Getting smart with energy

Hot, sleepy and remote it may be, but the heel of Italy’s boot is near the forefront of the move to smart grids that is revolutionising energy and utilities in cities and regions across the world.

The Apulia region’s partially EU-funded Puglia Active Network project exemplifies many of the new approaches to energy that smart grids are enabling. It has already installed 1200 of a planned 8000 new electricity substations, with real-time monitoring to minimise downtime and better plan maintenance; and it aims to put more than 3 gigawatts of renewable energy into the grid (that’s roughly a third of London’s peak-time usage).

It is giving smart meters to 30,000 customers, who will be able to reduce wastage by accessing up-to-the-moment information on their energy consumption; it is even installing more than 250 charging locations for electric vehicles.

The key to the Apulian venture – as to all smart grid deployments – is that it’s not about a single technology, or even a single deliverable. Rather, it’s an approach to the smarter, more sustainable use of resources in cities (and sometimes rural areas too) on the principle that sustainability is essential to urban success, and energy is an essential part of sustainability. This involves multiple different technologies, as well as education of consumers, businesses and government.

The application of the smart grid is not limited to electricity, either, although that’s what we’re focusing on here. For example, in the southern California city of Glendale, the Glendale Water and Power utility has employed Itron technology to provide 33,000 smart water meters as well as 83,000 smart electric meters to end users.

The Glendale smart grid includes in-home displays giving residents information and control over their electricity and water usage and costs, as well as automation within the distribution network to match supply to demand. A similar project – also using Itron equipment – in China’s Tianjin Eco-City covers water, heat and gas.

But whichever specific utilities are involved, the goals and principles are the same. A smart grid can cut costs by improving the efficiency of utility suppliers through real-time information, analysis of the flow of power, fault detection, and self-healing, as well as making customers’ own usage more efficient and therefore cheaper. (Indeed, a poll of energy professionals by the tech supplier Opower found that they saw customer-driven value-added services as by far the biggest benefit.)

The smart grid can also help achieve environmental goals (regulatory or otherwise) by reducing customer demand, while projects frequently involve the addition of renewable energy sources and customer-generated power too.

Bottom-line savings as well as environmental benefits can be huge and swift. The first-ever large-scale smart grid deployment to homes, for example, was Telegstore from the Italian utility Enel, rolled out in 2005 and delivering annual savings of €500 million from an original project cost of €2.1 billion. “Smart grids could be Europe’s shale,” the European Commission vice-president for energy union Maroš Šefčovič has said. And worldwide, Juniper Research expects that smart grids will deliver $18.8 billion in cost savings in 2021.

Stakeholders

To maximise these benefits, the smart grid needs to involve at least three main groups: the suppliers, the consumers (be they businesses or households), and the actual devices that use electricity. Existing infrastructure is improved, digital intelligence is added, information is disseminated, and business models may be revamped.

A wide-ranging project conducted by Commonwealth Edison (ComEd) in Illinois illustrates some of the ways in which the smart grid philosophy can affect all the stakeholders, and the sheer scale of deployment that’s needed to give appreciable results in a densely populated area.

ComEd’s goals included minimising the impact of power cuts through cutting their frequency by 20%, and their duration by 15%: an objective which, although it superficially only concerns the supply side, also involves consumers and their devices because it can be excessive demand from these which causes blackouts. The company also wanted to enhance customer satisfaction by giving its customers more information and better service.

The technology employed, too, spanned the supply and demand sides of the grid. By the end of last year, ComEd had installed 1.8 million smart meters at the premises of its Illinois customers, with every one of the 3.8 million total expected to receive them by 2018. It has also begun a five-year plan for the automation of electricity distribution, including the automatic detection of faults, re-routing of power around them, and swift dispatch of maintenance crews to fault locations – all in the interest of those projected downtime reductions.

Lightbulb moment

Targeting power-consuming devices as well, ComEd started a pilot of smart street lighting. This is a favoured choice for many utilities and local governments embracing the smart grid because scale can be achieved relatively quickly and easily through smartening the street lighting operated by a single body – much more simply than convincing thousands of consumers to install smart white goods in their homes, for example.

