A couple of months ago, NGK Insulators reported that one of its sodium sulfur (NaS) battery installations caught fire. The battery system itself is owned by Tokyo Electric Power Company (TEPCO) and it is installed at the Tsukuba Plant of Mitsubishi Materials Corporation.

Although NGK Insulators commercialized its NaS battery in 2001, the company is still dealing with safety issues. In many ways, NGK Insulators is at a disadvantage. Because of the success of the company’s NaS battery, there are now over 155 installations of NaS batteries globally with at least 140 systems installed in the TEPCO service area alone. In addition, many of these systems have been behind the meter installations for businesses looking to improve energy cost management and ensure power quality. These installations are in one of the most populous cities in the world (estimates vary, but Tokyo is typically in the top 20 of population dense cities).

In sum, this means that the company has approximately 140 sites, 200 MW, and ten years’ worth of battery installations in the most populous city in Japan. Relative to the number of installations and megawatt hours of energy delivered over the past decade, it is of little surprise that there has been an incident involving the company’s batteries. A fire at a remote substation or wind farm is one thing, however, most installed NaS batteries are behind the meter installations, which puts people’s safety at risk.

NGK Insulators is taking this fire seriously and has suspended manufacturing and installations until the company is satisfied the cause has been identified and resolved. Consequently, the firm has also revised its earnings through the end of its fiscal year, in this case March 31, 2012. The more important observation is that safety for grid-scale energy storage is an important technical issue for developers to address, and in many cases, it is an issue that may not come to light until systems are already installed and running for a good period of time.

So far, NGK Insulators has done everything it can to ensure the safety of its product, which is the best strategy for the company, the product, and the industry. In time, we may see other examples of these types of failures with grid-scale battery storage. As such, safety, manufacturing quality, and perhaps even safety of inputs will begin to differentiate vendors more and more.

At the IBM Smarter Cities forum in Rio de Janeiro last week, I had the chance to go behind the scenes and take a first-hand look at Rio’s smart city project. My main impression is that the project represents one of the purest emerging examples of a smart city project that is simultaneously developing smart solutions on multiple fronts – natural disaster management, public safety, health, utilities, to mention a few – and is starting to achieve a true “system of systems” – nirvana in smart city terms. This level of integration and interoperability across city agencies – and the successes Rio has had so far – bodes well for the smart city opportunity not only in emerging markets but worldwide.

The City of Rio de Janeiro has accomplished this by deploying smart technologies ranging from broad, continental-scale weather tracking down to mobile device-enabled notification systems for potholes and burnt-out streetlights. The centerpiece, of course, is the Rio Operations Center, which features Latin America’s largest screen and dozens of stations that provide visualizations of real-time data feeds. Within the center, 35 city agencies work together to synergize their responses to city events. (One interesting detail is that the operators wear uniforms modeled after NASA that create a sense of camaraderie and homogeneity across the historically separate city agencies, which creates something of a spectacle.)

To provide an example of how this works: If heavy rains cause flooding in a specific portion of the city, the operations center coordinates teams that notify citizens ahead of time via text message, close down the streets, mobilize ambulances, and shut down electricity distribution systems in the neighborhood to prevent electrocution. These processes are all pre-determined via standard operating procedures (SOPs). On the city side, bringing all these agencies under one roof helps break the silos that perennially plague the smooth delivery of city services. And, on the citizen side, it certainly helps that Brazil’s mobile device and networks are exploding, providing the platform for vigorous smart city app development and citizen involvement.

But technology is only one part of the winning recipe for a smart city. One persistent barrier echoed many times at the event is that smart city projects often rely heavily on the vision and initiative of specific mayors and administrations, which typically face four-year election cycles. The timetables required for certain types of infrastructure – particularly those involving high-tech and high initial capital expenditures – don’t always fit neatly into mayoral terms. Indeed, Rio’s mayor, Eduardo Paes, who spoke at the event, described the challenges of making progress on the project despite his uncertain future as mayor. Selecting smart city technology measures that optimize in terms of high net-present value, ease of deployment within a tight timeframe, and high PR benefits for the mayoral office seem to be emerging as the most pragmatic smart city solutions that address this challenge.

What differentiates Rio from other smart cities is the added challenge of managing its favelas – shantytowns perched on steep hillsides throughout the city that have historically received little in the way of city services or regulation – and integrating them with Rio’s urban fabric. These areas are among the most vulnerable to disasters such as mudslides as well as important symbolic testing grounds for Rio’s ability to serve even its poorest citizens as scrutiny of the city mounts in the lead-up to the 2014 World Cup and 2016 Olympics. From the perspective of a smart city, the favelas also provide opportunities for infrastructural “leapfrogging,” installing smart systems that could catapult these portions of the city to levels found in the rest of the city using state-of-the-art technology.

All in all, though, the event provided a clear picture of the concrete progress that’s being made on the smart city front and, in particular, the unique opportunities afforded by cities in emerging markets.

In a Washington Post op-ed entitled “Before Solyndra, a long history of failed government energy projects,” Steven Mufson argues that the U.S. government’s dismal record of failure in backing alternative energy technologies proves that the feds should get out of the business of backing energy programs altogether.  Citing the Clinch River Breeder Reactor, the Synthetic Fuels Corporation, and other examples, Mufson asks a question much on the minds of free-market economists and conservative members of Congress these days: “Should the government even be in the business of promoting particular energy technologies?”

Essentially a re-write of a study by the libertarian think tank the Cato Insitute – which also mentions the Yucca Mountain nuclear waste repository ‑ Mufson’s essay joins a growing chorus of commentators calling for an end to all energy subsidies.  There are several flawed assumptions behind these arguments, including the fallacy of composition – the confusing of the parts for the whole.  In this case, opponents point to highly public failures like Solyndra and Beacon Power, which declared bankruptcy last month, and claim that all of federal loan guarantees and other forms of backing for advanced energy projects are inherently doomed. 

