Single-phase earth fault research and judgment method based on multi-algorithm normalization analysis
A single-phase ground fault and normalization technology, applied in the field of distribution network, can solve problems such as line tripping, endangering system insulation and equipment safety, and power outages for users
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0033] Embodiment 1. A single-phase ground fault research and judgment method based on multi-algorithm normalization analysis, comprising the steps: step 1, using the pre-trained deep learning model to analyze the working condition data (i.e. data The operating condition data obtained by image feature extraction) is used for fault identification to determine whether a fault occurs; step 2, using the pre-trained decision tree to identify multiple operating condition types for the operating condition data that is determined to have a fault.
[0034] 1) Input data image feature extraction
[0035] The selected input signal is the A, B, C three-phase current waveform on the line, and the waveform length is the data from 4 power frequency cycles before the fault occurs to 8 power frequency cycles after the fault occurs. To visualize the input data, the waveform sequence must first be matrixed. Calculated at a sampling rate of 1K, the input waveform sequence is a=[1×3072]-dimension...
Embodiment 2
[0051] Embodiment 2. On the basis of Embodiment 1, this embodiment also includes the steps of result verification after the obtained multi-working-condition type discrimination results, including:
[0052] The fault recognition rate, an evaluation index in traditional pattern recognition, is introduced, and its calculation method is as follows:
[0053]
[0054] In the formula: T is the input known failure test sample set; C is the number of correctly identified samples in the set.
[0055] In order to reflect the recognition stability of the model for various fault signals, the evaluation index of wrong (missing) fraction rate is introduced, and its calculation method is as follows:
[0056]
[0057] In the formula: N is the set of various fault test samples input; M is the number of wrong (missed) samples in the set.
[0058] The test uses a variety of different types of faults (including grounding, short circuit, inrush current, accumulation, power recovery, and powe...
Embodiment 3
[0067] Embodiment three, as image 3 As shown, on the basis of the above embodiments, this embodiment provides a single-phase ground fault research and judgment method based on multi-algorithm normalization analysis, which also includes the following steps after completing the identification of multiple working conditions:
[0068] According to the model and classifier formed in step 1 and step 2, the incoming waveforms are matched and analyzed one by one, and the steady-state data of the real-time system is superimposed at the same time for multi-algorithm normalization analysis, and the fault identification results obtained by classifying steps 1 and 2 are used , using steady-state information such as voltage, current, and network topology to conduct comprehensive analysis, form fault intervals and form operable fault handling strategies, and complete all fault research and judgment logics.
[0069] The specific analysis method is as follows:
[0070] a) Obtain the waveform...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com