Forum

You need to log in to create posts and topics.

New: openEEMeter integrated into opentaps

A few weeks ago I posted that we had added meters, Green Button XML meter reading data, and integrated OpenEEWeather to get weather data.  Now we've completed the integration of OpenEEMeter as well.

OpenEEMeter is an open source project which provides Measurement & Verification (M&V), which it means can be used to calculate energy savings using a statistical baseline model of energy use.  Together with Green Button XML and OpenEEWeather, the combination can help you analyze the effectiveness of your projects.

For more information about this integration, including a video and screenshots, please see https://opentaps.org/2020/01/30/green-button-xml-openeemeter-added-opentaps-mv/

Hello Michael,

So in answer to your questions about our development plans:

  • Watt Time for carbon intensity of electric grid -- Haven't thought of that yet.  What could this be used for?
  • Utility rates via OpenEI for comparison to other rate classes -- Yes, we're planning on this.  We're looking at getting this capability from NREL SAM.
  • IFTTT (for connecting to home devices to change energy usage.) -- This would be more for VOLTTRON to do.  VOLTTRON has a lot of capabilities connecting to building automations systems using BACNET and MODBUS, and somebody has been working on a HomeKit integration for it as well.
  • Local weather data (TMY3 or other) - We're currently getting this from OpenEEWeather, which is part of OpenEEMeter.  I suppose we can get it from another source if needed.
  • ResStock or URBANopt modeling techniques/code for comparing to OpenEEMeter M&V and considering investment scenarios - No, never thought of this one.  How would it work, and what could it do?
  • Solar panel inverter data, like SolarEdge, using its API and Guide - This can be done again using VOLTTRON which supports MODBUS protocol.
  • Third-party electric suppliers - No, was not aware this existed either.  Is this for integrating with other retail electricity options?  What would be the use case?  Comparing different rate plans with different demand charge or time of use to see which one is best?

My interests in these specific uses are based on eliminating barriers to development of local DER for cost savings, microgrids/resiliency, and investment for individual or aggregated projects.

  • Watt Time: useful for time interval data analysis of benefit case and price risk under a carbon policy change to individual users/buildings.
  • IFTTT:  useful for residential-scale solutions, similar to HomeKit. VOLTTRON seems more focused on commercial products, but still may be the best place to handle automation data, so I will exclude going forward.
  • ResStock or URBANopt modeling techniques/code for comparing to OpenEEMeter M&V and considering investment scenarios: useful for analyzing a community of buildings (i.e. 100,000) with public data sources. OpenEEMeter M&V is focused on evaluating impact of an investment after an expert has determined the correct remedy and applicable costs. ResStock enables evaluation of any size portfolio of improvements to a specific building or groups of buildings. URBANopt enables evaluation of a group of building for sharing larger energy infrastructure. ResStock helps find the right buildings for the right "fixes" for the right costs. OpenEEMeter helps confirm that the executed "fixes" resulted in the agreed energy savings. Both ResStock/OpenEEMeter are necessary. Both should have similar results, but will never be the same (for reasons of modeling). ResStock and URBANopt are NREL solutions that leverage EnergyPlus and other valuable energy modeling software.
  • Third-party electric suppliers: useful for customer recommendation analysis and cost savings available. Using the GreenButton energy data (hourly or daily intervals) allows evaluation of historic data to calculate ICAP tag. Most residents or small businesses do not know their options, nor if they are likely to save any money if they switch to Time of Use rates. (Yes-Comparing different rate plans with different demand charge or time of use to see which one is best)

Thanks for this.  I'll have to spend some more time looking at the different projects you mentioned.

Generally our plan now is to move opentaps forward so that it could be used to manage and then finance a portfolio of projects.  So we just finished features to calculate the financial value of metered savings (such as those from OpenEEMeter) and will next aggregate them into transactions (payments.)  Then we'll show you how to set up and distribute those payments to investors in the projects.

Related to this, did you say you had some experience with EnergyPlus modeling?  I was thinking about putting in a second M&V option: Calculating savings against building models from EnergyPlus (IPMVP Option D.)  I read that ModelKit (params) even has some existing prototype building models.  What do you think of this?

Yes, I have some EnergyPlus experience.

