SMIGHT

It is difficult to say exactly which is ‘better’. As with so many dichotomies it comes down to where you place the emphasis. Smart Meters are certainly better at getting highly granular data but tend to be hindered by data privacy rules and the ability to process and use such high volumes of data. Substation monitoring on the other hand is much easier to roll out but doesn’t offer the same level of granularity. So perhaps is it not a question of either/or, it is more of a question of how they can be used together to supplement each other.

Define better. As you might expect there are competing requirements and there isn’t a simple answer. With unlimited resources one could say that a complete rollout of smart meters would generally be better as the higher granularity of data leads to a much ‘shaper’ image of what’s happening in the network. However, there are some limitations to bear in mind. First is about data, how it is communicated, and what can be done with it. The key points being:

• What is being measured?
• To what level of accuracy?
• How quickly is the data being communicated? i.e. how ‘live’ are the data.
• Who has access to the data and in what form?
• How much location data is combined with the measurements?

Then there’s the question of how much does it all cost. A smart meter costs ca 150€, then the installation costs say 50€. So, let’s multiply this out for a typical residential area served by a 6-feeder secondary substation. Let’s say there’s 150 households, that would cost ca €30,000 to deploy smart meters. To deploy substation level monitoring on all the feeders would be around 6-7 times less expensive.

Our view at SMIGHT is that in an ideal roll out smart meters are the best solution to the problem of bringing transparency into the low voltage grid. However, we observe that this ‘ideal roll out’ hardly ever happens. Sometimes the data is transmitted too late to be useful, sometimes the necessary data isn’t transmitted for it to be useable, sometimes the roll out level isn’t sufficient for it to be a complete solution, and so on and so forth.

On top of this, one must bear in mind that smart meter data by itself isn’t useful (for the purposes of network monitoring and control), it needs to be fed into a digital network model ‘a digital twin’. In and of itself, this is a very useful tool. But once again the theory meets reality. For a digital twin to work the data needs to exist and be accurate. Currently, from what we see in Germany at least, this is often not the case. So, DSOs need to both invest in the digital twin software and conduct a considerable amount of work to get the data into a useable state – whilst at the same time hoping that smart meters will be rolled out rapidly and the data will be suitable. From our point of view this is quite a few aspects that need to come together for this model to reach its potential. At SMIGHT we prefer to take a somewhat more pragmatic approach. We believe that DSOs can achieve 80% of the value of such a setup with only 20% of the complexity and cost. We believe that moving up the chain to the level of the distribution cabinet and secondary substation delivers all the transparency that required in a far easier and manageable way. One major point is also that the DSO is in control of the process. The substations and distribution cabinets belong to the DSO so it has complete control over what it does whereas smart meters require interaction within the property of end consumers.

At the end of the day, smart meters have benefit beyond grid operation such as enabling dynamic tariffs and helping end users to manage their own electricity usage and therefore cost. Plus, there is a very strong mandate in almost all countries to roll out smart meters. So, it isn’t really a question of whether we should adopt one option or the other. It is more of a question of whether smart meters are sufficient for low voltage grid monitoring and control? As stated above if they are used to feed an accurate digital twin, then the answer is yes. But if that digital twin doesn’t exist or is not very accurate then it makes sense to complement the smart meters with LV monitoring devices.