TL;DR
This paper introduces a novel Equivalent Heat Source (EHS) model for AC arcs in short copper-core cable gaps, significantly reducing computational complexity compared to traditional Magnetohydrodynamics (MHD) models. This EHS model is then used to develop a method for assessing the fire risk associated with these arcs, providing valuable insights for electrical safety standards.
What Problem Does This Paper Address?
Electrical equipment aging, faults, and damage can lead to electrode gaps breaking down and generating AC arcs. These arcs produce high temperatures that can ignite cable sheathing, causing severe fires. While Magnetohydrodynamics (MHD) models can accurately simulate these arcs by considering coupled physical fields (electric, temperature, flow, magnetic), they are computationally intensive, especially for studying random AC arcs. This complexity hinders the development of efficient fire risk assessment methods. Existing research has primarily focused on DC arcs or welding arcs with constant heat sources, leaving a gap in understanding the dynamic heat generation of AC arcs and its implications for cable fire risk.
Methods
The researchers first established a 2D axisymmetric AC arc MHD simulation model for a short gap in a copper-core cable. This model was validated against experimental data for arc voltage and current. The heat source distribution obtained from the MHD model was then fitted to develop an Equivalent Heat Source (EHS) model. This EHS model divides the AC arc's heat source into 16 independent segments. To optimize these segments and reduce temperature prediction errors, a correction matrix was introduced. This matrix was optimized using a combination of a backpropagation (BP) neural network for prediction and a genetic algorithm for finding the optimal correction matrix. Finally, this optimized EHS model was used to simulate the thermal characteristics of AC arcs and calculate the probability of cable fires under various random AC arc generation scenarios.
Datasets
Not discussed in the paper. The study relies on experimental data for validation and simulation results for model development.
Key Insights and Ideas
- Bridging the Gap: The paper successfully demonstrates that a simplified EHS model can accurately represent the complex thermal behavior of AC arcs in short gaps, offering a computationally efficient alternative to MHD models.
- Dynamic Heat Source: Unlike previous work on constant heat sources, this research captures the time-varying nature of AC arc heat generation, which is crucial for accurate fire risk assessment.
- Optimization is Key: The combination of neural networks and genetic algorithms proved effective in optimizing the segmented EHS model, significantly reducing temperature prediction errors.
- Randomness Matters: The study highlights that the randomness of AC arc generation (both in timing and number) significantly influences cable fire risk, with increased arc activity and current leading to higher probabilities of fire.
Key Results
- The MHD model's AC arc voltage and current outputs closely matched experimental results.
- The optimized EHS model achieved minimum temperature errors of 5.8%, 4.4%, and 4.2% compared to the MHD model for AC arc peak currents of 2 A, 4 A, and 6 A, respectively.
- The study established a method to calculate the probability of PVC cable fires based on random AC arc generation.
- It was found that for a given number of AC half-waves, increasing the number of randomly generated AC arcs and the arc current leads to a higher probability of fire risk (first and second levels).
Limitations and Future Work
- Scope of Arc: The study focuses on AC arcs in short gaps of copper-core cables. Future work could explore longer gaps or different cable types.
- Experimental Data: While experimental validation was performed for arc voltage and current, direct experimental validation of the fire risk assessment itself is not detailed.
- Material Degradation: The model primarily focuses on temperature distribution and fire risk based on temperature thresholds. Further research could incorporate more detailed material degradation and combustion models.
- Real-World Complexity: The random generation of AC arcs is a simplification. Future work could explore more realistic fault scenarios and arc initiation mechanisms.
Summary updated