Energy and Climate Change CommitteeWritten evidence submitted by The Institution of Engineering and Technology

Executive Summary

1. In the course of this inquiry we predict that the Committee will receive a large amount of apparently contradictory evidence because the subject of smart metering and smart grid spans so many disciplines: power system planning; energy retail; IT; telecommunications; meter design and manufacture; regulation; consumer affairs; behavioural analysis and more. Few, if any, individuals or organisations have practical knowledge in more than two of these areas. This is why technical programme management is so crucial.

2. When the Institution of Engineering and Technology (IET) addresses smart metering issues we draw on the expertise of professional engineers and particularly the IET’s three Policy Panels on Energy, IT and Communications. However, we are not qualified to comment on the social and behavioural aspects, highly important as they are.

3. In this evidence we draw attention to some key themes that we believe will be helpful to the Committee.

4. Our key points are:

(a)The roll-out of smart meters is a complex programme with significant risk. The timeline for roll-out should be determined based on what can reasonably be achieved rather than driven to meet political goals without consideration of practicalities. The potential consequences of a poorly conceived or overly rushed roll-out include cost escalation, poor functionality and rejection by consumers.

(b)The costs in the business case need to be tested against best current knowledge; costs are likely to have escalated as the system functionality and architecture has been clarified.

(c)The benefits in the business case need to be tested to ensure they remain deliverable by the roll-out programme.

(d)Even with a well-planned roll-out this is a complex business and behavioural change project supported by significant IT infrastructure. It will be vulnerable to cost overruns, delays and degradation of functionality unless well managed. The Committee may wish to enquire into these aspects, including the proposed management arrangements and key contract terms with suppliers.

(e)The IET is pleased that the meter specification has been developed to include functionality to enable a future smart grid, as that is where the main benefits are likely to lie. Smart grid benefits are not currently included in the business case, which means we do not have an accurate picture of smart metering costs and benefits.

(f)We are aware of significant work on end to end system security, but have not seen details as this is not in the public domain. End to end security is a property of the metering system as a whole and cannot be verified by proving that individual components of the system are secure. We recommend the Committee’s enquiries include security aspects given their vital importance to public confidence and the system’s role as part of the critical national infrastructure.

Smart Meters and Smart Grid

5. The Coalition Statement in May 2010 stated “We will establish a smart grid and roll out smart meters”. While there is much talk about smart grid and smart meters there is still confusion about what these terms actually mean. Smart meters are just that; our existing meters are going to be replaced by new meters that have more functionality and, most importantly, can exchange data with our electricity supplier. The smart grid is a much bigger concept in which smart meters play one part. The IET has consistently argued that the deployment programme for smart meters must be designed and implemented as part of a wider plan for smart energy grid infrastructure as part of a whole system approach to energy supply and demand.

Ability of Meters to Facilitate Time-shifting of Demand

6. Smart meters will support a range of new energy services that are expected to emerge in the coming years, including tariffs to encourage flexible electricity load to be shifted to periods of low demand, reducing the overall costs of the electricity system.

7. Recent studies by Glasgow Media Group and Chatham House have found that the public is not motivated to change behaviour in response to climate change because they hear so much conflicting evidence on the subject. However, when they were introduced to the concept of threats to energy security (which was new to them) this had a much more significant impact on longer term behaviour. This indicates the potential for improved efficiency if customers were more knowledgeable about energy.

8. Many of the benefits will be more easily obtained, or even increased, if in-home energy management systems thrive and this will depend to quite some extent on whether the energy companies embrace the concept of variable tariffs where energy costs more when it is in high demand. Steps need to be taken to ensure that the Government’s laudable aim of reducing the number of consumer tariffs does not stifle development of tariffs which reward customers who are able to shift demand away from peak demand or supply-constrained times.

9. The SMETS 2 consultation shows that DECC have taken the in-home energy management point on board and the so-called Consumer Access Device will provide historical and current consumption data for that meter (or for that consumer on that meter, as the data on the meter is deleted when a consumer moves house). What the householder used the energy for exactly is out of scope for the smart meter but could well be in scope for a home energy management system.

