Metering Communication: Don’t bet on the wrong standards

Examines the role of current interoperability standards and calls into question whether these are truly interoperable. Examining PRIME, G3-PLC, and OSGP and determining whether the interoperability is truly across the entire system, or just within a few reference layers of the whole communication model.

Growing Smart E-Meter Projects: Hype vs. Reality

The number of pilot projects with smart e-meters, where double-sided communication has been used successfully continues to grow. This growing reference information enables us to evaluate both single technologies, and the requirements for more complex solutions. There is no doubt that plenty of nice-looking paper or electronic presentations have been made where the results of these projects has been described.

Customers and final users (utilities) are persuaded that there is no necessity to come up with anything new, because the answer already exists and everything is operating perfectly. In fact, it is hard to receive the real results of realised projects and because of this, it is hard to evaluate them reasonably and in a provable way.

Utility Sector Unprepared for Smart Metering

First of all, you come across to the discovery of zero, or (in better case) very small technical readiness of distribution system operators. The truth is, that development in the power industry was evolving and traditionally, very conservative. The possibilities proposed by automated meter reading were sceptically received by utility technicians when initially put forward.

Today, we could say that technical progress caught them absolutely unprepared from a technical point of view as well as from the readiness for choice from an offered solution point of view. Political decisions that brought modern appliance with all their advantages and disadvantages to households, and requirements for smart meters (with new communication technologies) deployment, in order to solve power balances, have been received as a real shock to power engineers.

It would appear there are no experts in the power sector that could really understand the technical possibilities and who were able to suggest useful solution. In the better case, teams of experts were only being developed.

Managers with a lack of technical knowledge started to use one word that became a mantra for them – “interoperability.” They argued with another sector (in this case telecommunication), where standards are clearly defined and interoperability brings lower prices and vendor independence. However, they do not take into account, what preceded the standardisation found in the telecoms sector.

A Misguided Approach to Standards

In the beginning, there was a proprietary solution that was, first of all, technically tuned, and, subsequently, as a stable solution, released to the market. Let’s wise up the contest of single solutions preceded to final choice of the best one. Moreover, the standard in telecommunication solves everything in all layers of communication model ISO/OSI. Producers have totally clear instructions and customers have exactly described procedures in order to check the solution has been delivered as specified.

Nothing like this has happened in the power sector. We are only witnesses of the remarkable media messages of several chipset producers. Just to clarify: we can talk about interoperable solution only under the condition that single activities in single layers of ISO/OSI model are precisely described. It is too few to describe them by a norm that gives only limit recommendations or describes data exchange in the highest layer (i.e. seventh-application). Unfortunately, such a defined “standard” is not sufficient for full interoperability; it is a halfway solution only.

What is wrong with the Standards?

Nowadays, there are three so-called “standardised” solutions demanded (more likely offered), that should be interoperable, based on the statements of the producers. These are: PRIME, G3-PLC, and OSGP. The first two solutions are based on modern modulation OFDM, and they promise high transmission rate. But wait a second, the enthusiasm calms down immediately after we find out that the DLMS protocol is used at the application layer, which is the same nonsense as a situation where a Formula One Ferrari pulls a 30-ton semitrailer.

PRIME

Datagram is spread among all subcarriers. If a couple of them are missing, the whole datagram is not delivered, and everything must be repeated. If you set the highest rate of modulation, the processor is not able to repair errors (FEC corrector is off), which results into very low communication rates and overall unreliability.

G3-PLC

The lower number of subcarriers leads to slightly more stable system, but proclaimed communication rate is (if you read the documents from the producer carefully) guaranteed in frequency band 10 – 500 kHz. In CENELEC band (10 – 150 kHz), the communication rate only reaches 4.5 kB/s! If you add requirement for DLMS, even the dilettante finds out that this is not the right way. One of the G3-PLC projects reveals a final communication rate of 2.5 B/s (this is not a typing error).

OSGP

This solution does not show a high communication rate at all. It is very stable and the number of installed e-meters is really impressive. This is where the positives, however, end. The application layer contains the interoperable OSGP protocol (and you will receive all possible support to it), but if you want to communicate, there is no other way than to purchase a chipset from the only producer in the world.

All information necessary for communication support is not public, and some of them are protected by patent. If the customer’s requirements on one hand, and producer’s supply on the second hand meet each other, everything is ok. If not, it is bad luck to you.

Summary

If we summarize, we can draw the following (not very encouraging) conclusion: all interoperable systems are based on a single chipset producer. The appearance of higher numbers of producer is illusory only as alternative solutions are based on the same signal processor (uncompetitive from a price point of view).

Moreover, these systems are designed as AMM systems and they are not able to integrate new requirements of Smart Grid (control decentralisation, time response guarantee, significant support of multi-utility solutions, etc.). It is thus necessary to return to the beginning and suggest a solution that will comply with technics.

Binary Data and the Need for Efficiency

Let’s switch on the light to the depth of whole problem and have a look to the following small analysis: at e-meter level, the measured data are stored at the registry that is present in the binary system. For illustration, number 999,999,999 (there is no higher number in e-meter) can be presented by 4 Bytes (i.e. 32 bits).

Code system	Number of Bytes necessary for presentation
Binary System	4
ASCII Code	9
UNICODE	18
Wide UNICODE	36

If we want to see this number in older system ASCII, we will need more than twice that amount of space. If we use wide UNICODE characters, the number of Bytes increases by ten times. The amateur can say this is trifling, but we need to remember that data is transferred in serial mode one bit by one bit!

In the past, it was necessary to make an order in values, so that the power distributors could understand each other. That is why OBIS codes and the COSEM system were suggested. The address space of the OBIS code has a similar structure to an internet address, it is defined at 6 Bytes, which gives address space of 248, (number 281,474,976,710,656). Though, you need to receive at maximum 100 registers from the e-meter!

The Path Forward for Smart Grids

It is clear that the structured approach defined at COSEM and OBIS is very transparent, and it really looks good on paper. If such a system is suggested and used by large consumers, where costs for communication channel construction and operation are negligible, there is no reason to cancel it. However, in a situation of serving millions of places, and with no possibility (from cost, or technical reasons) of using high-speed connections, it is not possible to pretend that it will go all right this way.

Let us define requirements addressing power distributor needs, stressing cyber-security requirements, involving the final customer, and (last but not least) accepting the characteristics of the transmission channel. The golden rule says the system is as strong as its weakest link. And due to the fact that Smart Grids include all power devices, a similar analysis could be used for control distribution network systems as well.

Want to discover our Metering Communication solutions?