The Importance of Aggregate Whole-Building Data ... - ENERGY STAR
The Importance of Aggregate Whole-Building Data for Benchmarking ENERGY STAR Data Access Network: Module 3 Agenda What is Aggregate Data and Why is it Important? Understanding Data Aggregation and Data Privacy Technical Considerations for Aggregating Data Resources What is Aggregate Data and Why is it Important? Defining Aggregation Monthly consumption data Reflects total usage across the whole building Combined monthly value for each energy type, regardless of number of tenants, accounts, or meters Aggregation Addresses Two Primary Goals of Data Access Streamline the benchmarking process for the building owner by reducing number of inputs Maintain privacy for individual tenants when providing
data to the building owner Streamlining the Benchmarking Process Portfolio Manager only requires monthly energy consumption at the whole-building level Users can set up and track multiple individual meters for each fuel type, but Many building owners would prefer not to deal with more granularity than is needed for the tool to work 6 Balancing Data Access and Data Privacy Utilities typically cannot provide tenant energy consumption to building owners without explicit tenant authorization Without complete, whole-building consumption data, benchmarking will not provide accurate results Many utilities and building owners see aggregation as a win-win solution for all parties 7 Understanding Data Aggregation and Data Privacy
8 Using Aggregation Thresholds to Protect Customer Privacy Need to avoid possibility for disaggregation or reidentification of tenant-level data Many utilities have chosen to establish an aggregation threshold Key consideration: what is the minimum number of tenants/accounts required for aggregation to effectively obscure individual customer data? Most utilities selecting a threshold between 2 and 5 tenants/accounts How Do Aggregation Thresholds Work? Typically expressed as a minimum number of tenants (or accounts) located at a single building If met, utilities will provide aggregate data to building owner without prior approval from tenants If not met, building owner must obtain tenant-level authorizations before utility can release aggregate data 10 Key Considerations Regarding Aggregation Thresholds Selecting the right threshold number
Many utilities settling in the range of 2 5 tenants Higher threshold fewer buildings able to receive data for benchmarking Establishing a clear process for buildings that dont reach the threshold Streamline tenant authorization process as much as possible (e.g., online portal or electronic forms) Ensuring clarity for building owners Is the threshold defined by number of tenants, number of meters, or number of accounts in the building? How is the aggregation threshold applied in the case of multibuilding properties (e.g., garden-style apartments)? 11 Keep Up With the Latest Information on Aggregation Thresholds Across the US www.energystar.gov/utilitydata 12 Technical Considerations for Aggregating Data 13
Key Technical Considerations for Providing Aggregate Whole-Building Data Meter-to-building mapping Identify all meters/accounts at a given property Ensure that all individual meter/account-level consumption data can be compiled into a single aggregate record Calendarization of individual meter records Necessary step if individual meters have different meter read periods Ensure that aggregate consumption record accurately reflects when consumption occurred Meter-to-Building Mapping: Overview Must identify all meters/accounts at the building being benchmarked. However, most utility data systems do not use the concept of a building Utilities may need to find creative approaches to identify which meters/accounts are associated with specific properties Once these associations have been made, then consumption can be aggregated across multiple meters/accounts Meter-to-Building Mapping: Utility-Led Utility makes the initial effort to identify all
accounts/meters at a property, based on limited input from building owner (e.g., building address; meter number for an owner-paid account) Typically includes an opportunity for building owner to review and provide feedback, to ensure that the accounts/meters identified by the utility correctly reflect all tenants Examples: Baltimore Gas & Electric, ComEd, Eversource (Boston/Cambridge, MA), PECO, Puget Sound Energy, Xcel 16 Meter-to-Building Mapping: Customer-Led Building owner required to identify all meter numbers associated with the building, then provide list to utility Utility prepares aggregate data set from the specific meters identified Examples: Pepco, Rocky Mountain Power, Pacific Power 17 Meter-to-Building Mapping: Comparing Options System Design UtilityLed
CustomerLed Customer Experience More effort anticipated (need to build out mapping queries, logic/criteria for inclusion, automation, etc.) Less effort anticipated (system just needs to be able to query for consumption data from specific meter numbers
provided) Utility leadership in obtaining wholebuilding aggregate data is likely to be seen as a valueadded solution Ensuring Accuracy Requirement to identify all meters will present a burden to some owners (e.g., large office or multifamily with hundreds of individual meters)
Utility may not be able (and isnt expected) to get 100% of the mapping correct on its own Requires review/input from building owner to confirm accuracy Utility assumes that meter numbers provided by owner are complete If meters are left out, utility has no way to know Ongoing Maintenance Ideally, the utility should track when individual meters
need to be added or removed from the aggregate total This information can be stored and updated in a central database, to ensure that the monthly aggregate wholebuilding consumption being compiled for a building is always complete and current 18 Calendarization: Overview Multiple meters at a property may not all have the same monthly read period Calendarization is the allocation of meter consumption data to a standard calendar month Allows summing of meters with different start and end dates 19 Calendarization: Different Approaches