10. The future opportunities for more real-time integration between home/office energy management systems and the local grid in order to manage congestion locally is seriously lacking in the present approach. This could stifle future local community schemes to manage their own network assets. The Committee may wish to inquire as to how competition at the local level will be stimulated and how the use of data in real-time at a network level can be brought forward under the frameworks currently under consideration.

Security and Inter-operability

11. Throughout the programme, the IET has repeatedly stressed that secure operation of individual components of the smart metering system, though important, cannot guarantee system security. End to end system security is critical. A new design element, the Consumer Access Device (CAD) has recently been added to the draft SMETS 2 HAN system and the repercussions of this have not yet been fully worked through. The fact that a significant change to system architecture has been added at this late stage cannot fail to ring alarm bells. The tight time constraints should not be allowed to compromise rigorous end to end security analysis and testing of the resulting system.

12. Also time is needed to develop the UK ZigBee interface for the Home Area Network (HAN). The UK requirements have doubled the length of the ZigBee specification which means that the time taken for testing the interoperability of all components has risen exponentially—now estimated by the Chair of the UK ZigBee Alliance at an additional two years. Meanwhile testing alone is not sufficient as a methodology for achieving end to end security of a large complex ICT enabled system.

13. It is likely that almost all meters will be manufactured outside the UK. While this has implications for balance of payments, it also raises concerns over system security due to embedded software contained in the meters. The Committee may wish to inquire what steps are in place to protect and ensure system security. This aspect may be particularly difficult to assess since much of the hardware is likely to be imported containing embedded software for which the source code is not readily available for inspection and assessment.

Lessons from Previous IT-enabled Business Change Projects

14. A rushed solution to legitimate technical concerns is highly likely to lead to programme failure later down the line. Completion targets should be set according to engineering reality not political deadlines. The key principles derived from study of the causes of major IT-enabled project failure are now well documented and we recommend that DECC studies the report “Engineering Values in IT, a joint report by the Royal Academy of Engineering, The Institution of Engineering and Technology and the British Computer Society”.1

15. Historically projects that involve both new, leading edge technology and operational concepts suffer large overruns on budget, even assuming design specification and objectives are met. This can be especially true for IT based systems where insufficient design verification and testing has been implemented.

16. The Committee may also wish to explore with DECC what steps they are taking to protect consumers against cost overruns by the companies responsible for IT and communications contracts.

Cost and Value for Money

17. It is important to recognise two aspects to the assessment of value for money:

(a)the simple relationship between the costs of smart metering and scope for reducing individual bills, and

(b)the benefit to the UK as a whole of consumers, in due course, being able to time-shift their electricity use to flatten out peaks in demand (or more closely match their demand to the production of electricity from intermittent renewable generation at any given time). This would reduce the amount of new generation that needs to be built, and/or the amount of low efficiency high-carbon peaking plant that would need to be deployed during times of peak demand or low output from renewable generation. It would also reduce the need for costly and disruptive electricity transmission and distribution network reinforcement—involving for example the need for additional overhead lines in rural areas, or street excavations to lay larger capacity cables in urban areas.

18. For a long time DECC had a Smart Metering Programme which recognised cost benefit analysis of smart meters without having a strategy for a smart grid or placing any value on the contribution of smart grid infrastructure investment.

19. In terms of value to the nation, the time of use element of energy consumption, particularly of electricity, should be given greater prominence in public debate. This is because both the cost and the carbon intensity of electricity varies according to the supply and demand at any particular time. Overnight electricity will be plentiful and low carbon coming from nuclear and wind (in most weather conditions). At times of peak daytime demand, or low wind generation, the least environmentally friendly and most costly generation has to be pressed into service. As we move towards meeting our 2020 decarbonisation targets this will be even more the case. The electricity that comes on at the flick of a switch will be more costly to produce and be less environmentally friendly at peak times than it is during periods of low demand and it is in everyone’s interest that those that can defer their use are incentivised to do so, leaving more supply available for those whose need is urgent.

