Once an organization has a grasp on all the energy data at its fingertips, the logical next step is figuring out how to use that information.
In my last post, I discussed the various kinds of energy data and how to collect them. But how do you align data collection goals with energy management initiatives across the enterprise?
High-resolution data from smart meters or other interval metering technologies are significantly more robust than utility bills, but this increased granularity is not always required. There are some energy management use cases that just require utility bill data.
First, what is the value proposition? There are a variety of estimates, and in some cases they compound savings across a full energy information system (EIS), rather than by use case. Lawrence Berkeley National Lab calculates that EIS can help realize 17 percent median site savings and 8 percent median portfolio savings, and most other sources estimate savings at 10 to 20 percent.
Energy reporting and benchmarking
This is one of the first steps that almost every enterprise will go through when establishing an energy management program.
Benchmarking can answer questions such as “how am I doing?” and “which sites are performing well, and which are laggards?” In addition, energy management programs typically pull data from many utilities and many other sources (interval and/or utility data), so centralized reporting of this information yields valuable insights.
Most industry benchmarking, such as Energy Star, is based on utility bill data. So, if the goal is simply to have Energy Star scores across your enterprise, your organizations will just need utility bills.
At the same time, benchmarking with interval data can help to identify buildings that have particularly expensive peak demand charges or inefficient startups and shutdowns. The interval data will provide more detail about where and when these problems occur, making it easier to address them.
That said, some firms might find that using utility bills to benchmark a portfolio provides enough information before selecting specific buildings to undergo a detailed energy audit.
This is a nebulous concept within the energy management space. Most organizations will attempt to estimate future energy costs and consumption based on past performance. This normally is part of a corporate-wide annual budgeting initiative.
Some firms will then track against these projections throughout the year. This has been a core use case for energy management for years, if not decades.
There are a few different approaches to developing a forward looking estimate, from just adding an inflation percentage to the current year’s consumption to using a more complex model with multiple variables, such as changes in square footage, utility rates or overall energy use.
Once the variables are set for the coming year, a budget estimate of energy use and cost can be generated. The challenge is that these budgets are only as good as the underlying assumptions. If you think that footage will grow by 5 percent and peak demand will go down by 2 percent, there are a variety of tools and services to build this budget. But the budget won’t be accurate if these assumptions don’t materialize.
Interval data can add granularity to the budget while also adding significant complexity. Some organizations use interval data to provide daily and weekly budget visibility. The budget is generated based on monthly utility bills, but it is tracked based on interval data.
Most building professionals want to know as soon as possible when actual performance deviates from the budget. The granularity of interval data can provide advance notice if a building is on track to exceed its budget. By the time the utility bill arrives, and it is too late to avoid the overage.
Interval data can give building professionals time to make changes that bring the budget back in line. If the budget value still happens to be exceeded, interval data will provide a better story around why it happened.
This is an increasingly common strategy that requires interval data. The goal is to reduce peak demand charges by reducing demand at the times in which a building is using the most energy, to avoid costly peak demand charges.
Battery storage vendor Stem reports that demand charges can make up 50 percent of the utility bill. The building must have visibility into interval demand, which is not commonly provided on utility bills.
A utility bill may identify the single point in time each month that peak demand was set, but without knowing if there are points throughout the month when the demand was nearly as high as the peak, it is difficult to build a demand reduction strategy. For example, avoiding a peak of 1,000 KW isn’t cost effective other weekdays have a peak of 990 to 995 KW.
A utility bill may not provide this full picture, which will make it hard for building professional to understand the opportunities to reduce demand and build a plan to do so.
Real time data from meters, rather than utility-owned smart meters, will enable alerts to be set that proactively warn users of potential peak demand thresholds.
These data enable control scenarios to be implemented that automatically reduce demand, such as the dimming of lights or small modifications to temperature setpoints. Utility bills may identify a peak demand problem but can’t solve it.
Settling the tab
There are a few kinds of utility bill validation. With only utility bills, a review of rates and check for billing errors may yield significant cost savings.
Accenture reports that 1-2 percent of all utility bills contain errors. This is especially true for firms that have hundreds or thousands of utility bills and rarely review them in detail. At the same time, comparing values from interval meters to the utility bill statements may identify even more discrepancies and billing errors.
But the process for challenging a utility bill based on what likely are non-revenue grade interval meters can be complex and varies by utility.
The goal of measurement and verification (M&V) is to calculate energy savings from efficiency projects and operational changes. Instead of comparing energy use before and after a project is implemented, M&V enables a comparison of actual energy use to what the use would have been if the project was not conducted.
This is more accurate, since occupancy and weather changes might impact energy use after the project. Without the efficiency project, energy consumption would have been even higher. M&V approaches typically include building a regression model of pre-retrofit energy use with variables like weather and occupancy.
M&V is the foundation of performance contracting and many utility-sponsored energy efficiency programs. There are industry standards from the Efficiency Valuation Organization and ASHRAE, which detail how M&V should function.
These standards do not dictate the use of a particular data set, and in many cases, utility bills are used to build a baseline model and conduct M&V.
At the same time, Lawrence Berkeley National Lab and others have been actively investigating how interval data can improve M&V, generally finding that accurate regression models can be developed with interval data. Additionally, LBNL surveyed some of the leading commercial products that use interval data for M&V finding that the state of the market is strong.
M&V can be conducted with utility bills or interval data, but moving forward, interval data-based models will show a variety of advantages and the gap between these two data sources will widen.
Alerts and anomaly detection
Alerts can replace reactive, schedule-based operations to a more pro-active effort. Especially when software solutions provide advanced workflow- and role-management capabilities, alerts that direct building professionals to problems right when they occur (or even before they occur), is more effective than finding issues when a utility bill arrives or when an occupant files a complaint.
Utility bill data can drive billing alerts, but the market is moving towards more interval data-based alerting capabilities.
In addition to interval energy data, trend data from a building automation system (BAS) can drivealerts. Insights from this data, such as current chiller performance, can identify specific equipment problems.
These are a few of the primary use cases for energy data. Firms looking to build an energy management plan should consider data availability and think about the desired use cases and outcomes.
While interval data does provide benefits above and beyond what can be accomplished with utility bills, there are some scenarios in which bills are satisfactory.
Source: Joseph Aamidor’s “GreenBiz 101: Putting your energy data to work” on GreenBiz