Date of Award

Fall 11-18-2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Engineering Physics

First Advisor

Neven Simicevic

Abstract

This research was motivated by the overly simplified claims, from a previous study, that attempted to explain the cause of the reportedly repeated lightning strikes at certain locations in the chemical manufacturing plant. The findings, without any model, claimed that the electronic monitoring and control devices’ fire damage occurred, due to the raised electrical ground potential during lightning strikes. It proceeded that the raised electrical ground potential caused a power flow reversal, which exceeded the ratings of the affected electronic control modules, damaging them. This claim is inconsistent with published research work in the failure mechanisms of industrial control systems, caused by lightning strike induced effects such as electric and magnetic fields. The critical questions to be answered that may confirm or refute this claim are: Given the well-researched and understood lightning related ground potential rise, as well as electronic process control modules failure mechanisms, would a rise in electric ground potential cause damages to a power module or the device input/output module? If all the failed electronic modules are certified to be identically rated, why do only a few, not all, fail when reverse biased? Why do these electronic failures not occur during every lightning strike, given the high lightning strike frequency of the chemical manufacturing plant’s location on the keraunic map. The overall objective of this research project was to use a detailed physics-based model, and carry out numerical modeling of lightning strikes at the chemical plant, by incorporating the actual geometrical structures and conditions. Firstly, the stepped leader, which is propagated at a slower speed of about 1 to 2 x 105 m/s, relative to the speed of light (3 x 108 m/s), was formulated using the Poisson equation to describe the local electric and magnetic fields. Secondly, the Maxwell’s time dependent equations were used to describe the fast propagated (1 x 108 m/s), ground originated, cloud-oriented return stroke. The third part of the study analyzed the model response to the inputs that include the geometry of the plant’s structures, as well as the existing lightning protection system. Model simulation results were correlated with the electronic controllers’ failure mechanisms that better explained the observed failures during lightning strikes in certain locations of the chemical manufacturing plant. The Finite Difference Time Domain (FDTD) model simulation to study the possible effects of the return stroke on the chemical plant’s ground potential, indicated that, the damages due to the raised electrical ground potential during lightning strikes’ attachment to the ground or grounded structure in the area is less or comparable to the damage that is caused by induced currents. Because of the long vertical and horizontal runs of the instrument loops that connect the instrument sensors with the I/O cards of the electronic controllers, the circuits are more vulnerable to high voltages resulting from the coupling of lightning induced electromagnetic fields than they are to the raised ground potential. The surge protectors on the input sensor circuits of the electronic control modules were verified to be absent, confirming that the simulated induced voltages, due to lightning strike induced electromagnetic field interferences, in the sensor circuits, will far exceed the typical rating of 5 V, resulting in a probable I/O module failure. The I/O module happened to be the commonly failed module in the affected electronic process controllers.

Download

Share

COinS