Fault Currents from Battery Energy Storage Systems charging vs discharging 1

[rockman7892]

When looking at grid connected Battery Energy Storage Systems (BESS) i'm trying to understand if there are any differences in battery contribution to faults occurring on AC collector system (secondary of GSU)between when batteries are in charging mode vs discharging mode.

I read a paper recently that discussed the level of battery fault contribution to both AC and DC faults to be depended on the battery stage of charge (SOC) with higher charge state producing higher fault contribution than a lower state.

From that I understand fault contribution to AC system from BESS to be depended on battery charge state (100% charge state producing max)but not related at all to weather batteries are in a "charging" vs "discharging" mode. There was a confusing table in the paper which seemed to indicate that during discharging with a fault on the AC system the batteries would drawing power vs when in a charging state they would contribute fault current to AC fault.

Perhaps I mis-interpreted results incorrectly which is why I pose the question that other than stage of charge are there any affects from charging vs discharging cycle that impact fault contribution. Assuming there is not I"m assuming worst case contribution from a modeling case is modeling batteries/inverter as a generator w/ fault contribution based on inverter/battery specs (typically around 1.10 PU).

 

[dpc]

On the ac side, I assume any fault contribution from the dc side has to come via the inverters? The inverters should limit current and shut themselves down within a few cycles to avoid self-destruction in the event of an ac side fault.

 

[JJ_Roy]

As dpc noted, the contribution is largely based on the inverter characteristics. Have a discussion with the BESS manufacturer on how their protections work.

This NERL report also gives a somewhat good insight. Understanding Fault Characteristics of Inverter-Based Distributed Energy Resources

 

[iop95]

When power module (IGBT/SiC or similar) from inverters are gone (in short), battery will be connected to AC for time that fuse clear short-circuit.
Depending of DCbus voltage level, switching/protection equipments capacity at shot-circuit, may apear huge DC short-circuit currents that are very difficult to clear.
Power module short-circuit capability and turn-off are very limited.

 

[EnergyTransfer]

Good question. I didn't see that study.

Large battery systems typical in these systems do have interesting fault condition characteristics.

For the DC side. If a fault occurs on a battery circuit, the current available is extremely high. Whether charge or discharge, or even SOC likely has little effect. That is unless internal cell temperature is high due to this charging or discharging. The internal impedance of cells increase with temperature, which may be the highest impedance in a faulted circuit. I think they are probably considering the wrong issues. Their test may have shown results as described, but without understanding the cause, it can be misleading.

For the AC side. As another mentioned, IGBTs can only deliver so much current. When faulted, they stop. Every industrial inverter I've seen has controls to respond to a fault by stopping IGBT signal input. I usually have more trouble specifying circuit protect to ensure the inverter can clear an overload. A high quality inverter can only source 150% of rated output for a few minutes. Faults would need a shunt trip wired from the inverter to the breaker or it'll just remain closed and the inverter remains off until reset. So the state of charge on the battery should not matter. But a lower quality inverter may have trouble converting as battery voltage during discharge is reduced. And the battery internal resistance will increase, making it harder for the cells to source demand current.

On my designs I have all these issues considered. I use worst case scenario power delivery to dimension and specify all components. This includes clearing capacity of all protectors to cover any overloads. It is not so simple a task. I now always include Emergency Power Disconnects to open all in and out circuits for inverters if inverter is melting metal. And I include Emergency Power Disconnects to open Both the battery side and the charger side of every battery system, to fully isolate these bi-directional conductors. These are located insight of the systems but outside the danger area. I also design every system as fully isolated, A and B systems with 100% capacity on each side. If you read between the lines, you may catch on to why my designs are NOW so conservative. There is something really special about a high frequency, high voltage, arc fault that cannot clear the fuse, while melting the steel chassis into a puddle. Never count on any protector on an inverter.

For even more fun, did you know fire extinguishers do not extinguish these? It's a bit like spitting at the sun. Even the super bad handheld from the firetruck blew it out for a quick pause but nope, it only looked like it paused. The only answer is to disconnect power.

I got to learn that a fire extinguisher is only effective on FIREs, defined by has a flame which sustains and spreads heat for combustion. An electrical fault without flames is not a fire. It might start or feed a fire, but it is not a fire.