Flexible Renewables and Their Implications towards System Reliability

2018 was definitely the year of renewables triumphing totally over conventional energy sources. Except for the ridiculous energy politics going on on my home island, everywhere else seems to realize what the right track is.

In Germany where I am currently living, while coal lobbyists are still trying to keep the industry alive as long as possible, renewables again hit record high electricity production this year; more than 40% of the public used electricity now comes from renewables, and for the first time ever renewables beat the entire coal power plant fleet in terms of total energy production.

As IEA puts it, Germany is not alone in facing a rapid increase of wind and solar power. By 2023, the agency forecasts that 33 countries will have up to 20% variable renewables in their power systems, up from 12 countries in 2017.

As we have been saying over and over again here, what this transition means to the power system is a total paradigm shift towards flexibility. However, the time has come to move a step further for the transition advanced regions; for them, the state of art discussion is how to harvest the flexibility potential of variables renewables.

A few months ago I have already mentioned slightly what this discussion is about. Since then, concrete researches were conducted, and more reports and policy advices were released. Needless to say, this is a very dynamic and young field of study, with many new opportunities ahead.

Today we will demonstrate one aspect of why flexible renewables is becoming more of an attracting concept; that is, its potential contribution to more value of providing system reliability.

System reliability sounds vague, and it is truly being misused or misunderstood very often. Here, loss of load expectation (LOLE) would be the kind of indicator we would use to demonstrate the performance of reliability.

As more variable renewables get onto the grid, LOLE resulting from inadequate generation (our conventional imagination of how a load shedding or blackout event would happen) would actually decrease; on the other hand, LOLE resulting from insufficient flexibility of the power system would increase.

In the study of CES-21, higher renewable penetration yields higher LOLE due to insufficient flexibility but lower LOLE in total.

That we already learnt in previous studies. But then one might wonder, why don’t we just replace all the conventional energy sources into more flexible alternatives? It turns out to have to do with the must run power of a power system, the reserve capacity that is required to keep power systems stable should contingency events take place.

Traditionally, this system must run is provided by conventional power plants, because the market involved with it sets many constraints that restricts VRE from participation. Under this dispatch structure, there would exist a limit for which VRE penetration can reached without curtailments or storage.

Since VRE plus storage are (not yet) viable in a very large scale, it follows that curtailment would be the better option if it is the VRE operators who are asked to provide power system flexibility. However, if we are to curtail the renewable plants anyway, why not let them provide some of the control reserve? VRE with wuch dispatch scheme are considered flexible renewables.

This supply curve shows the order of cost effectiveness for each flexibility measures; in almost all of the literature, storage systems will always be in the rightmost side of the curve.

This would give to a seemingly counterintuitive result: with curtailment planned beforehand, flexible renewables will allow higher VRE penetration rates at a lower cost.
(Or from another perspective, flexible renewables ensure a higher value of the VRE fleet in the system.)

This is mostly because that by providing some of the must run capacity with flexible renewables, the needed conventional power capacity to be on-line and stand by during times of excess VRE will be reduced. Then in these time intervals the system can integrate more VRE than the conventional dispatch mode.

In the study of Energy and Environmental Economics, Inc., it is shown that dispatch solar panels with their full flexible potentials can actually be more cost effective and reduce curtailment.

Also, because of the deployment of flexible renewables, the ramping requirements for conventional power plants will also be smoothed. This will decrease the loss of load expectation due to insufficient flexibility. Since insufficient flexibility will become the main reason for LOLE, deploying flexible renewables will actually improve system reliability more than dispatching renewables as the current way.

As a side effect, flexible renewables will also make inflexible conventional power plants more obsolete; the simulation on a local system in Tampa perfectly captures this well by showing how less the share of coal will become when the entire solar fleet turns fully flexible.

When solar are dispatched with full flexibility, the capacity reserve market will have less demand, meaning that fewer flexible conventional power plants will be on line; this will reduce the need for inflexible power plants that operates as “baseload”.

Flexible renewables are no wonders of the future; the technology and know-how already exist, and were tested at field. Of course, to make it a reality, the electricity market must be reformed such a way that this type of dispatch becomes economically more feasible.

This is why, in the new report of Wind Solar Alliance, it was suggested that the following measures, among many others, should be implemented for a market that encourages flexibility operations:

1. Renewables should be able to participate in all kinds of ancillary services, which is the key for them to contribute to the system must run.

2. Capacity value calculations should be based on rigid studies of effective load carrying capability; this is the kind of study to find out how much ideal capacity does a system need, if the power generation source of concern is removed, such that the LOLE of a year remains the same. In this sense, many grid operators assume less capacity value to VRE than they deserve; while some operators assume the capacity value of wind to be less than 10%, 15–20% might be a more accurate estimation.

3. Meanwhile, traditionally the capacity value of conventional power plants may not take into account correlated forced outages due to weather events. For example, a typhoon might cause all the nuclear reactors in the region near its route to ramp or shut down.

As more and more renewables are getting onto the grid, the idea of flexible renewables will become more important. That said, in most regions renewable energy sources are still in their cradle, so enhancing the flexibility of conventional power plants is still a priority for them. When and how flexible renewables should gradually replace some of the ancillary services of conventional power plants, is therefore a study of policy cost effectiveness and awaits further investigation.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store