Observed Time-weighted allocation: apportion a given periods consumption to a month based on the percentage of days in billing period that fall in that month Simplified allocation: assign all of a given periods consumption to a single month based on a specified criterion, such as: The month in which the majority of billing days fell for that record; or The month in which the meter read period started (or ended) 20 Calendarization: EPAs Recommended Approach EPA recommends that utilities use the time-weighted approach to calendarization Consistent with Portfolio Manager functionality Ensures that Portfolio Manager will be able to perform appropriate weather-normalization Ensures that the same result would be obtained if individual meter consumption records were entered directly into Portfolio Manager 21 Calendarization Example Account
Meter Start Date Meter End Date Tenant 1 12/16/2014 1/15/2015 Tenant 1 1/16/2015 Tenant 1 2/16/2015 Tenant 1 Consumption (kWh) Month 1 Prorated
Consumption Days in Month 1 Month 2 Prorated Consumption Days in Month 2 Monthly Total for Consumption Used for Entry into PM (kWh) 120,000 16 61,935 15 58,065
116,129 March 188,157 Common Area 12/5/2014 1/5/2015 20,000 26 16,774 5 3,226 Common Area 1/6/2015 2/5/2015
19,500 26 16,355 5 3,145 January 19,581 Common Area 2/6/2015 3/5/2015 19,200 23 15,771 5
3,429 February 18,917 Common Area 3/6/2015 4/5/2015 21,300 26 17,865 5 3,435 March 21,293 Calendar Month
Total Consumption Used for Entry into PM (kWh) January 390,290 February 368,982 March 403,131 22 Calendarization: Additional Considerations Portfolio Manager must be able to see data through the last day of a given calendar month, in order for that month to be included in metrics calculation May need more than 12 monthly consumption records per meter in order for Portfolio Manager to register a complete year of data across multiple meters When aggregating multiple meters, each meter may actually need to include up to 14 entries in order for the
aggregate value to reflect 12 complete calendar months 23 Calendarization: Illustration of Additional Considerations Example: seeking to obtain metrics for calendar year 2016 (1/1/16 12/31/16) 4 meters each comprise a full year of data, but dont align for the calendar year in question 2015 2016 2017 December January February March April May June July August September October November December January 1 10 20 31 1 10 20 31 1 10 20 28 1 10 20 31 1 10 20 30 1 10 20 31 1 10 20 30 1 10 20 31 1 10 20 31 1 10 20 30 1 10 20 30 1 10 20 30 1 10 20 31 1 10 20 31 Meter 1
Meter 2 Meter 3 Meter 4 1/1/16 - 12/31/16 12/10/15 - 12/9/16 2/1/16 - 1/31/17 1/10/16 - 1/9/17 Need to obtain and account for additional meter readings for, in order to see the full calendar year in question 2015 2016 2017 December January February March April May June July August September October November December January 1 10 20 31 1 10 20 31 1 10 20 28 1 10 20 31 1 10 20 30 1 10 20 31 1 10 20 30 1 10 20 31 1 10 20 31 1 10 20 30 1 10 20 30 1 10 20 30 1 10 20 31 1 10 20 31
Meter 1 Meter 2 Meter 3 Meter 4 1/1/16 - 12/31/16 12/10/15 - 12/9/16 2/1/16 - 1/31/17 1/10/16 - 1/9/17 24 Coming up in Module 4 Exploring the mechanisms through which utilities can deliver aggregate data to customers Understanding the unique value of Portfolio Manager web services as a delivery mechanism Reviewing various approaches in use by utilities for receiving and responding to data access requests Additional Resources 26 For Further Exploration
ENERGY STAR Interactive Utility Data Access Map ENERGY STAR Utility Data Access Fact Sheet Better Buildings Energy Data Accelerator: Best Practices for Providing Whole-Building Data Better Buildings Energy Data Accelerator: Guide to Data Access and Customer Confidentiality Better Buildings Energy Data Accelerator: Statistical Analysis of Data Access and Privacy 27 Thank you! If you have any questions, comments, or other feedback on the contents of this module, please email: [email protected] 28
The dog begins to salivate upon hearing the tone (neutral stimulus becomes conditioned stimulus). Did you follow the changes? The UR and the CR are the . same. response, triggered by different events. The difference is whether conditioning was necessary...
*The students were confused by the exam results. We use adjectives which end with "-ing" to talk about a person or a thing gives us that feeling. It means; How a situation appears to sb or sth. EXAMPLES: *This programme...
Braden and Braden Q Go-Live Day. On Thur., Jan. 26, we will use the Pressure Ulcer Dashboard as a reference for patients with active Pressure Ulcer. For each of those patients Systems Support Services staff will assist the patients nurse...
Historical Materialism. Workers are forced to do the impossible. The Marxist approach to the interpretation of history is known as historical materialism.. According to this theory, the history of society could be categorized by its economic characteristics.
Dormancy Requirements: Some seeds require more dormancy than others: Seeds lacking true dormancy—common vegetable garden seeds; ready to sprout, just need H. 2. O and warm-enough temperatures. Seeds endowed with deeper forms of dormancy—often taken from the wild
Section 4: Allows any Indian not residing upon a reservation, or for whose tribe no reservation has been provided, to secure an allotment upon public lands. Section 5: Provides that after an Indian person is allotted land, the United States...
Office Supplies. Extraordinary costs for office supplies that are used for project specific activities outlined in the proposal may be included in the budget and directly charged to the project. paper, envelopes or computer paper, used as part of a...