Substation Arc Flash


Substation arc flash events are critical failures within the utility industry that can result in significant harm to life and damage to property. An arc flash is the light and heat energy produced from an arc fault, an electrical explosion or discharge that results from a connection through air to ground or another voltage phase in an electrical system. This event within substation operations occurs when a qualified electrical worker makes an error around the presence of energy at a switch or circuit breaker.


A Trajectories Company advisory team partnered with utilities to build a Trajectories® model of switching errors in a substation setting.

The project revolved around developing a predictive risk model for substation switching, which ultimately resulted in identifying 6000 paths of risk within substation switching tasks. Within the model 5 elements were identified that could materialize into significant risk factors when conducting a substation switching task. They are as follows:

  • Proximity – the amount of distance personnel can manage from the hazard while switching.
  • High Intensity Flash – Potential for arc flash to be above 8 cal/cm2
  • Complex vs. Complicated Systems – Complex systems are characterized by interconnectivity and interdependencies, which often results in a higher rate of error (specifically a wrong mental model). In contrast, layers of many simple structures characterize complicated systems that generally have lower error rates attached to the activities within those operations.
  • Equipment condition – the physical condition of equipment and the ability of operators to assess the reliability of the equipment during operations.
  • Exposure rate – number of operational activities that expose personnel to the potential for an arc flash.


As an example, when a qualified electrical worker is conducting a switching task and three out of the five identified elements emerge as risk factors that exist within the switching operation; the risk of an arc flash event is potentially increased by an order of magnitude. The most hazardous combination of factors consisting of a switching task where the acting operator is working in tight quarters/proximity to equipment with high calorie arc flash potential (above 8 cal/cm2), that they are unable to evaluate and believe to be de-energized, due to a wrong mental model or misunderstanding of power flow. This path of risk was our most dangerous combination and most likely to result in harm.


In summary, the Trajectories® model identified all potential errors, mistakes, and other failures at each point in the process which could lead to an adverse event, and then assessed the robustness and efficacy of existing controls to prevent these failures. 


It is important to note that utilities have a highly reliable switching system in place; currently most utilities average approximately 2 events per year. In building the models most of the trajectories were not of much concern, but the highest risk trajectories accounted for nearly all the predicted events in year. What also came out of the analysis was that these high-risk trajectories all centered around technician mistakes.

  • The highest risk trajectories centered around a technician having a wrong mental model regarding their position or position of the device (i.e., switch open or closed) and switching without proper authorization.
  • The other high-risk trajectories involved human error, such as when the switchman moves to open the wrong switch or inadvertently skips a step during switching. These mistakes may happen often in the field, but they can be caught by different defenses, such as switchman checker (if available), three-part communication, basic steps of switching, and distance in cases of remote switching.
  • Inadvertent encroachment was also identified as in area of concern when working on equipment in a confined space that is within a couple feet of the technician. The modeling team accounted for an inadvertent encroachment with either body or tools, where a technician backs up into a switch rack or drops a tool onto a switch. These trajectories are high-risk because there are currently no defenses for this type of incident.


The high-risk trajectories accounted for ~90% of the predicted risk and involve the above stated factors. Each of these high-risk trajectories are either undefended or involve only one or two defenses.


Several possible interventions were reviewed with the modeling team, and only those identified as reasonably applicable were considered and recommended. Most of the recommended interventions apply to different sections of substation switching tasks. Therefore, a combination of these interventions will likely yield the best chance in preventing the negative outcome from occurring.

  • The first recommendation involves reinforcing the requirement to read out loud position verification on device and program in the basic steps of switching to help minimize drift. The analysis indicated this could have a significant impact on wrong mental models. Reading the position on device and program out loud forces the brain to not just read the position but to read and comprehend the position (think of the difference between hearing and listening to comprehend).
  • The second recommendation asks utilities to limit the amount of solo switching in their system and if the switchman is solo, notify the system operator during their three-part communication so the system operator is aware that the switchman is operating in a solo environment and can potentially offer more assistance if needed. The modeling team believes the system operator’s awareness could reduce the active failure rate for switchman position errors due to wrong mental models in solo switching programs. The modeling team understands that solo switching is sometimes required by system conditions and cannot be removed completely.
  • The third recommendation looks to develop a best practice for switchmen to identify the called out stop point within program steps related to permission/authorization. By creating or identifying a stop point on the program during switching we can reduce the rate of switching without authorization, effectively reducing the risk of this trajectory by 50%.


The trajectories analysis predicted if these interventions or versions of them were to be implemented within utility switching systems, it could result in a 67% reduction of total events.

The trajectories analysis predicted if these interventions or versions of them were to be implemented within maintenance aircraft taxi operations, it could result in a 98% reduction in total events.

Leave a Reply

Your email address will not be published. Required fields are marked *