The major flaw, however, is that Leonard, Mufson, and other critics hardly ever point out the biggest energy subsidy of all: the $4 billion or so in “incentives,” tax breaks, and other taxpayer donations to big oil companies.  In 2009, Exxon Mobil, the largest American oil company, posted a record $45.2 billion profits and paid little or no income tax.  “The tax breaks for oil have a long history — the so-called percentage depletion allowance for oil and natural gas wells dates to the 1920s — and have withstood repeated efforts to kill them,” The New York Times reports.

Those efforts have included at least three attempts by President Obama, all of which have been blocked by bipartisan opposition in Congress.

It’s worth repeating the explanation familiar to all students of energy economics: by paying little to nothing in taxes and enjoying sizable federal incentives for new exploration and drilling projects, the oil and gas industry is shifting the social costs of fossil fuel reliance (global climate change, industrial disasters like the Deepwater Horizon oil spill, huge infrastructure projects, and public health costs) onto American taxpayers.

For free market commentators, though, that’s nothing compared to the $535 million in loan guarantees handed out to Solyndra – a company pursuing a technology that will unquestionably be critical to the post-oil energy economy of the 21^st century, regardless of the fate of individual companies pursuing solar power.  

“An effort to eliminate all energy subsidies without instituting better alternative policies should be understood for what it is,” wrote Michael Levi, an energy analyst at the Council on Foreign Relations, in a recent blog, “a recipe for cementing the dominance of traditional fossil fuels against their competitors.”

At present, there is no widely acknowledged or precise definition of a virtual power plant (VPP). The very use of the term “virtual” connotes something not solid, temporary, and perhaps even fleeting. In many ways, these attributes are accurate. Unlike a large coal-fired power plant or nuclear reactor, most VPPs are fairly invisible to the naked eye. Yet, VPPs can provide extraordinary value and services to transmission and distribution (T&D) grid infrastructure as well as to myriad stakeholders engaged in the provision of electric power.

Growing investments in the full range of distributed resources – renewable distributed energy generation (RDEG), demand response (DR) load, advanced storage, and plug-in hybrid vehicles (PHEV) – throughout the world will require new business platforms to manage this increasing diversity and complexity. The increasing variability of both generation (from solar and wind) and loads (due to DR and PHEVs) will require more sophisticated and decentralized decision making, embedding intelligence deep down throughout the system’s distribution lines.

Some observers provide such a broad definition of a VPP that virtually the entire power market could be segregated into regional VPPs. For example, The California Renewable Energy Collaborative (of the University of California) claims that each utility system could be considered a load-following VPP. With structures like Competitive Renewable Energy Zones (CREZ), which are currently being developed in large states, such as California and Texas, central station power plants like concentrated solar power projects or large utility-scale wind farms could also be viewed as VPPs. Going down to the distribution level of grid service, a net zero community’s energy infrastructure is a VPP, as is a net zero energy building, according to this perspective. All have costs associated with supply, delivery, storage, and efficiency components; they also all require integrated planning and operations.

As a result of this wide divergence of what constitutes a VPP, Pike Research has come up with a narrower, yet still broad definition of a VPP that encompasses four different primary segments: DR-VPPs, supply-side VPPs, mixed asset VPPs, and wholesale auction VPPs. What one company calls a VPP is often dismissed by a competitor. In fact, one could argue that the vast majority of today’s DR resources do not even meet the following Pike Research definition of a VPP: a system that relies upon software systems to remotely and automatically dispatch and optimize generation, demand-side, or storage resources (including PHEVs and bi-directional inverters) in a single, secure web-connected system.

Ironically enough, the most vibrant and commercially advanced segment of VPPs does not involve (virtually) any power generation at all! Like the microgrid sector segment of the smart grid, these so-called DR resources are most advanced in the United States. With strong regulatory and policy support from the Federal Energy Regulatory Commission (FERC), this segment is undergoing rapid changes as new technologies empower utilities and end-user customers alike to respond to changing market conditions when real-time and dynamic pricing programs are offered in organized markets.

With the backing of the FERC, DR simply turns conventional wisdom on its head. Instead of firing up a fossil fuel peaking power plant on a hot afternoon, it is far more cost effective and efficient just to lower demand by an equal amount. One of the best ways to capture the essence of a this VPP market segment is with this definition: the ability to tap resources in real time, and with enough granularity, to control the load profiles of customers, aggregate these resources, and put them up on a trader’s desk.

Considerable debate exists as to whether managed load reductions and dynamic pricing programs should both fall under the VPP umbrella. For the purposes of this report, the entire universe of DR activity is presented. However, it is the dynamic pricing segment – and particularly automated DR – that most clearly matches up with VPP vision. Using this narrower definition would reduce DR-VPP capacity by over half, but still represent the largest single category of VPPs by 2017. As a sign of expected surge in growth, utility dynamic pricing capacity grows from 4.8 GW in 2011 to 29.9 GW by 2017, a 35.6% CAGR.

Between now and 2017, DR-VPPs (both managed load reduction and dynamic pricing) will lead the way in terms of projected capacity additions within the forecast period, especially in the United States, with 44 GW already available today, virtually doubling in capacity to 85 GW by 2017 in the average scenario. As documented in the new report from Pike Research, DR is also now slowly gaining traction in Europe and more rapidly taking off in the Asia Pacific region as smart grid deployments expand. Unlike the U.S. market dynamics, DR is less about shaving peaks and more about balancing variable RDEG and large-scale renewables in other parts of the world. In terms of Europe, Denmark and Germany are leading the way in the supply-side and mixed asset category of VPPs. But more about those models in a future blog post.