Abstract
Power-grid faults pose a great threat to the economy and to social stability. In this paper, the detrended fluctuation analysis method is used to investigate the scaling properties of power grid faults and the correlation between faults and meteorological factors that are closely related to power grid faults. Multifractal detrended fluctuation analysis showed that the fault time series were multifractal. Further investigations revealed the origins of multifractality in power-grid fault time series, and the results showed that the temporal correlations in the data represent the distribution of the returns, which are a significant source of multifractal scaling. Then, cross correlations between the fault and the four meteorological factors were investigated using the detrended cross-correlation analysis method. The results showed that maximum wind speed has a considerable impact on the number of transmission system line faults per day, whereas daily precipitation, daily mean air temperature, and maximum wind speed have a considerable impact on the number of distribution system line faults per day.
Introduction
Power grids are among the most complex machines ever built, and their failure carries staggering economic and social costs. In regions like China’s Hunan Province, over 90% of these failures are triggered by meteorological events such as lightning, ice storms, or wildfires. Traditionally, these are viewed as isolated incidents. However, this research, published in IEEE Access, argues for a "global dynamics" perspective, applying advanced fractal mathematics to analyze years of fault data. The study investigates whether fault patterns exhibit "monofractal" (simple, consistent scaling) or "multifractal" (complex, time-varying scaling) properties to better predict long-term risks.
Methodology and Data
The researchers utilized two primary datasets from Hunan Province (2015–2018):
- Grid Fault Data: 2,497 occurrences across 531 high-voltage transmission lines (≥110 kV) and 5,217 distribution lines (10–35 kV).
- Meteorological Data: 81,030 records including relative humidity, precipitation, air temperature, and maximum wind speed.
To analyze this non-stationary data, the authors employed Detrended Fluctuation Analysis (DFA) and Multifractal Detrended Fluctuation Analysis (MFDFA) to detect long-range correlations and determine if faults exhibit "memory" (where a fault today influences the probability of one tomorrow). They also used Detrended Cross-Correlation Analysis (DCCA) to quantify the power-law correlation between grid faults and weather variables.
Key Insights: Scaling and Memory
The analysis revealed that power grid faults are not random. The calculated Hurst exponents (derived from scaling exponents $\lambda$) were greater than 0.5—specifically 0.74 for transmission and 0.77 for distribution systems. This indicates a "positive correlation" or "memory" within the system, meaning past events influence future occurrences.
Multifractal Nature of Grid Failures
The study confirmed that power-grid fault time series are multifractal. This complexity is primarily driven by the "fat-tailed" distribution of the data, meaning rare, extreme weather events have a disproportionately large impact on the system's mathematical structure. The multifractal spectrum exhibits a characteristic unimodal bell shape, which serves as a signature of this high complexity.
Weather Triggers and System Resilience
The research highlights a clear distinction between transmission and distribution network vulnerabilities:
- Transmission System Line Faults (TSLF): High-voltage lines are generally more resilient, with maximum wind speed identified as the primary driver of failure (often due to conductor galloping or windage yaw).
- Distribution System Line Faults (DSLF): These lower-voltage lines are the "weak link," showing significant sensitivity to a broader range of factors, including daily precipitation and mean air temperature.
Cross-correlation analysis also revealed semi-annual cycles (~180 days) in fault patterns, aligning with seasonal shifts in the region.
Conclusion
By identifying the fractal "signatures" of power grid faults, utilities can move beyond reactive maintenance. While the specific correlations found in this study are influenced by the unique terrain of Hunan Province, the methodology provides a universal framework for designing climate-adjusted, resilient infrastructure. Future work aims to leverage these findings to predict the scale of rare "10-year" or "50-year" fault events, allowing for proactive infrastructure design.
Summary updated