EnergyPlus Software:

Current Status, Challenges

& Opportunities

IMA Special Workshop

Brent T. Griffith
June 12, 2013
NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

EnergyPlus Development Team

DOE A. Roth
LBNL P. Haves, F. Buhl, M. Wetter, T. Hong, T. Nouidui, S.
Vidanovic, G. Sawaya
NREL B. Griffith, E. Lee, E. Bonnema, P. Tabares
ORNL J. New, B. Fricke, B. Shen, M. Bhandari
PNNL K. Gowri, W. Wang, D. Kang
FSEC L. Gu, R. Raustad, B. Nigusse, C. Sharma
GARD Analytics M. Witte, J. Glazer, R. Henninger
DHL Consulting L. Lawrie
UIUC R. Strand
OSU D. Fisher, M. Mitchell
2

Scop si obiective
Formulare Problem: Evaluarea echitabil i corect a
diferitelor msuri de eficien energetic pentru toate
tipurile de cldiri, necesit un cadru de analiz general
bazat pe fizica si sisteme de modelare integrate.
Impact: Energy Plus permite economia de energie n moduri generale:
prioritizeaz tehnologii, cuantific economiile pe ntragul, promoveaz
colaborarea tehnic cu artizanii de pe pia
Market Stimulation delivers integrated design tools, supports product
sales, quantifies value of efficiency measures, supports analysis for design
guides, education
Codes and Standards offers analysis for prescriptive levels, performance
path compliance, quantifies progress across revisions
EnergyPlus is a comprehensive, continuously updated and maintained, fully
supported and documented, free, and open-source solution!
3

EnergyPlus Overview
Engineering Modeling Tool for Time-series Building Performance Calculations
o
o
o
o
o

Energy Resources (use, peaks, cost, primary/source, pollution)

Thermal and Moisture Loads (mechanical design)
Thermal Comfort
Water Consumption
(Indoor Air Quality and Daylighting)

Numerical Characteristics
o
o

o
o

Time-stepping simulation (1 to 15 minute steps, for up to a year or more)

Forward predictions from a detailed description of building
Stochastic drivers from scheduled occupant behavior and controls
Large (and growing) number of component models in a modular framework

Dynamic models (where storage is important)

Quasi-steady models (where storage can be neglected)
First-principles models (where input data available)
Empirical black box models (where input data restricted)
Select between models of different fidelity (as input and application allow)

EnergyPlus Overview
EnergyPlus is large production software branded by DOE

604,600 lines of code (Fortran2003, some C++)

689 input objects (e.g., Fan), 55,143 fields
Runtime programming language
3,490 output variables, 89 summary tables
4,259 pages of documentation, 775 pages of validation reports
Online helpdesk, support for 3rd-party interface developers

EnergyPlus Overview
Development process: 6-month release cycle

Collaborative effortbroad expertise base

Distributed teamtwice-monthly calls, annual planning meeting
Programming standard
New feature proposal review and approval process
New projects selected from list of requests (370 items)
DOE guides priorities
Development team estimates effort
Consensus-based selections during planning meeting

Bugs from helpdesk tracked, prioritized, fixed and 2nd party

verification
Regression testing, timing checks
Validation test suite (ASHRAE Standard 140, others)

Market Impact
Third-party software based on EnergyPlus
US: AECOsim Energy Simulator (Bentley), Simergy (DOE, CEC, Trane, H-Q,
NEEA), OpenStudio (NREL), MLE+(UPenn), N++(ExpertApp),
CleanUrbanEnergy (CUE)
International: SMART ENERGY (India), DesignBuilder (UK), JEPlus (UK),
GreenSpaceLive (UK), VisualEPlus (China)

Organizations switching to EnergyPlus

California Energy Commission Title 24

Trane Company TRACE 800, development
Natural Resources Canada research, development
National Research Council Canada research
FSEC EnergyGauge software for FL energy code
Xcel utility design assistance program software
National Grid utility design assistance market research
PNNL COMcheck for IECC and ASHRAE Standard 90.1
Others under NDAs
7

EnergyPlus Overview
Major pieces
o
o

o
o
o
o
o
o
o
o
o

Input/Output
Operating conditions: weather, occupancy, lighting, plug
loads
Surface heat transfer: conduction, convection, radiation
Zone air heat balance model
HVAC air system: heating/cooling coils, fans, economizers, air
terminals, etc.
Plant (hydronic) system: chillers, boilers, pumps, cooling
towers, etc.
Sizing
Water systems
Commercial refrigeration: display cases, compressor racks,
etc.
On-site generation: photovoltaic, combined-heat and power
Runtime language for custom controls and modeling
8

