Text Classification and Fault Diagnosis Method of Power Dispatch Alarm Signals Based on Similarity
A technology of alarm signal and power dispatching, which is applied in the direction of control/regulation system, electrical test/monitoring, program control, etc., can solve problems that affect the correctness of fault diagnosis, chaotic alarm signal timing, failure to diagnose faults, etc., and achieve output diagnosis The results are accurate and reliable, which is conducive to rational use and the effect of improving the accuracy of fault diagnosis
Active Publication Date: 2019-04-12
ZHEJIANG UNIV +2
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the time scale information of the alarm signal in the power dispatching system is only accurate to the second. When an accident occurs, the alarm signal is sent out at a speed of several per second. Therefore, the alarm signal in the accident phase is often in a chaotic time sequence, which will affect the accuracy of the method. The correctness of fault diagnosis, and this method can only diagnose a single fault, and cannot diagnose several different faults that occur at the same time
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0058] In order to verify a similarity-based power dispatching alarm signal text classification and fault diagnosis method proposed by the present invention, the inventor used the C++ programming language to develop and implement the fault diagnosis program, and implemented the implementation on a PC configured in the following environment Example testing and verification.
[0059] Intel Core i5-4250U CPU 1.30GHz, 4GB RAM, Microsoft Windows 10x64, Visual Studio 2013.
[0060] The accident samples and real-time alarm signals used in this embodiment are all from the dispatching system database of Zhejiang Electric Power Dispatching Command Center. The accident samples are taken from the typical accidents and their corresponding alarm signals stored in the database from March 2014 to April 2017. In order to ensure compliance with the actual operation of the dispatching system, the real-time alarm signals use the data stored in the database in March 2014. All the signals received...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a power dispatching alarm signal text classification and fault diagnosis method based on similarity. A power dispatching system will receive a large number of alarm signals in the fault process of a power system. How to classify the alarm signals and diagnose faults according to the alarm signals is the problem to be solved in the invention. According to the method, an alarm signal text is vectorized, real-time alarm signals are extracted using a sliding time window, the real-time alarm signals are classified by calculating the similarity between the vector spaces of the real-time alarm signals and accident sample alarm signals, and faults are diagnosed according to the text classification result. A variety of different faults occurring to the power system at the same time can be diagnosed using the method. The method can be used in the field of power dispatching system alarm signal text classification and fault diagnosis.
Description
technical field [0001] The invention belongs to the technical field of fault diagnosis of electric power dispatching system, natural language processing of alarm signal and Chinese text classification, and in particular relates to a text classification and fault diagnosis method of electric power dispatching alarm signal based on similarity. Background technique [0002] Power system security is the core of building a strong and smart grid. Although the theory and technology of modern power systems continue to develop, failures that endanger the safety of power systems are still unavoidable. Fault-related alarm signals will be generated during the fault process of the power system. The power dispatching system needs to use these alarm signals to quickly and accurately perform fault diagnosis, judge the fault location, fault type and fault equipment, identify incorrectly operated protection and circuit breakers, and assist Dispatchers or operators handle failures. However, w...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G05B23/02
CPCG05B23/0243G05B2219/24065
Inventor 汪崔洋江全元唐雅洁朱炳铨项中明高强童存智吴利锋徐祥海李跃华
Owner ZHEJIANG UNIV