The ComEd lighting pilot involves 750 fixtures in Lombard and Bensenville. Already using energy-efficient LED lamps, they should have their consumption trimmed still further by the addition of functions such as remote control and scheduling, while failure and maintenance alerts improve service. It’s hoped that total operations, maintenance and energy costs for the 750 street lights can be reduced by as much as 65%.

Street lighting fixtures themselves can also be used to support other elements of a smart city. GridComm, for example, connects Internet of Things sensors to lights to measure weather, pollution and traffic within a city, transmitting data over the lighting control network.

In ComEd’s case, as well, the lighting project will heavily depend on communications technology – in this case based on the wireless platform from Silver Spring Networks – which can form the basis of more ambitious smart city apps. Information underpins the smart grid as it does every aspect of the smart city, so rapid, inexpensive communication is crucial.

For instance, a hypothetical smart grid system could involve factories telling the intelligence at the core of the grid that they are going to hike their power consumption in the next few minutes; and the grid then assessing the local temperature before suggesting to heating or air conditioning systems in other premises that they could turn themselves down, reducing demand to compensate for the factories’ increase. While refrigerators that proactively adjust their own power requirements are certainly better than nothing, this kind of linkage among multiple consumers and producers of power clearly takes intelligence to a new level.

Power operators have a potential advantage here in that Internet of Things connectivity can be provided over the power lines themselves, for example to enable monitoring and control of electricity usage. Widening this technology to provide full Internet connectivity (known as BPL for “broadband over power line”) remains a largely experimental hope thanks to technical challenges, but the lower-bandwidth requirements of passing smart grid information back and forth seem to be more achievable although they are still not mainstream.

Even at the relatively straightforward level of empowering consumers to manage their own consumption, communications technology is all-important. Smart meters (often known as AMI, for “advanced metering infrastructure”) typically feed data back to the systems that control them at least every hour and sometimes more frequently.

One company, Bidgely, is taking this even further with a customer-facing application called HomeBeat that demonstrates how communications can transform power consumption. The HomeBeat Energy Monitor, which costs only $35 per unit, feeds data from homes to the company every few seconds; Bidgely’s systems then analyse this and feed it back to the householder via Web or mobile, for example notifying them when they reach a preset consumption level. “People are used to instant gratification,” Bidgely’s founder and CEO Abhay Gupta has said.

Smart thinking

It’s not just communications that are enabling this kind of application, though: the number-crunching is equally important. Bidgely, for example, has attracted much interest from utilities with its employment of machine learning for a technique that it calls “energy disaggregation”, and describes as “the science that itemises consumers’ whole house energy data” – enabling the usage of individual appliances to be extracted from the household’s overall energy consumption figures, without physically monitoring the actual appliances.

In similar vein, smart grid technology specialist Tendril operates a platform for analysing consumer energy use (which it is now expanding from North America into Europe and Australia) based on a simulation model called True Home. This is claimed to deliver accurate predictions of consumption based on data such as past metering, weather and demographics, and can be applied to new customers without a billing history too.

Smart meters are invaluable in performing these analyses, and are spreading rapidly: in Europe, for example, it’s forecast that well over €50 billion will have been spent on them between 2015 and 2020, reaching 170 million homes and businesses.

But even customers who lack these AMI devices can benefit from the intelligence of the smart grid approach. Opower, for instance, was working along similar lines to Bidgely and Tendril when it developed an algorithm that breaks a customer’s energy usage down into categories (such as heating, appliances and so on) which can then be applied to ordinary meters read once a month as well as the smarter ones that report every quarter of an hour. Essentially, it assumes that the micro-level understanding of consumption patterns achieved through smart meters will translate to the situations where only dumb meters are found; it also builds in data on weather, household characteristics and other factors.

This kind of intelligence can help utilities as well as individual consumers, as when Opower analysed 812,000 customers in three U.S. markets to develop “load archetypes”, or typical patterns in the power demand of different kinds of users. Understanding these enables utilities to improve the match between supply and demand, achieving smart grid goals without necessarily having smartness in the grid itself – a further illustration that it’s all about the approach, not the nuts and bolts.