A Protection Engineer's Diary: A Date with Disturbances
An ideal working place for any engineer is an error-free environment. We have seen a lot of memes about software engineers and zero compile errors. Seems a far-fetched idea to me!
As an electrical engineer, more importantly, a protection engineer, you are not only expected to safeguard the power system and its equipment but also anticipate events and disturbances which could disrupt the power supply. And all this within a span of a few milliseconds while importing and exporting millions of units of power. This is the reason we have protective devices like your Intelligent Electronic Devices or relays, making the decisions for you. These IEDs recognize patterns in your electrical parameters like current, voltage, frequency and impedances under the system conditions and take decisions according to a set of pre-programmed actions. But how do you, as a simple human, understand what happened to your system and why the IED acted like it did?
Enter the Disturbance Recorders (DR). It is an inbuilt storage within the IED to save the events and disturbances as detected by the IEDs. Most importantly, it is timestamped. Meaning, anyone could paint a picture on the sequence of events that had led to the reaction from the IED. Now, this changes the way we analyze system events completely. The Standard Operating Procedure (SOP) for any event is to download from the Disturbance Recorders (DRs) and the Event Recorders (ERs), sort them by time and see the story unfold before you.
Then, why is it difficult to analyze the system events and identify the root cause analysis? Like I mentioned earlier, there is no ideal workplace, especially for protection engineers with a lot of variables in their system. The CONFIGURATION of the DR plays a particularly significant role in the way they generate outputs and reports. So, in this article, let us take a deep dive and go on a date with the DRs.
DRs can be poorly configured in some of the following ways, and though list is not exhaustive, it covers the most common mistakes by the protection engineer.
No / Wrong Date and Time Setup
The most common reason for difficult and incorrect analysis is the poor settings of the date and time. With time synchronization available in most of the substation and utilities, this may not be a recurring problem, but in case of its unavailability, industries are dependent on using an appropriate time setting and hoping that the internal clock of the IED maintains the accuracy over time. This time setting may also be lost, in case of any internal battery failure and rebooting of the relay. There is no workaround over this, other the using a time synchronization setup within your plant or utility, and the initial setup of all the IEDs are correct.
Improper Analog Channel Selection
Selection of channels also play a key role in the ability for right analysis. In case of basic numerical relays, the number of channels is limited, meaning that the right analog channels need to be configured for us to make the best possible analysis for any possible system event.
There are two main pitfalls associated with trigger assignments - not assigning enough triggers for all system conditions and assigning too many irrelevant triggers. The former setting means that some of the system events are not recorded in the DR, meaning that we cannot perform any root cause analysis. Assigning every trigger, even the unnecessary and continuous signals, means that the DR is either always recording or it is filled with trivial data. This leads to my next point - poor memory allocation. For example, in case pickup signal is given as a trigger for DRs, and there is a lot of known voltage fluctuations in the system, then DR records all the undervoltage and/or overvoltage events, filling up the memory.
Poor Memory Allocation
The memory within the IED is fixed and cannot be expanded. Meaning that all the DRs and ERs, which occupy memory space, are finite in number. The space required for a particular DR is defined by its recording length, pre-fault, and post-fault time. Once the memory of the IED is full, then the newer DRs or ERs are not stored. In case of some IEDs, the older ones are deleted to make space for the new records in a FIFO concept.
No Correct Labels
This is the most common human error, which results in confusion and chaos during analysis. We properly fail to give the correct or easy-to-understand labels for the channel, which causes uncertainty during analysis.
Recommendations and Conclusion
It is highly recommended to read the relay manual before deciding the recording strategy. Some relays may delete old DRs and ERs to make up space for new ones. In this case, it is better to have a schedule for downloading the records at regular intervals. On the other hand, if there is no automatic deletion, we need to plan for regular cleaning and clearing up the IED memory.
Disturbance Records play a leading role in identifying system events and conditions which could be avoided in the future. But with poor configuration and improper strategy, the output data may not serve the purpose.
DISCLAIMER: In this article, the internal disturbance recorders of IEDs have been discussed. The standalone DRs have not been considered.