DC micro-grid fault detection method based on data driving

By adopting a data-driven fault detection method in the DC microgrid, using the state space model and unknown input observer, the problem of fault detection under unknown system parameters is solved, and efficient fault detection and system stability improvement is achieved.

CN119986238AActive Publication Date: 2025-05-13NORTHEASTERN UNIV CHINA
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202510068984.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-01-16
Publication Date
2025-05-13
Estimated Expiration
2045-01-16

AI Technical Summary

Technical Problem

The prior art is difficult to implement fault detection in DC microgrids, especially when system parameters are unknown.

Method used

Using a data-driven method, the state space model of the DC microgrid is constructed, input and output data is collected, and the fault signal is detected using subspace identification technology and unknown input observers.

Benefits of technology

It realizes fault detection when the system parameters are unknown, and improves the fault detection capability of the DC microgrid and the stability and safety of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119986238A_ABST
    Figure CN119986238A_ABST
Patent Text Reader

Abstract

The invention provides a DC micro-grid fault detection method based on data driving, and relates to the technical field of DC micro-grid fault detection and safety control. Under the condition that parameters of a micro-grid system are unknown, input and output data are obtained through system operation, and an unknown input observer is constructed for fault detection. Compared with a single system, the method has the advantages that non-measurable interconnection items are estimated in a data-driven framework, and a system matrix is identified. In addition, on the basis of designing an unknown input observer, the design of the observer is converted into solving a linear matrix inequality and considering a robust performance index and a sensitive performance index, and finally fault detection of each DC motor is realized. According to the method, under the condition that only input and output data are obtained, the designed detector can still complete a detection task, in addition, according to the method, a large amount of system operation process data are used, online use can be achieved through direct programming on a computer, and the cost is quite low.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • DC micro-grid fault analysis method based on Pearson

    CN118501614A

  • Wind power torque fault-tolerant control method based on data driving output feedback

    CN112748663A

  • Direct-current micro-grid fault and network attack distinguishing detection method and system

    CN115473210A

  • Fault detection method for RLC (Radio Link Control) fractional linear circuit

    CN115508665A

  • Large-scale industrial process distributed fault detection method based on subspace identification

    CN117075560A