研究目的
To detect and identify the fault location and faulty line in multi-core control and instrumentation cables in nuclear power plants using a clustering algorithm based on time-frequency domain reflectometry results.
研究成果
The proposed cluster TFDR method successfully detects and identifies fault locations and faulty lines in multi-core cables, as verified by experiments with open and thermal fault scenarios. It enables on-line condition monitoring and quantitative assessment of cable health, supporting condition-based maintenance in nuclear power plants.
研究不足
The method may be susceptible to noise and crosstalk, and the accuracy depends on the design of the incident signal and the clustering parameters. It is specific to multi-core cables and may require adjustments for different cable types or environments.
1：Experimental Design and Method Selection:
The study uses time-frequency domain reflectometry (TFDR) combined with a K-means clustering algorithm. TFDR involves designing an optimal incident signal, measuring reflected signals, and analyzing them using time-frequency cross-correlation and phase synchrony. The clustering algorithm groups data to classify faulty and normal lines.
2：Sample Selection and Data Sources:
Experiments were conducted on a 4-core Halogen Free Insulation (HFI) cable, simulating two fault scenarios: an open fault at a terminal block and a thermal fault induced by a heating chamber.
3：List of Experimental Equipment and Materials:
Equipment includes an arbitrary waveform generator (AWG), a digital storage oscilloscope (DSO), a signal processing system, a heating chamber, and cables. Materials include the 4-core HFI cable.
4：Experimental Procedures and Operational Workflow:
The incident signal is generated by the AWG and applied to the cable. Reflected signals are measured by the DSO and processed using time-frequency analysis. Clustering is performed on cross-correlation and phase synchrony results.
5：Data Analysis Methods:
Data is analyzed using K-means clustering with Euclidean distance, and results are interpreted to identify fault locations and faulty lines.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