What lessons can be learned from successful smart meter implementation and usage elsewhere in the world?

20. Caution will be required in interpreting responses received to this question as definitions of what a smart meter is and what constitutes success vary markedly around the world.

21. An additional distinction is that in most other countries the metering is provided by the distribution company, (often a vertically integrated energy supply and distribution arrangement) which makes the system required significantly less complex than the retailer-led model adopted in the UK.

22. No other country has attempted to install a system of such scale and complexity involving so many stakeholders as that being planned by the UK. Consequently this initiative will have to address many technical and business change challenges not addressed in other countries.

23. The UK’s decision to make energy retailers responsible for delivery as they have the “relationship” with the customer, now looks less useful due to the extent to which retailers are mistrusted by customers and the importance of encouraging customers to switch supplier. Additional safeguards have been drawn up (The Smart Metering Installation Code of Practice) to prevent suppliers from taking advantage of that relationship for marketing purposes.

Should vulnerable customers and the fuel-poor be first in line for smart meters so they can get the benefits sooner?

24. No. We are of the opinion that the advantages of smart metering to these groups have been over-stated. The fuel-poor would gain far more benefit from a good standard of building insulation and modern programmable thermostats set up correctly.

25. That said, it will be important to ensure that neither vulnerable nor fuel-poor consumers are disadvantaged due to potential technical difficulties in commissioning smart meters in multi-occupancy residential buildings. Early results of trials have shown that high-rise buildings, especially those with communal (typically basement or ground floor) metering positions, will present challenges in terms both of signal penetration and communication with in-home displays. Vulnerable and fuel-poor consumers could be disadvantaged if delays in enabling smart meters in such buildings precluded the possibility of those consumers taking early advantage of new time-of-use tariffs, though we leave comment to other specialists as to whether these groups of consumers would make use of and benefit from such tariffs.

Are consumers’ concerns about privacy and health being addressed adequately?

26. The IET’s Biological Effects Policy Advisory Group chairman, Professor Tony Barker, has advised that consumers’ concern regarding any health issue with the roll-out of gas and electricity smart meter infrastructure is likely to be related to the wireless technology that might be adopted. In the absence of smart meter infrastructure decisions, it is presumed that any wireless communication will be comparable in power, if not lower, than existing technologies, with a low data-rate, and is likely to use existing communications technologies such as WiFi, Bluetooth, mobile phones and others. Any ensuing health issues would therefore be covered by, for example, the latest IET Position Statement on “The Possible Harmful Biological Effects of Low-Level Electromagnetic Fields of Frequencies up to 300 GHz”.2 This concludes that the balance of scientific evidence to date does not indicate that harmful effects occur in humans due to low-level exposure to EMFs.

Will DECC’s current approach to roll-out, including on procurement and establishment of the central Data and Communications Company, deliver an optimal data and communications strategy?

27. The data and communications strategy cannot be described as “optimal” given the complexity inherent in the original programme design.

28. The complexity of the UK programme results from:

(a)a more fragmented energy supply industry than most other developed nations;

(b)the need to include both gas and electricity metering, whereas many other EU countries only need to consider electricity; UK consumers typically heat their homes by gas which is why smart gas meters are important in the UK but not mentioned in the EU targets; and

(c)policy design having placed responsibility for smart metering with retailers rather than with Distribution Network Operators.

29. These inherent policy obstacles to good engineering design either existed or were created before the current Implementation Programme was instigated and before staff with relevant technical experience were recruited, and despite IET engagement setting out the case for a regional (DNO-led) approach. The reasons for the approach adopted was to enable competition in the provision of smart metering by retailers, but this will come at the cost of not adopting engineering best practice.

30. The result is that we now have a centralised Data Communications Company (DCC) responsible for receiving and storing all data and passing it under agreed privacy conditions to:

(a)energy retailers which have a relationship with the customer at any given time (daily data for billing purposes, more granular only if access granted by the consumer);

(b)the distribution network companies which have a statutory obligation to plan and operate their distribution networks efficiently. For electricity distribution network operators in particular, the availability of half-hourly data will be critical to identifying and addressing emerging network constraints. This will become increasingly important as additional demand due to electrification of heat and transport, and the effects of reverse power flows due to growing numbers of solar PV installations, begin to erode existing distribution network capacity headroom; and

(c)new entrants to the smart energy system such as Demand Aggregators, subject to customer wishes.

