Case study

PG&E Battles High Impedance Faults.

PG&E is addressing the challenge of detecting dangerous high impedance faults (HIFs) that can lead to wildfires by partnering with Gridware to implement the Active Grid Response (AGR) platform, which enables real-time detection and situational awareness of hazards, exemplified by a recent incident where an oak tree on a power line was swiftly identified and mitigated before it could ignite a fire.

Customer

PG&E

PG&E, like all electric utilities, faces a significant challenge in detecting high impedance faults (HIFs) — a type of abnormal electrical condition where a conductor makes unintended contact with the ground or another object, but does not draw enough current to trip conventional protection devices.

These faults are particularly dangerous because they often do not produce obvious physical signs, can persist undetected for extended periods, and pose serious risks for wildfire ignition, public safety, and equipment damage. Traditional relays and breakers are not designed to detect the subtle and erratic current signatures of HIFs, leaving a critical blind spot in the utility's monitoring and protection systems. 

A branch hits the line after dark, doesn’t trip anything, and no one sees it—but we still found it and shut it down safely. That just wouldn’t have happened a year ago.

Once the power is out, line workers are dispatched with no insight into the hazard’s type, location, or severity. They must patrol miles of power lines, often in dangerous conditions, searching for the issue before action can be taken to restore power. This hazard awareness delay becomes even more consequential in PG&E’s expansive, vegetation-dense service territory, where the early identification of such faults is essential to mitigating wildfire risk and ensuring grid resilience.

Enhancing EPSS with Gridware’s Active Grid Response Platform

In late 2022, PG&E partnered with Gridware to pilot the AGR platform across a handful of high wildfire threat areas to evaluate performance across several use cases:

  1. Detect incipient & high impedance hazards, that have a high probability of eventual ignition, but are otherwise undetectable through conventional protection methods.
  2. Enhance situational awareness by monitoring de-energized lines during storm response operations and PSPS events.
  3. Reduce unknown outages and patrol times by predicting event causes in advance of troubleshooters arriving on scene.

PG&E has begun to evaluate the next deployment areas to enable broad coverage of high-risk circuits to augment existing wildfire mitigation measures ensuring a layered approach to further enhance the safety of their hometowns.  

Real-time Prevention of a Forest Fire

On July 5th, at 8:40 pm, an oak tree began to crack in a remote area of Northern California and fell into an energized 21 kV line.  The branch remained on the line, smoldering, without enhanced protection settings operating. There was a several day stretch over this period of red-flag warnings, including the 5th, with temperatures exceeding 100 degrees and strong winds. In this scenario, because it was dark and the tree did not dramatically displace the line, this would not have been caught by a member of the public, and would’ve continued to slowly smolder on the lines until the branch carbonized and created a low-impedance path to ground. At that point, a larger electrical fault current would flow, and very likely the branch would combust resulting in sparks landing on dry fuel.  With a larger fault current, protection would have operated and then the lines would have de-energized, but not before sparks flew. If AMI existed downstream (unknown), they could have been made aware of the outage, but would only know it occurred on a lateral downstream of a blown fuse.


With Gridware AGR’s multi-modal sensing, the initial impact and line displacement was picked up in real-time, before the tree had a chance to carbonize and result in an electrical fault, outage, and possible wildfire ignition. PG&E was able to dispatch a troublemen to the exact location of issue, where *in the dark*, he was able to quickly spot, de-energize the line, and remove the tree.