The spirit of this new process will be a transition from central to local, from analog to digital and from big/heavy to small/light systems. A disruptive transformation is mandated because of the ending economical life of electricity infrastructure in developed countries and the need for faster and more economical solutions in developing countries due to being unable to meet the high cost of conventional central approach. Without a doubt, the key to this process will be the word “distributed” which will disrupt all the processes from the structure to the operation of the electricity grid. You will also come across this word quite frequently in this article.
The above image shows the Grid 2.0 structure that awaits us. Distribution grids are evolving from the longstanding structure. This was colossal and central into a new grid structure where stakeholders are more distributed and in constant communication with each other.
Antoine de Saint Exupery, the author of the book called “Little Prince” which most of us would have read in our primary school years, has a great phrase for this situation: “Your responsibility is not to foresee the future, but to enable it.”
We need to talk about and discuss how we can enable and plan for this future. We are heading towards it. In the end, there is no doubt that the water will find its way with the flow but what will it cost and how long will it take?
We are at a crossroads where we better hurry up and focus on adoption/application of technologies (which we’re quite good at as a society in general) and take a route well-planned. That’s why we need to start planning the transformation of distribution grids which has a potential to become a show-stopper in this overall transformation. We need to understand the problems ahead of us by looking at the situation both in detail and with a bird’s eye view to start developing the correct and most economical engineering solutions within a well-defined and planned framework. Hence the main question is: What are the problems on the horizon especially for electricity distribution companies in the distribution grid structure visualized above?
We need to highlight an important aspect here. Although it is true that distributed and flexible systems are marching their way in, we still need a physical infrastructure (which is the distribution grid) for distribution of the generated electrical energy. The main point will be that the limitations, inadequacies or flexibility of this distribution grid will dictate how flexible our end systems will be.
A very simple example would be the technical constraints such as transformer/feeder loading capacity, voltage rise or reactive power flow for exporting to the grid for all rooftop PV systems fed from the same distribution transformer.
The most prevalent distributed generation source is expected to be solar energy because of increasing efficiency through technological developments and ever-declining equipment costs. Even today the least cost of energy resource is identified (US Department of Energy Report) as solar energy in terms of LCOE. So the first expected problems with the increasing share of distributed generation are the constraints due to bi-directional energy flow and technical quality problems in systems designed for a one-directional flow.
On the overall the fact that the energy is going to be consumed at the location it is generated is going to provide a huge improvement in terms of system efficiency. Technical losses usually experienced in the transmission and distribution of centrally generated energy will also be reduced the big time for a specific period of the day. Now the key phrase in the previous sentence is “for a specific period of the day.”
The specific period of the solar energy due to its direct dependence on irradiance of the sun and the increasing prevalence of distributed generation systems has a closing effect of the gap between generation and consumption leading to a change in the conventional load graph towards becoming what we call “duck curve” as it can be seen from the below figure.
As a matter of fact, it is actually even worse for some emerging economies so that it is actually called shark curve which happens due to very sharp increase in the demand because of lower efficiency appliances and lack of demand management policies as explained below:
The change in the curve in this manner has brought about voltage regulation problems in the grids designed for consumption at the endpoint. The distribution transformers are set according to peak hours when there is no or little solar generation coupled with high consumption. However, during the daylight hours, total demand will decrease because of solar energy generation (and in some places energy will start flowing back to the distribution grids) and there will be a voltage rise problem in the overall system. High voltage during sunlight hours and low voltage in the evening will be one of the main challenges that the electricity distribution companies will need to handle.
On the other hand intermittency because of rapid changes in the weather and the concentration of generation facilities in a certain geographical area will make it hard for the generation forecast with high accuracy beforehand. For this very reason, resources will be utilized much more effectively if the new investments or improvements are made taking the solution of this problem into account.
Typical Voltage Profile
There might be some arguments over “Duck Curve” being improved and even made better by storage technologies. And yes, storage is one of the areas that we will definitely touch on but even if we look at Australia which is way far ahead from most countries in this area as can been seen from the following report: Energy Storage
The event there we see that the rooftop PV installations come close to 2M whereas distributed storage is around 33k. So there is a big lag until the storage catches up with distributed generation installation numbers. This phase difference between storage systems and PV installations is highly likely to cause the load profile and voltage regulation problems discussed above.
When the problems such as voltage regulation caused by the high prevalence of distributed generation start posing a serious problem for the electricity distribution system to be operated in a safe and reliable way, the solutions to be applied will also have to be distributed in nature parallel to the problem. At this point what we call Grid 2.0 will embody itself and bring together an architecture where not only energy but also data is flowing in a multi-directional way and existing assets are utilized in a more effective manner and problems are minimized by means of smart systems.
The point to be highlighted here is that the solutions will be distributed as well. In a system like distribution grids where there are many many assets (transformers, line, consumption points) distributed solutions mean new or additional investments in a lot of points. And of course, this requires time, money and effort or resources in short. The best economical way to prevent this is to deploy ICT systems at the highest possible level to enable all existing or new asset to be utilized more effectively. Integrating the distributed PV and storage systems into this equation will enable solutions for distribution grids to two fundamental problems: the necessity of distributed systems and inadequate resources.
Reactive Power Control
Another problem on top of voltage regulation is going to be a reactive power flow and the technical losses stemming from this, causing also the voltage problem indirectly. Even when the active power demand is low, there will still be demand for reactive power (for air conditioners, pumps, etc) and the technical losses stemming from reactive power in the distribution grids can start becoming more than technical losses stemming from active power. Reactive power flow and especially cause efficiency to decline dramatically in point fed by long and small diameter lines. The best method to overcome this would be to meet the reactive power demand in the same or close location.
Reactive power compensation can start becoming a requirement for residential buildings as well as in time. This also means an additional investment for end users. At this point, being able to utilize the existing assets (especially PV systems) will prevent additional (and possibly to be idle in the future) investments. To be able to utilize the rooftop PV systems as a reactive power compensation system will become a solution for both the technical losses due to reactive power flow and additional investment requirement.
To summarize the discussion points above:
Distribution grid systems that have been conventionally designed and operated for one directional power flow are going to have multi-directional load flow and this will bring voltage regulation, reactive power flow and additional technical losses and these are serious problems that can bring the system to an inoperable point.
The systems to solve these problems will also have to be distributed in nature and will bring together a substantial resource requirement. For this very reason, we need to look into systems that utilize the existing assets more effectively by operating them with integrated information technologies. It is very important that the distribution companies to have some control over and manage PV generation, battery, EV charging stations (which are quite ready for IT-OT integration) and also guide the end users in their choices for these systems to prevent costly system augmentation. That’s why we think that distribution companies and prosumers are going to be stakeholders that will be in constant interaction and that they can start taking steps towards enabling this interaction straight away.
Alper Terciyanlı, Ph.D.