A second M&V option (IPMVP D or otherwise) depends on your goals. For my goals of 1-4 family residential whole-building energy use improvements and aggregation services, Option D is less useful. Here are some reasons:

  • The industry participants (utilities, ISO, aggregators) are moving toward Pay for Performance (which leans toward OpenEEMeter with actual savings instead of Option D projected savings);
  • Few investments in residential buildings prove their full value in less than 12 months (i.e. time in each season, time for savings "drift", time to capture multiple portfolio solutions), which is on par with OpenEEMeter and other performance systems (EnergySTAR Portfolio Manager, Dynamic LEED, etc)
  • My customers respond better to "actual savings" than "predicted savings". Actual saving matches their experience long-term. Predicted savings sets their expectations, yet may help "close the deal".
  • The ongoing performance data of improvements is more valuable than a snapshot in time (Option D), whether calibrated to energy use or not. An aggregator expects to have multiple revisits/touch points with the customer and needs ongoing energy use feedback to diagnose, analyze, and craft next solution.

If you set up whole-building energy modeling for purposes like proposing solution alternatives (i.e. What If scenarios), as opposed to M&V, then I think there is a good rationale to pursue it.

If your focus is small-medium commercial, then you need to explain more of the customer rationale and use case you aim to solve. Modelkit (params) and the experts at BigLadderSoftware helped many groups setup and integrate these for commercial projects. My previous team at Hitachi Consulting used params for portfolios of small-medium-large commercial buildings undergoing DER and EE upgrades across the US. I just need more help understanding opentaps initial markets and users before my comments are more useful.

The use case for opentaps is like this: A building owner enters into an agreement where if the building's energy use is less than an agreed amount, the owner would get paid.  Then an investor would finance improvements to the building's efficiency, and the owner would use the savings to pay the investor back.

The agreed amount of energy use would be based on a model, which could be statistical like OpenEEMeter or an actual simulation of the building.  That's where I thought building models built with EnergyPlus could be used.  opentaps would then be used to get the data for the model, run the model, get the metered energy use for the building, compare it against the modeled energy use, calculate the savings and payments due, and disburse the payments.

If you're familiar with Measured Energy Efficiency Transactions (MEETS), this would be similar but a more generalized: It could be for a short or long payback project and involving either utility or other off-takers who pay for the building's energy efficiency.

Where the difference in M&V methodology comes in will depend on the type of project and its payback.  If you have short-term projects with fast paybacks (Retrocommissioning and controls optimization), portfolios of smaller buildings, or buildings with highly predictable energy usage, then I think OpenEEMeter's statistical model is a good fit.  If you have long-term projects with long paybacks, like renewables or building envelope upgrades, then you would probably need an actual model of the building that everybody could agree upon.  This is especially the case if the building itself or usage could change, which would require people to agree how to change the model. If you look at the DOE's M&V Guidelines (https://www.energy.gov/sites/prod/files/2016/01/f28/mv_guide_4_0.pdf), they recommend IPMVP Option D for some of the bigger retrofit projects.

opentaps itself, though, is agnostic and M&V model independent.  Whatever the building owner, off taker (utility), and investor can agree upon, they can put into opentaps and run it to manage and finance the project.

Another variant is a group of smaller buildings or 1-4 family residential homes.  I think they're a great opportunity, but you have to properly incentivize them.  One idea is to "gamify" energy savings by putting the participants into a pool and giving social badges and rewards, in addition to just monetary payments.  This could also be done with opentaps by setting up non-monetary transactions to encourage efficiency.

Anyway, sorry for the long reply, but please let me know what you think.  Thanks!

If you have long-term projects with long paybacks, like renewables or building envelope upgrades, then you would probably need an actual model of the building that everybody could agree upon.

Renewable energy, or DER in general, are metered for production and therefore more easily isolated with other M&V protocol options. Building envelope changes, which tend to be deeper and drive different savings impacts at the whole-building energy meter based on space use and is not easily metered, is most appropriate for simulations or whole building metering. I understand why you'd like to have both options.

Another variant is a group of smaller buildings or 1-4 family residential homes.  I think they're a great opportunity, but you have to properly incentivize them.

The MEETS structure, or opentaps as a generalized version, could meet the residential market, but it requires full utility buy-in or aggregation-level involvement, like Community Choice Aggregators. Individual contracts can be made with residents/homeowners, but a utility is not involved until the portfolio reaches a minimum participation size in standardized markets, like demand response. While gamification is possible, and even achievable in reliable ways through companies like Opower and Tendril, it's a saturated solution that utilities have typically exhausted and is not as long-lasting/durable as deep retrofits.

For your interests in whole-building analysis, calibrated models of individual buildings can be done with params or EnergyPlus, but portfolio analysis of energy models can be done with ResStock and ComStock (residential and commercial, respectively), with final calibrations and modifications available for the final parties to negotiate, as will be required for every contract for the baseline and its savings adjustments.

Make sense?

Yes, thank you!  I'll look into ResStock and ComStock some more and see what we could do with them.