Getting the Most Out of Automated Metering Infrastructure
Utilities and grid operators find themselves increasingly at a loss, and with an increasingly pressing need, to acquire data and gain insight into rapidly changing energy consumption and behavioral patterns “behind the meter” as customers deploy intelligent, adaptive solar and battery energy storage systems and a panoply of smart, energy efficient home appliances, devices and equipment. They're intent on maintaining their primary, privileged place in power and energy markets, but they need to move quickly and fundamentally restructure their business lines and organizations if they're to succeed, something they have not needed or have been designed to do.
A host of young, digital, distributed energy tech-driven power and energy market players has emerged, leveraging and capitalizing on innovations that pave the way forward towards digital, decentralized and decarbonized energy. Utilities are keeping a close watch on this new branch of the industry ecosystem, making strategic investments in promising startups, often via venture capital or private equity arms, and collaborating with them on increasingly substantial and ambitious projects.
Having recently announced a strategic business development partnership and systems integration initiative, Israel's mPrest and Houston's Innowatts believe they have an end-to-end grid modernization, predictive analytics and management solution that stands out from the growing crowd. We continue our conversation with Innowatts' Senior Vice President Bob Champagne to gain insight into this collaborative effort and the resulting Grid Modernization “System of Systems” here in part two of this two-part series.
Utilities have been investing in and putting substantial amounts of capital at risk in bids to deploy smart meters and automated metering infrastructure across their service territories. News of early results regarding enhanced grid operations and cost savings have been encouraging, but utilities still lack a “critical lens” into customer-sited power generation and consumption. That puts utilities and their customers at risk when something goes wrong, and leaves utilities ill prepared to realize the full value of their smart meter and AMI investments, Champagne explained in part one of this series.
Generally speaking, the latest distributed energy resources management systems (DERMS) can measure and forecast the power output of DERs and the net power and energy they dispatch to the grid. They have difficulties when it comes to discerning, or forecasting, actual “behind the meter” consumption, which isn't metered, Champagne told Energy Central.
“That poses a big risk to the grid, for example given a co-load pick-up scenario where an outage occurs,” Champagne elaborated.
As per safety regulations, behind-the-meter solar PV systems automatically disconnect from the grid and shut down in the event of an outage. That renders them unable to enhance energy resilience by continuing to supply power to homes and businesses during and in the wake of outages, as has been highlighted in news reports and industry analysis regarding the effects of Hurricanes Harvey and Irma.
Problems, including follow-on outages, can occur as these behind-the-meter solar PV and battery energy storage systems recycle and come back online. “PV systems take various amounts of time to recycle, but they still have to draw on the grid, which represents actual consumption,” Champagne said.
The need for better data
That consumption, as well as the amount and rate of electrical energy they deliver to the grid isn't directly visible to utility grid managers, he continued. Similar scenarios can play out when an extreme weather or celestial event -- say an ash cloud or solar eclipse -- affects an area. They also play out more frequently, for example when fast moving cloud cover moves in and obscures solar panels, resulting in sharp, substantial drops in output and corresponding increases in load.
“Utilities need access to better data – more granular and at higher frequency – and they need to be able to simulate, predict and act on information, at times immediately,” Champagne said.
The root of the problem lies in the longstanding, fundamental approach utilities take when forecasting load. That, in turn, lies at the core of the commodity energy purchasing and selling they undertake, as well as the capital expenditures they make to enhance and maintain grid infrastructure.
Essentially, utilities take a top-down approach to load forecasting, starting at the level of aggregated load data -- recent and historical. That's used to construct load profiles and curves across time – minute-by-minute, hourly and/or daily, for example. Theoretical models, statistical tools and scenario analysis, along with weather data and load factors, are then used to forecast loads across sections of the grid down to the level of individual customer sites.
That's in the best of circumstances. “There may be just one such assumed load shape in the case of some electricity, heat and gas customers. They are generally very good at forecasting peak demand, but nothing beyond peak demand,” Champagne explained.
“DERs mostly affect load during ramp up and ramp down, but the load shapes are way off and load forecasts are off by a mile. Think about the impact of a forecast load profile: it's used to set up trading blocks and hedging strategies that are based solely on peak demand. They're missing out on the entire rest of the curve.
A creative approach to behind the meter load aggregation and disaggregation
“If you took a poll of 20 utilities, 18 might say all their grid planning is done on the basis of these load shapes. It's a big problem – we're talking about an essential service and a $2 billion industry market, probably $1 trillion or more when the retail side is factored in.
“Load data cannot be re-aggregated as part of producing a forecast from the top-down, at the feeder or substation level, which is typical of most conventional approaches."
Champagne says Innowatts' DERM and DR (demand response) applications can make a big, beneficial impact on about half that trillion dollars by virtue of taking a creative, market-proven approach to the problem of utilities having limited access to AMI and smart meter data. That includes resolving bandwidth challenges so that they can acquire the data they need in real-time.
The Bottom Line
Utilities need to be able to adjust on the fly given any change in grid network topography – the distribution of load and available energy resource capacity at any location at any point in time. They can take three different approaches in doing so, Champagne explained:
- Use top-down consumption intelligence and forecasting;
- Gather and collect real-time data down to the meter level, process it to produce load flow models and planning frameworks; or
- Make use of Innowatts' predictive forecasting and have actionable data and information available at any point in time from any grid node, including planning and engineering functions.
“In a competitive market, a 10 percent improvement [in load forecasting] can result in a 2 percent improvement to gross margins, so the potential gains are enormous. The financial gains for a utility that serves 1 million customers could be in the range of $30 million to $40 million a year.
“On the grid side, this helps utilities right-size the grid, makes for better, more strategically targeted investments based on real consumption profiles and forecasts. That makes for a lot less guesswork when it comes to asset capex and placement, and also addresses the issues of grid reliability and resiliency.”
No discussions yet. Start a discussion below.