Site and Operating Conditions

Location (lat, lon, elevation, timezone, etc)
Weather (design and hourly weather data, interpolation, sky
models, etc)
Scheduled Drivers
Occupancy (number of people, schedules, metobolic rates, etc.)
o Occupant-driven Loads (can be 30 to 80% of total energy use)
o

Lighting power and schedules

Appliance loads and schedules
o

Controls
Thermostat schedules
Hours of operation
Seasonal changeover
9

Building Loads: Surface Balances

3D geometry of planar surfaces
Inside Face Surface Energy Balance

Wall

Room

Shortwave radiation from

solar and internal sources
Longwave radiation
exchange with other
surfaces in zone

Conduction
from
outside, qki

Outside Face Surface Energy Balance

Surface conduction model options (1D)

o
o
o
o
o
o

Transfer function
Finite difference (2 schemes)
Combined heat and moisture finite difference
Moisture penetration transfer function
Internal Source/Sink (radiant HVAC)
Layer-by-layer detailed window model (ISO)

Longwave radiation from

internal sources
Convective heat exchange
with zone air

Shortwave radiation,
including direct,
reflected, and diffuse
sunlight

Conduction
into wall,
qko

Longwave radiation
from the
environment
Convective exchange
with outside air

Wall

Outside
Face

Surface convection coefficients (45 models)

10

Building Loads: Zone Air Balances

Zone Air Energy Balance
dTz N sl
Cz
Qi
dt i 1

N surfaces

i 1

hi Ai Tsi Tz

N zones

m C T
i 1

zi

Tz

minf C p T Tz msys C p Tsup Tz

Zone Air Moisture Balance

airVz CW

N sl
dWz
kg masssched load
dt
i 1

N surfaces

i 1

m W
W
N zones

Ai hmi airz Wsurfsi W

t
z

minf W Wzt msys Wsup

i 1

zi

Wzt

t
z

Zone Air Carbon Dioxide Balance

airVZ CCO 2 t

t
z

t t
z

N sl

kg
i 1

*10
6

masssched load

N zones

m C
i 1

zi

Czt minf C Czt msys Csup Czt

11

HVAC System Types

Ideal Air System (for load calcs)
Zone HVAC Forced Air
Self-contained, single-zone units
o Unit heater, window air-conditioner, etc.
o

Central Forced Air HVAC

Flexible topology for built-up air handlers
o Variable Air Volume, Dual Duct, etc
o

Radiant-based HVAC
Controlled by recalculating surface and air balances
o Chilled and/or heated slabs, radiative baseboards
o

12

HVAC and Plant Topology

Outdoor Air

Plant Supply Side

Cooling Loop

Supply Fan

ChW Pump

Outdoor Air
Mixing Box

Bypass

CC Plant Demand Side

Cooling Loop
Bypass

Chiller

Relief Air

East

Bypass

Return Air Mixer

VAV Box:ReHeat

Zone Air Splitter

Zone

Plant Demand Side

Heating Loop

West

Condenser
Bypass

Cond. Demand
Side Loop
Boiler

HW Pump

Zone

Condenser
Bypass

Plant Supply Side

Heating Loop

VAV Box:ReHeat
Bypass

North

Cond. Pump

Cooling
Tower

Zone
VAV Box:ReHeat

Cond. Supply Side Loop

13

HVAC Component Types

Fans (5 models)
Coils (24 models)
Air Terminal Units (14 models)
Dehumidifiers (2 models)
Humidifier
Evaporative Coolers (5 models)
Heat Recovery (3 models)

Individual, low-level component models are building blocks of HVAC

systems

14

Plant System Model

Multiple closed loops interact with each
other
Each loop breaks into two half loops
(demand and supply)
Node-and-branch-based topology
Parallel branch flow resolution
Fluid properties (water, glycols,
refrigerants)
Load dispatch and operation (sequential,
uniform, optimal, or component
setpoint)
Components called in flow order
Dynamic capacitance model at half loop

mc T Q
interface
T T

t
tank

t t
tank

p inlet

mc p

pump heat

A
1 to m
Components
B
n Splitter
C1

Cn

...

1 to i
Components
D1

1 to j
Components
Dn

n Mixer
E
1 to k
Components
F

mc p
exp

M tank c p

mc pTinlet Qpump heat

t

mc p

15

Plant Component Types

Pumps (4 models)
Chillers (11 models)
Boilers
Ice Storage (2 models)
Water thermal storage (2 models)
Water heaters (3 models)
Cooling towers (5 models)
Heat exchangers
District heating and cooling
Load profile

16

Sizing

Autosize inputs, program must fill data before simulating

Automatic design of mechanical equipment capacities
Building zone loads during design weather conditions
Find maximum capacity needed
Calculate sizes: Zone Loads => Air System => Plant

17

Water Systems

Mains temperature
Precipitation
Water using equipment (with hot water temperature-control)
Storage Tanks
Rain collector
On-site well
Cooling tower consumption
Cooling coil condensate collection
Evaporative cooling consumption
Green roof irrigation
Drain water heat recovery

