We are hiringCome build the future of the grid with us.Find your role

Event analysis and reporting

Every relay record in one place

Protection engineers were pulling each relay record by hand from six sources. Now the files land in one list, parsed and screened on arrival, with the report pre-filled.

Renewable owner, operations engineering, ten plants to start
PRC-030-1, PRC-029, PRC-028-1, PRC-004-6

The situation

When a relay tripped at one of this owner's plants, the protection engineer logged into the control network and pulled every relay that had picked up, usually several per event. Each file was opened one at a time. The rest of the picture came from about six places: fault recorder data, relay records, inverter logs and fault codes, SCADA and plant controller logs. The write-up went to a third party. In the engineer's words: "it's a very manual process for us right now."

Under PRC-030-1 the analysis of an unexpected loss of output has a 90-day clock, the corrective action plan a 60-day clock, and the records have to be kept for three years. PRC-029 and PRC-028-1 are judged on the same records.

What was required

Every disturbance record for a plant in one place, each one read for what it contains, screened against the standards that apply, and brought into the report the standard expects, before the clock runs out.

What we did

We turned on the Events page for the owner's plants. Every record for a plant now sits in one list: SCADA day exports, COMTRADE fault records and relay event reports, each with its plant, day, result and review status. Files arrive by upload or by an automated drop into the plant's events folder.

A file is parsed the moment it lands. The trigger instant, the relay that recorded it, the station, the event type, the sample rate, the record length and the time source are read off the file, and the relay is matched to the plant's protection profile. Each record is screened against the ride-through, disturbance-monitoring, misoperation and output-loss standards, and anything that falls short is flagged before anyone opens it.

One click opens the matching workflow with the event data already filled in. The engineer writes the root cause, the findings and the corrective action, and issues the report from the same screen.

What we found

Within a week, the owner's existing relay records were loaded and records from ten plants were run. A single day's SCADA export is 86,400 rows; nobody reads it by hand anymore.

The outcome

The engineer still does the engineering. What is gone is the download, the file juggling, the retyping and the version chasing around it.

Compliance under PRC-029 and PRC-030 is judged on real disturbance records, and the clocks start whether or not anyone asks. One list, screened on arrival, with the report pre-filled, is how an owner proves every event was handled on time.

At a glance

  • 6sources an engineer pulled by hand per event
  • 1list, every record screened on arrival
  • 10plants running in the first week
  • 90 / 60 daysthe analysis and corrective-action clocks it tracks
  • 86,400rows in one day's SCADA export, read automatically

Related

Talk to an engineer

Written from the delivered record.