31. The solutions to this multi-layered communications requirement are quite rightly constrained by:

(a)the requirement for a very high level of security in a system which will form part of the National Critical Infrastructure; and

(b)cost and how this is accounted for, where the benefits accrue and over what period and with what degree of certainty.

32. Smart metering is part, and only part, of a complex system. The Smart Metering Programme has always struggled in articulating its real objectives and benefits. However, the real gain will be seen when a smart grid is deployed in the future, and without a smart grid much of UK energy policy will not be readily deliverable. Flexibility needs to be included in the DCC for future expansion as part of the overall systems approach.

33. Much current debate around smart metering communications concerns how much pre-investment in the wide area network (WAN) should be made to protect future smart grid functionality. It is agreed that oversizing the WAN at this stage is not necessary, but providing a clear future upgrade path for both data handling capacity and latency is critical.

34. It is desirable to equip the distribution network for further liberalisation in the future. However, the ability for new stakeholders at a community level to access real-time data at a feeder level for innovative community-based solutions is highly constrained by the current proposals. Bandwidth and latency requirements mean that the cost of sending both transactional data (required in non-real-time) and near real-time data3 will inevitably be higher than if only transactional data was sent to the DCC. As indicated above, this is clearly demonstrated by not being able to size the DCC WAN for the functionality that will be required in the future. Other options have previously been proposed to circumnavigate this problem both by the ENA and the IET but these have been declined in favour of a single data stream via the DCC.

The need for a Design Authority

35. This is one of the first major inter-disciplinary infrastructure projects to cut across several Government Departments. The need for such cross-cutting technical and programme design will increase.

36. As indicated throughout this document, a whole systems approach requires that a “systems of systems” delivery is in place in order for many of the systems benefits to flow to the correct part of the supply chain. Often, those who invest are not the natural beneficiaries on their part of the fragmented supply chain. New business models need to be developed to allow the new investment to bring forward the joined-up system benefits that will only be available if different parts of the supply chain act in unison.

37. To enable this, multiple departments and regulatory bodies will need to act together to allow the market structure to evolve in a more holistic manner. In the end Government has the overriding oversight of the different market stakeholders and it is therefore important that the technical infrastructure is designed to maximise the commercial flexibility that a whole systems approach can deliver. The IET recommends that Government considers how best to create a technical Design Authority to meet this need.

38. Much of the promise of Smart Grids is not in singular projects (as in smart meter roll-out) but in the way all of these projects are allowed to become part of a wider plan between, for instance, heat, transport and health. This cannot be achieved with the current structures.

Timescale

39. The IET strongly recommends that the start of mass roll-out should be pushed back rather than compromising design issues and security to meet an arbitrary political target. To move forward before having a fully stable and tested design and product could lead to the deployment of sub-optimal equipment, poor customer experience and will risk the success of the entire project.

40. Recruiting and training sufficient installers with the required technical and customer education skills is a significant challenge. There is concern that programme delays will compress the roll-out period which would stress an already challenging timetable by requiring an even greater number of installers over a shorter period. The responsible course of action is to delay the end date of roll-out in keeping with any delay in the start date.

41. EU targets require that member states install smart meters to 80% of electricity consumers by 2020. The earlier completion date in the UK is unjustified. (However, this does highlight one of the differences between the UK and most other EU countries. UK targets rightly need to include gas meters as well as electricity.)

About the IET

42. The Institution of Engineering and Technology (IET) is one of the world’s leading professional bodies for the engineering and technology community and, as a charity, is technically informed but independent of network company, equipment supplier or service provider interests. This submission has been prepared on behalf of the Board of Trustees by the IET’s Energy Policy Panel in collaboration with the IT Policy Panel and the Communications Policy Panel.

February 2013