Fault diagnosis device for three-phase asynchronous motor based on digital twin and transfer learning

Abstract

The invention discloses a three-phase asynchronous motor fault diagnosis device based on digital twins and transfer learning, including a terminal sensing system for sensing various electrical parameters of a three-phase asynchronous motor; used to construct a three-phase asynchronous motor digital twin and a fault diagnosis model remote data center; build a three-phase asynchronous motor fault diagnosis model through deep learning, and train it through a large number of rich data sets generated by digital twins, which greatly improves the accuracy of fault diagnosis model detection; through transfer learning, only need With a small amount of new samples, the fault diagnosis model can be updated, and the robustness and stability of the fault diagnosis model detection for three-phase asynchronous motors can be improved.

Description

technical field [0001] The invention relates to the field of three-phase asynchronous motor fault diagnosis devices and methods, in particular to a three-phase asynchronous motor fault diagnosis device based on digital twin and transfer learning. Background technique [0002] As the most important motive force and driving device in industrial production, three-phase asynchronous motor is the core equipment in coal mine, energy, chemical industry and other industrial fields. Once the three-phase asynchronous motor fails, the entire industrial operation system may face the risk of complete paralysis , causing unpredictable economic losses and even endangering the lives of on-site workers. Therefore, the fault detection and fault diagnosis of three-phase asynchronous motors have important research significance. [0003] At present, most of the fault diagnosis of three-phase asynchronous motors is in the stage of on-site detection and diagnosis, and a small number of them have ...

AI Technical Summary

Problems solved by technology

[0002] As the main driving force and driving device in industrial production, the three-phase asynchronous motor is its core equipment in coal mines, energy, chemical industry and other industrial fields. Once the three-phase asynchronous motor fails, the entire industrial operation system may face the risk of complete paralysis , causing unpredictable economic losses, and even endangering the lives of on-site staff

On-site diagnosis requires experts to go to the work site for measurement and diagnosis. Limited by space, work efficiency is low

Remote virtual simulation, limited by sensors and communication technology, cannot be mapped in real time, so the accuracy of three-phase asynchronous motor fault diagnosis is low, and the real-time performance is poor

With the continuous development of big data and artificial intelligence technology, the fault diagnosis methods of three-phase asynchronous motors are also continuously improved. The existing data-driven fault diagnosis methods for three-phase asynchronous motors need to collect a large amount of actual data before they can be used. Training, but the three-phase asynchronous motor will inevitably be affected by its own wear and working environment changes during use. The fault diagnosis model trained before is no longer applicable. Retraining requires collecting a large number of three-phase motors in various states. Asynchronous motor operation data, time-consuming and low efficiency

Images