We perform high-energy arcing fault, or HEAF, fire modeling using a computational fluid dynamics (CFD)-based tool to determine the zone of influence of a HEAF for different types of high-energy electrical cabinets. HEAFs can pose a fire risk to nuclear power plants.
For questions about this testing, please contact Debashis Basu at +1 210 522 8333.
A HEAF can occur in an electrical system or component through an arc path to ground or lower voltage, if sufficiently high voltage is present at a conductor with the capacity to release a high amount of energy in an extremely short time. Normally, HEAFs are caused by a failure within an electrical installation, such as an electrical switch, circuit, junction, or distributed panel, which may result from a defect, an exceptional service condition, or faulty operation. A HEAF event initiates with an arc that heats and ionizes the surrounding gases at several thousand degrees Celsius, making it electrically conductive, and the temperature increases further as a result of self-heating induced by the current. In an electrical installation, the energy input from the arc to the air leads to a sudden pressure rise followed by a release of hot gases, creating a hazard for the surrounding equipment and nearby operators. Simulation of the HEAF phenomenon involves modeling the plasma-state gas as well as thermofluid analysis. The plasma-state gas is affected by the electromagnetic force induced by the current, the plasma-state gas pressure gradient, and the rate of heat generation. The electromagnetic field significantly influences the overall flow and the thermal field in the computational domain.
With faster processors and better numerical techniques, computational fluid dynamics (CFD) tools have revolutionized engineering design and optimization, limiting expensive experimentation and providing virtual solutions with short turnaround times. Today, CFD is used extensively to analyze a wide variety of applications, from aircraft wing design to sportswear manufacturing.
Our integrated multidisciplinary approach incorporates code customization, analytical model development and applications, and experimental investigation to accurately and effectively solve complex fluid flow and heat transfer problems, including:
- Fire dynamics simulation
- Conjugate heat and flow analysis with multimode heat transfer and phase change
- Complex turbulent unsteady flow and acoustic analysis
- Integrated flow and thermal analysis of engineered systems with interface modeling
- Mesh-free, particle-based computing and smoothed particle hydrodynamics
- Particulate and droplet flow simulation
- Free surface flow and fluid-structure interaction
- Advanced turbulence and heat transfer modeling in fire dynamics simulations
Engineers also have expertise in CFD code modification, algorithm development, and experimental benchmarking to address specific client needs. They have provided extensive CFD research and technical assistance to a variety of programs. The broad spectrum of staff experience is applied to:
- Numerical simulations of fire and vapor suppression devices
- Multiphase flow with phase changes in fire modeling
- Fluid flow and transport analysis of pressure vessels
- Chemical process streams
- Fire dynamic analysis of onshore and offshore structures
- Flow and thermal study of mixing chambers
- Fire dynamics simulations for LNG tanks and pressure vessels
Our Fire Technology Department has computational facilities and resources, including ANSYS-FLUENT that is available with multiple serial and parallel licenses ensuring that multiple simulations can be carried out simultaneously. ANSYS-FLUENT is available on Dell PowerEdge M820 Red Hat Linux-based cluster boxes with 160 computer nodes. In addition, we have a custom-built stand-alone cluster with 96 processors and 512 GB memory. Dedicated workstations for performing pre- and post-processing of models are also available. Numerical modeling capabilities include both FDS and CFAST, as well as a variety of open source software, including OpenFOAM and code SATURNE. FDS is available on the PowerEdge M820 Red Hat Linux-based cluster with 160 computer nodes, and also on the custom-built stand-alone cluster with 96 processors. pre-processing tools are available to generate the geometrical model and computational grid as well as post-processing tools such as TECPLOT and MATLAB to visualize the results.