Towards to the Development of Virtual Testbed for Net Zero Energy Communities

Abstract

As a step towards future energy smart cities, net zero energy communities (NZECs) attract international attention and can be important contributions to combat the global climate change. However, the complexities of the NZEC, derived from the interactions between the various subsystems and components of the community energy system, make the design and control of NZECs difficult. To address this problem, we propose to develop a virtual testbed based on Historic Green Village, which is a real NZEC located in Anna Maria Island, Florida. This paper first introduces the design of the virtual NZEC testbed and the hybrid thermal-electric energy system that serves the Historic Green Village. We then show the implementation of Modelica models for the ground coupled heat pump subsystem, which is a part of the hybrid system. We also present some preliminary simulation results of the ground coupled heat pump subsystem.

Full text

ASHRAE and IBPSA-USA SimBuild 2016
Building Performance Modeling Conference
Salt Lake City, UT
August 8-12, 2016
TOWARDS TO THE DEVELOPMENT OF VIRTUAL TESTBED FOR NET ZERO
ENERGY COMMUNITIES
Dong He1, Sen Huang2, Wangda Zuo2,*, Raymond Kaiser3
1Chongqing University, Chongqing, China
2University of Miami, Coral Gables, FL
3Amzur Technologies, Inc. Tampa, FL
*Corresponding Author: w.zuo@miami.edu
ABSTRACT
As a step towards future energy smart cities, net zero
energy communities (NZECs) attract international
attention and can be important contributions to combat
the global climate change. However, the complexities of
the NZEC, derived from the interactions between the
various subsystems and components of the community
energy system, make the design and control of NZECs
difficult. To address this problem, we propose to develop
a virtual testbed based on Historic Green Village, which
is a real NZEC located in Anna Maria Island, Florida.
This paper first introduces the design of the virtual
NZEC testbed and the hybrid thermal-electric energy
system that serves the Historic Green Village. We then
show the implementation of Modelica models for the
ground coupled heat pump subsystem, which is a part of
the hybrid system. We also present some preliminary
simulation results of the ground coupled heat pump
subsystem.
INTRODUCTION
As a step towards future energy smart cities, net zero
energy communities (NZECs) attract international
attention and can be important contributions to combat
global climate change. According to the U.S.
Department of Energy (U.S. Department of Energy
2015), “a net zero energy community is an energy-
efficient community where, on a source energy basis, the
actual annual delivered energy is less than or equal to the
on-site renewable exported energy”.
In the past, much research has been conducted to achieve
net zero energy buildings (Fabrizio, et al. 2014, Moore
2014, Lu, et al. 2015). To move from net zero energy
buildings to future energy smart cities, an intermediate
step is to implement the net zero concept in communities.
In addition, past studies (Managan 2012, Athienitis, et al.
2015) show that it will be more effective to achieve the
net zero energy goal at a community scale than at a single
building because of the following reasons:
•Supplying renewable energy can be easier at the
community level because sites are typically
available to accommodate larger-scale, high-
efficiency renewable energy supply system
installations, such as solar photovoltaic, micro-
wind turbine, Combined Heat and Power (CHP)
and geothermal energy systems.
•Cogeneration plants that run on biomass or
waste can meet both the electricity and heating
needs of buildings across the community.
•Different buildings (commercial and
residential) with different occupancy patterns
and varying load profiles can be balanced to
flatten load profiles across a community,
because their time-of-use rates are different.
•Other energy systems, such as water supply,
wastewater treatment, and transportation can
also be included when considering the energy-
saving opportunities for the community.
There are a few studies of NZEC published in the
literature, such as (Gaiser, et al. 2014, Marique, et al.
2014, Orehounig, et al. 2014). However, there are only a
handful real NZEC reported in the literature. One
example is the West Village at Davis, California
(Gaiser, et al. 2014). It is an all-electric campus, where a
solar photovoltaic (PV) array and a digester, which
supplies a generator for additional electricity, are
installed. It also uses a heat pump with a backup electric
heater to supply the domestic hot water.
One reason for the scarcity of the real NZECs is the
difficulty of its design and operation. Computer
simulation can be used to support the design and
operation of NZEC. Table 1 summarizes the current
NZEC simulation studies and tools they used. These
studies have shown great potentials in terms of feasibility
and economics of NZECs. However, there are still some
limitations in the current studies. First, the models in
current studies are mainly designed for predefined
systems and cannot be easily tailored to accommodate
different systems and designs.
Second, many of existing modeling tools of the studies
do not support the unconventional system topologies
(Wetter, et al. 2006, Wetter 2009). As a result, these
models are usually designed for a subsystem of the
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
125

