Power system faults – traditional generation compared to Inverter connected plant

Introduction

This is the third article in a series which is exploring various power system issues that are being experienced as we transition from a network based on rotating machines to one that is more dependent on power electronics. Each article is self-contained and hence they can be read out of order but it will be also informative to also check out the other articles. ?The first two articles can be linked to here and here .

This article will explore how traditional generation and inverter connected plant responds to external power system faults. This is key to understanding how we can build networks which rely heavily on inverters.?

It would be fair to say that many in the industry have concerns about how the new technologies behave in response to faults – to the extent that some are advocating we keep or add spinning machines (in the form of synchronous condensers or constraint equations mandating connected generation) on the system.?Depending on system circumstances, this approach may indeed be necessary, however the reader should be aware that the authors have been somewhat disillusioned by many decisions taken to install/favor synchronous plant that has taken place in recent history.?It appears (to the authors) these decisions have been based on a prejudice favoring traditional technology rather than the considered technological agnostic approach that should be applied.?

This article will cover a fair bit of ground so below we have mapped out a prelude of the main points to be covered:

• ?What is a power system fault?
• How does traditional generation behave in response to a power system fault?
• How do synchronous condensers behave in response to a power system fault?
• How does wind/solar or battery storage inverter connected plant behave in response to a power system fault?
• Tabulation of the differences
• Is there a difference in the response of grid forming and grid following plant?
• Summary

What is a power system fault?

For the purposes of this article, we will consider the following to be power system faults:

• Three phase short circuit – whereby somewhere on the power system remote from the generator we are studying – the three line (phases) conductors are electrically connected to each other.
• Phase to phase short circuit – as above, except only two phases are electrically in contact with each other.
• Phase to earth short circuit – where one phase is in contact with the earth.

There are other more unusual faults that can occur, such as open circuit faults, inter-winding faults, unbalance conditions and phase to phase to earth faults – but these will not be discussed in detail in this article. ?In fact, for reasons of brevity, most of the discussion will focus just on three phase faults which are the easiest to analyse and describe.

Whenever a system fault (3-phase, phase to phase or phase to earth) occurs, there is a voltage collapse on the faulted phases at the location of the fault.??Generation on the power system responds by injecting more current into the power system in an attempt to maintain system voltage.?At this point, it should be noted that the behaviour of inverter plant and traditional generation differs – so each will be discussed separately.?

Traditional generation response to a fault

The typical sequence of events for a power system fault on a transmission system is as follows:

A fault occurs on a transmission line near the generation plant– this could be a flashover due to a lightning strike, or a tower collapse, or some other misadventure.?The fault causes the voltage on the power system to collapse on the transmission system near the fault location.

The collapse in voltage means the generator can no longer supply power to the transmission system. In AC power systems, Power = Voltage x Current x power factor, if the voltage is very low, the power that can be delivered is similarly very low, even if the fault current that flows from the generator is very large.

The turbine control systems (which are typically electro-mechanical design) are unable to respond instantaneously, so the mechanical power delivered by the turbine to the generator does not change significantly. There is a surplus of power generated compared to the power that can be delivered to the system which has to go somewhere – it is used to speed up the generator.

While the generator speed is increased, the overall speed of the power system (which is effectively the same as its frequency) remains constant. Accordingly the generator will get further and further out of synchronism with the power system while the fault is active.

If the fault is cleared quickly enough (by disconnecting the faulted transmission element from the power system using circuit breakers), then the generator will be pulled back into synchronism with the power system. It will swing backwards and forwards against the system a few times like a pendulum – before settling back into a stable operating state.

If the fault is not cleared quickly enough, the generator will “pole slip” against the power system and experience high mechanical stresses which could cause catastrophic damage to the machine.?Protection systems should trip the machine before this happens.?Typically the time of fault clearance has to be very fast ~ 0.1 – 0.25 seconds to avoid pole slipping.

The ability of a power system to maintain synchronism after a large disturbance (such as a fault) is referred to as “transient stability”.?

Synchronous condenser response to a fault

Synchronous condensers are effectively traditional generators except without any mechanical power input.?They spin using power sourced from the power system, which is typically a very small proportion of their rating (i.e. 1-2%).

Because there is very little power imbalance during a fault – there is little acceleration which means the risk of pole slipping is reduced.

However, like a generator, on fault clearance the synchronous condenser will swing backwards and forwards against the system a few times like a pendulum – before settling back into a stable operating state.?

As with generators, this latter behaviour can excite oscillatory modes in the power system, which can cause power quality issues and in worst case scenarios lead to oscillatory instability which could ultimately cause trips or equipment damage.?

Inverter connected plant response to a fault

The behavior of inverters in response to system faults can be quite different.?As mentioned in a previous article

• the biggest advantage of inverters is that they can do virtually anything,
• the biggest disadvantage of inverters is that they can do virtually anything.

