A workshop control method and system based on deep learning

Abstract

The invention discloses a workshop control method and system based on deep learning. The method includes: collecting data of the whole life cycle of the parts of the processing equipment; constructing an equipment failure prediction model based on artificial neural network, using the data of the whole life cycle of the parts to train and test the equipment failure prediction model, and using the trained equipment The failure prediction model outputs the predicted failure time data of the processing equipment; constructs the workshop scheduling model, and obtains the maintenance duration data of the processing equipment, and uses the maintenance duration data and the predicted failure occurrence time data to construct the workshop scheduling Special processing constraints in the model; solve the workshop scheduling model, obtain the workpiece processing plan with the shortest total processing time, and send workpiece processing instructions. The invention can further shorten the workpiece processing time, reduce or avoid rescheduling, and improve workshop production efficiency.

Description

technical field [0001] The invention belongs to the technical field of workshop control, and more particularly, relates to a workshop control method and system based on deep learning. Background technique [0002] Shop floor control refers to determining the processing sequence of multiple workpieces in a shop on multiple processing equipment. With the development of intelligent and flexible workshops, workshop data is becoming larger and larger, manufacturing systems become more and more complex, and dynamic events occur more and more frequently. Traditional workshop dynamic control methods are not enough to solve increasingly complex actual production problems. . [0003] In the actual production control process, unexpected events such as processing equipment failure and order insertion will occur. In the prior art, the most common is to use a pre-reactive scheduling method for dynamic events. Pre-reactive scheduling is a rescheduling process that modifies the original ...

AI Technical Summary

Problems solved by technology

[0005] In the prior art, the scheme of using deep learning method to predict the fault of workshop equipment is also proposed, but the prior art only stays at the level of fault diagnosis and equipment maintenance, which can prevent the dynamic event of workshop machine failure to a certain extent occurrence, but failed to fundamentally solve the impact of machine failures on workshop control

Images