18

Commercial Refrigeration

Display cases and walk-in coolers

Zone (moist) air interactions
Secondary loops
Heat reclaim to air or water
Refrigerated warehouse

19

On-Site and Renewable Generation

Dispatch (baseload, limit demand, track electrical demand,
thermal-led)
Inverters (3 models)
Batteries ( 2 models)
Transformer
Photovoltaic (3 models)
Microturbine (2 models)
Micro-Cogenerator
Fuel Cell
Wind Turbine
Flat plate hydronic solar collector
Photovoltaic/Thermal air or water collectors
Transpired air solar collector
20

User-Programmable Controls and Models

Energy Management System
Sensors
o Actuators
o

EnergyPlus Runtime Language (Erl)

Line code interpreter
o Domain-specific functions (psychrometrics, error messaging)
o IF-ELSEIF-ELSE blocks
o WHILE loops (new)
o

External Interface
Ptolemy Server
o Functional Mockup Interface (new)
o

User-defined HVAC and Plant Component Modeling Shells (new)

21

Challenges and Opportunities

Numerical Challenges
o
o
o

o
o
o

Solution speed
Robustness and stability (production software)
Different parts of the model have different response times
Non-linear models
Error detection (protect users from themselves and being misled)
Input data burden

o
o

Large domain (> 680 separate input objects)

Large data sets (realistic models require 10s of thousands lines of input)
Unavailable (need energy-related properties for everything, mfgs refuse)
Unknowable (how will occupants really use the building, building defects)

Cross-platform, portable, free software

Clear code authorship (issues for most libraries)
Compiler bugs
22

Challenges and Opportunities

Model Development Challenges
o
o
o

o
o
o
o

Academic efforts stop before models get into production software

Manufacturers cannot, or refuse to, provide data needed for energy
modeling
Mismatches in timescales and input burden (need to target whole-building)
Modular framework appeal (build it, but will they come?)
Licensing hurdles (now more historical)
Funding for implementation, but not new model development, physical
testing, and validation
Silod expertise misses opportunities to couple disparate models

Whole Building Energy Modeling Challenges

o
o
o
o
o

Geometry Simplification
Anti-simulation backlash (real world energy savings not always realized)
Graphical User Interfaces (many disparate types of users and use cases)
Input data libraries
Training, user support, and workforce issues
23

Challenges and Opportunities

Future Plans
Continue adding component models
Continue reengineering for speed, robustness, maintainability
XML input format
Major rewrite coming?
Gathering requirements
o Federal Funding Opportunity Announcement (FOA) planned
o Migrate from Fortran2003 to C++/Modelica
o FMU/FMI interfaces internal and external
o

24

EnergyPlus Applications based

on OpenStudio

OpenStudio Software Development Kit (SDK)

Enabling
Analysis for
Emerging
Technologies

Easy-to Use
Tools for
Practitioners

simuwatt
Mobile
Auditing Tool

Utility
Focused
Web Tools

Broad Market Penetration

Market Facing Modeling and Analysis Tools

OpenStudio Software Development Kit (SDK)

Modeling Engines (EnergyPlus, Radiance, Others)
Modeling Expertise

26

Xcel Energys Energy Design Assistance Program Tracker (EDAPT)

Reports
Template
Reports
Reports

EDAPT
automatically
generates
documents from
project data and
OpenStudio
output

Detailed
Portfolio
Tracking

27

simuwatt

Procedural Modeling

simuwatt Software Guided

Audit Workflow on Tablet

Key Goals

OpenStudio and Building

Component Library (BCL)

Analysis of Baseline and

Energy Conservation Measures

Reduce cost of investment-grade, level 3 audits below current cost of level 1 or 2

Produce higher quality, more consistent audits with greater residual value
Not simply a report that prescribes actions and quantifies savings

Data and models aggregate and can be reused for further cost reduction in EISA 2007
compliance, portfolio assessment, etc.
http://simuwatt.com/rd100.html
28

nationalgrid Customer Optimization For Energy Efficiency (COFEE)

Optimized
Incentives
1

1
EE Program
2

2
EE Program
n

.
.
.

EE Program

Program measures
are applied to
tuned model of a
customer building

Tuned Models

ROI,
Energy Savings,
etc.

ROI,
Energy Savings,
etc.
.

ROI,
Energy Savings,
etc.

.
.

Energy savings,
simple payback,
and more are
calculated from an
energy simulation

Customer
Cost Hurdle

Ranked and
Optimized
Marketing
Strategy for
Specific
Customer

Calculate incentive Ranked,

required to
customerovercome
specific
individual customer marketing
hurdle rate
strategy

Repeat Across Portfolio

29