NZEC and are difficult to be extended to provide a
holistic representation of the entire NZEC, which may
have multiple sub-systems for energy generation,
distribution and consumption.
Table 1 Review of current modeling research for NZEC
Ref Community
Name
Simulation Tools
Kwan, et al.
(2011)
City College in
Los Angeles,
USA
RETScreen and PV
Watts v.2.0 (Solar PV
system Output
Simulation)
Coninck, et
al. (2014)
A dummy
community
Modelica (the DHW
system simulation)
Gaiser, et al.
(2014)
West Village
in California,
USA
eQuest (Energy
Consumption
Simulation)
PolySun and SAM
(Solar PV System
Output Simulation)
Lu, et al.
(2014)
Qingshan Lake
district in
Hangzhou,
China
MATLAB (Exergy
Estimation)
Marique, et
al. (2014)
Two common
archetypes of
neighborhood,
Belgium
Townscope software
(Solar PV system
Output Simulation).
Orehounig,
et al. (2014)
A village in
Switzerland
EnergyPlus (building
performance simulation)
Hachem-
Vermette, et
al. (2015)
A solar mixed-
use
community, in
Calgary,
Canada
EnergyPlus (Load
estimation, energy
consumption for DHW,
lighting and equipment).
TRNSYS (the
mechanical systems
simulation)
Kilkis
(2015)
Ӧstra Sala
backe in
Uppsala,
Sweden
Rational Exergy
Management Model
Analysis Tool (Exergy
Efficiency Analysis)
Third, existing research is mainly focused on the annual
energy performance evaluation. Thus, they tend to
ignore either the dynamic pattern or the interactions
between different systems to accelerate the simulation.
For the same reason, these modeling tools tend to highly
idealize the control (Wetter 2009, Wetter, et al. 2011,
Huang, et al. 2014). Although this is acceptable for the
annual energy performance analysis, it can be a critical
issue for the study of system controllability and stability.
Last, most current studies are not based on a real system
due to the limited availability of NZEC. Thus, they may
miss the real-world operating characteristics. To provide
a realistic simulation platform for the NZEC control
evaluation and optimization, a virtual NZEC testbed
based on the real NZEC is needed.
To address those challenges, we propose to develop a
virtual NZEC testbed based on an operational system
located in Anna Maria Island, Florida. The testbed is
designed for general purpose and can be easily modified
for different needs. In this testbed, Modelica, an
equation-based object-oriented cross-domain modeling
language (Fritzson, et al. 1998), is used to simulate the
various energy systems of the community. Modelica has
been applied in simulation research in many aspects of
the building industry, such as model predictive control
of chiller plant (Huang, et al. 2014, Huang, et al. 2016),
design of energy and water efficient hotels (Miranda, et
al. 2015), computational fluid dynamics (Bonvini 2012,
Zuo, et al. 2014, Zuo, et al. 2015), integrated district
energy simulation (Ruben, et al. 2015), and building
integrated with electrical grid (Wetter, et al. 2015a,
Wetter, et al. 2015b).
In the following parts, we will introduce the design of the
virtual NZEC testbed and the energy system
implemented in the Historic Green Village. We will then
show the implementation of a Modelica model for the
ground coupled heat pump system, which is one of the
subsystems implemented in the Historic Green Village.
We also present some preliminary simulation results of
the ground coupled heat pump subsystem.
DESIGN OF VIRTUAL NZEC TESTBED
Figure 1 illustrates the framework of the virtual NZEC
testbed. It is mainly composed of four modules: data
acquisition, pre-processing, optimization, and post-
processing. The framework can be used at either the
design phase or the operation phase.
In the data acquisition module, the data is mainly
obtained from the design documents in the design phase
or the onsite measurements in the operation phase. The
design documents usually include the building structural
parameters, plant parameters, and initial heating/cooling
loads. In the operation phase, data for environmental
parameters, energy consumption, the state of the devices,
and user’s behaviors will be collected through building
sensors and transmitted in real time to the cloud based
database. The obtained data will be pre-processed to
ensure the data quality can meet the needs of model
simulation. Usually we need to reject bad data and ensure
the continuity of the data. Script languages such as
Python (Python 2015), can be employed to automatize
the process.
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
126