The following descriptions is what most inverter plant currently does in response to a fault, however the reader should keep in mind that changes to inverter software and settings may significantly change its response to power system faults.?This is still an area of active research.

As for the traditional generation case described above, a fault occurs on a transmission line near the inverter plant– this could be a flashover due to a lightning strike, or a tower collapse, or some other misadventure.?The fault causes the voltage on the power system to collapse on the transmission system near the fault location.

The collapse in voltage means the inverter can no longer supply power to the transmission system. In AC power systems, Power = Voltage x Current x power factor, if the voltage is very low, the power that can be delivered is similarly very low, even if the fault current that flows from the inverter is increased.

The inverter control systems (which are of power electronic design) respond virtually instantaneously, so that the power delivered by the inverter is cut off. Instead the controls switch to providing reactive current to the power system in an attempt to support the system voltage as much as possible.

On fault clearance the voltage recovers, and the inverters switch back to power delivery mode.

This switching between different control modes can sometimes cause stability issues with inverters. Specifically, if the control set point dead bands are not set with enough margin, it is possible for the inverter to oscillate backwards and forwards between the different states – causing power quality issues similar to (but usually higher frequency range) to the oscillatory stability issues that rotating machines can sometimes cause.?

Fault current transients

So far we haven’t mentioned how fault current differs between traditional generation and inverter connected plant.?This will be addressed below:

Traditional generation produces fault current magnitude up to about 5-6 times its rating – although if the impedance of the step down generator transformer is also considered, in practice it is usually only about 3 times the rating of the plant.?This fault current output is not sustained, it is controlled by the sub-transient and transient time constants of the machine, which are typically 0.1 – 0.3 seconds (for sub-transient), 2 – 10 seconds (for transient). The effect this has on the fault current output is to cause the generator to initially produce a large fault current which then decays in accordance with its time constants.?Usually the final fault current that a generator can produce is less than its rated load current, but it takes several seconds for the current to decay to this value.

In practice the fault will be cleared within less than a fraction of a second, so the initial and transient fault current is usually the most important consideration.?Sustained fault current can be important to calculate if slow protection systems based on overcurrent relays are used.

The figure below shows that in addition to a decaying amplitude, the fault current produced by a generator is affected by the point on wave that the fault occurs at. DC currents and waveform asymmetry will occur in many cases. ?Magnetic saturation effects in the machine can make calculation of the fault currents complex.?

Sustained fault current can damage generators – but fortunately this is a rare occurrence – usually the fault is cleared by protection systems long before thermal or magnetic forces damages the machine.

Inverter connected plant will produce fault currents in the range of about 1.2 – 1.3 times the inverter current rating.?The main limitation is the current ratings of the power electronics.?The output current must be restricted to avoid damage to the plant.?Fortunately this can be achieved by the power electronics control.

Unlike traditional generation, inverter connected plant can sustain fault current output at a constant value effectively indefinitely.?The fault current is always virtually symmetrical and does not contain decaying DC components.?This predictability makes it easier to design protection systems.

The smaller fault current produced by inverter connected plant compared to traditional generation is often thought of as a serious problem for the energy transition.?This is not the case as will be reasoned below:

• Transmission protection does not require large fault currents to operate reliably because it is usually based on distance relaying principals.
• ?In order to provide the same energy that a fossil fleet of generation produces, it will be required to install approximately three times the nameplate capacity in renewables.?Three times the number of power sources which provide about 1/3 of the fault current per unit tends to result in about the same fault level as the system based on traditional generation has.

Recently the NEM has legislated (what we believe to be poorly considered) rule changes aimed at charging renewable generation for "consuming system strength". This is impossible as was pointed out in a recent LinkedIn article which can be read here .?

Summary of the key differences

The table below summarizes the discussion above:

Grid Forming Inverters

As indicated above:

• the biggest advantage of inverters is that they can do virtually anything,
• the biggest disadvantage of inverters is that they can do virtually anything.

Changing the software control of the inverters can enable them to mimic the behaviour of traditional generation to a greater or lesser extent, so that the table of characteristics shown above may need modifying.?E.g. power swings become a possibility with grid forming plant, as can acceleration during a fault if the software is so designed.?Whilst grid forming has many desirable properties, mimicking the disadvantages of traditional generation does not seem to be a good idea.?We have an opportunity to design the control software of inverters to provide much better performance than we currently have. ?It is overly simplistic to prefer grid forming over grid following control strategies, in different scenarios each mode of control has advantages and disadvantages.

Summary

The behaviour of inverter connected plant compared to traditional generation during and after power system faults has been described.

Traditional generation has transient stability concerns, can pole slip and can participate in low frequency power oscillations. ?It can also produce significant fault current, albeit not entirely predictable.

Inverter connected plant can produce high frequency oscillations and can only supply limited amounts of fault current to the power system.

So long as the characteristics of each form of generation are taken into account in the power system design, reliable power systems can be designed, built and operated successfully. ?