components, it is quite difficult for the staff of the testing and maintenance department to detect the cause of failure of defective products. In the past, employees checked the defective products point by point based on visual inspection or using mea...

AI Technical Summary

Problems solved by technology

[0003] However, similar to this method of troubleshooting defective products based on manual experience, first, the reliability needs to be improved, and mistakes are prone to occur; second, it takes a long time and the efficiency of troubleshooting is low; The non-inheritability, when a new circuit board product arrives, employees need to relearn and summarize troubleshooting experience

Images