By Phil Kreveld
It’s not about the interminable ‘when the sun doesn’t shine or the wind doesn’t blow’—our network security challenges are right around the corner.
We are not designing networks specifically for renewable energy sources, with the exception of isolated microgrids. The current renewable energy climate, commercial and political, is clouded over. Already major wind projects are withdrawn for not being economically viable.
Nevertheless, the climate change and energy minister appears to be brimming with confidence—confident that renewable timelines for targets will be met—and that electrical energy tariffs will come down. However, willing it will not make it happen and the renewables transition is the prime problem creator.
Related article: ‘Dunkelflaute’ and weak grids
The critical points:
- electrical energy is vital to the functioning of the economy;
- electrical energy at lowest cost is vital for commercial competitiveness.
The first point relates to grid security; the second, to growth in the national economy. Most necessary engineering effort and expenditure required relates to (1). The second is about underlying economic principles, i.e., a free market with a bunch of players, or a national asset. In one, the consumer pays; in the other, the taxpayer. It is obvious that the renewable transition is responsible for very significant engineering problems. Whether the cost of solutions feed into electricity tariffs is only a matter of accounting. The costs will land somewhere.
The well-established, and the renewable technologies differ in only two material ways. One form of generation relies on the interaction between electrical current and magnetic fields; the new one, almost entirely on just electrical current. The new technology, when adapted to work in traditionally-engineered networks requires significant network engineering effort and expense. We are not designing networks specifically for renewable energy sources, with the exception of isolated microgrids.
The retention of the established technology, i.e., synchronous generation driven by low or even non-CO2 emitting energy sources would not attract the level of re-engineering of the grid needed for renewable energy sources. For example, gas turbines, high-temperature coal-fired boilers, low-temperature differential organic working fluid Rankin cycle turbines, and nuclear energy sources.
The salient issue in network design is the time span available for making network control decisions. These decisions are required when there is a sudden load change, a transmission link is lost, a fault such as a short circuit occurs, or generation capacity is lost. Alternating current (AC) networks are very susceptible to failure if decisions are not made within the time span of a few cycles (20 milliseconds/cycle multiplied by a factor of 3 or so). Inertia can extend the time to a few seconds.
Decisions that might need to be taken include load shedding, opening circuit breakers within the time increment before generators become unstable, switching in alternate transmission links, etc. A major concern is the loss in megawatt-seconds of inertia, to about half of that 15 years ago because of the loss of synchronous capacity. However, the latency permissible in control actions as a function of inertia in MWsec is not linear, i.e., it can suddenly ‘drop off a cliff’. It also depends on topology, i.e., location of generators and their electrical distance from faults, and changes in load consumption, etc. Note, this not just geographical distance but the transmission line impedance properties which affect electrical distance. Thus, various low points in inertia can exist, making these susceptible to voltage collapse.
Related article: Facts, fictions, and fables
The replacement of synchronous capacity by inverters, whether grid following or voltage forming, is affected by lack of real-life operating experience. It is added to by lack of centralised operation of the network. Centralised grid operation is a foreign concept in geographically extensive AC grids, but the change to inverters is bringing its necessity ever closer because of their far more rapid response to power, voltage, frequency and voltage-angle changes compared to synchronous machines.
To stave that off, measures such as the installation of synchronous condensers are required and that appears to be world practice. These can prevent voltage collapse and provide inertia. However, it also slows down the connection of renewable energy sources because of demand for synchronous condensers by many networks around the world.
No attempt is being made for a comprehensive review of network requirements based on inverter-only or mainly inverter-based grids. Such a review would game-play a series of credible and non-credible contingencies at some 3000 busbars and on transmission links, to predict events such as sub-synchronous oscillation damping, and voltage and reactive power variation. That would be the start of closing the gulf between problem creators and problem solvers!