The optimization module mainly includes the simulation
model and optimization engine. The models can be built
with different modeling tools. Also the optimization
engine can be chosen according to different optimization
objectives. In our designed testbed, Modelica was
chosen as the simulation program and GenOpt (Wetter
2004) /Matlab (Mathworks 2014) were chosen as the
optimization engine.
After the optimization process, the optimal results will
be accessed by different target persons. In the design
phase, the experts and designers can compare the
simulation results of various designs to further enhance
the design. During the operation phase, optimal control
variable values, or demand side management strategies
can be accessed by the operators or the energy managers
to improve the operation. The post-processing of the
optimization output data is also carried out using Python
automatically.
HISTORIC GREEN VILLAGE
The virtual testbed is based on a real NZEC, the Historic
Green Village. The Historic Green Village is located in
the Anna Maria Island, Florida. As shown in Figure 2,
the Historic Green Village is a mixed use community
mainly consisting of five mixed-use (retail, residential
and office) commercial buildings.
To satisfy the energy demand of these buildings, there
are three main subsystems in the village, including
electric energy subsystem, water-source heat pump
subsystem, and solar thermal domestic hot water
subsystem (Figure 3).
The electric energy system includes the PV, electric load,
and the distribution network. They form a micro-grid
which then interacts with the power grid. The water-
source heat pump sub-system includes water to air heat
(a) Building Layout
(b) Google Street View
Figure 2 Historic Green Village on Anna Maria Island, FL
Figure 1 Realization of the framework of the visual testbed
Data Acquisition
Optimal
Values
Optimization
Input File
Generation
Design
Documents
Cloud based
Database
Sensor &
Energy
Measurements
Enhance the design
Improve the operation
Experts
and
Designers
Operators
and Energy
Managers
Simulation Model
Optimal
Control
Optimization
engine
Optimal
Design
Design
Operation
Post-processing
Optimization
Output Data
Output File
Generation
Pre-processing
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
127

pumps, refrigerator racks, and a single ground-coupled
water loop with two boreholes. They provide cooling and
heating to all buildings within the community.
The solar thermal domestic hot water subsystem includes
three solar thermal domestic water heaters and is also
coupled with heat pumps for the purpose of heat
recovery.
Our preliminary work focused on the modeling of the
ground coupled water source heat pumps subsystem
shown in Figure 4. The ground source side consists of a
heat exchanger and two wells. The wells penetrate a
layer of limestone rock and are 450 feet deep. There are
nine heat pumps to provide cooling and heating to the
buildings. Each heat pump has two dedicated circulating
pumps. In addition, the water loop also provides directly
cooling for the refrigeration units in the General Store.
To ensure that the loop temperature stays with the design
range, a variable speed well pump controls the flow rate
of the ground water through the main heat exchanger
based on the exit water temperature.
MODELICA MODELS OF GROUND
SOURCE HEAT PUMP SUBSYSTEM
We modeled the ground source heat pump subsystem
with Modelica. A Modelica environment named Dymola
(Brück, et al. 2002) is used to compile the models and
perform the simulation. Dymola allows us to build the
system model in different levels. Using this advantage,
we can enhance the model reuse to speed up the model
development process. Also, we can easily identify the
model error to reduce the debugging cost.
Figure 5 shows the top level model of the ground coupled
heat pump subsystem. Some data need to be inputted
such as outside air temperatures, room air temperature
set points, switch of cooling/heating mode, set
temperature of the water leaving the heat pump,
cooling/heating load profiles, and other device
parameters. The main outputs of the model are energy
consumptions of individual heat pumps and room air
temperature.
Figure 6 shows the diagrams of the hierarchical models
for the heat pump system which includes the air side and
water side. Figure 6 (a) shows the model of one water to
air heat pump which include the packaged models of air
side and water side. Then Figure 6 (b) and Figure 6 (c)
show the detailed implementation of the air side and
water side.
Figure 3 Schematic of the energy systems of Historic Green Village
Figure 4 Schematic of ground coupled heat pump subsystem
Solar PV
Panel
Building
Solar Heater
Grid
Heat Pump
Heat Exchanger Borehole
Electricity
Domestic Hot Water
Heating /Cooling Air
Facility 1 Facility 2 Facility 3
Recovered Heat
Heating /Cooling Water
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
128

(a)
(b)
(c)
Figure 6 Diagram of heat pump module: (a) heat pump;
(b) air side; (c) water side
SIMULATION RESULTS AND ANALYSIS
Simulation results
Table 2 Designed cooling load for different buildings in
NZEC
Building
Name
Function
Area
(m2)
Index
Deigned
Cooling
Load
(kW)
Picklefish
Commercial
One
floor:
148.64
Com1
17.5
Pilsbury
Commercial
1st
floor:
112.22
Com2
7.0
Residential
2nd
floor:
77.11
Res1
14.1
Thelma
by the Sea
Commercial
1st
floor:
181.34
Com3
17.6
Residential
2nd
floor:
148.64
Res2
7.0
Sears
Commercial
One
floor:
101.07
Com4
14.1
Rosedale
Commercial
One
floor:
162.95
Com5
28.1
Because the measured data are not available at the time
of this study, we perform the simulation using the design
data.
Table 2 shows the design cooling load for each building
(residential and commercial). Based on that, we
determine the cooling loads profiles for each buildings
according to the ANSI/ASHRAE/IESNA Standard 90.1-
2007. We also assume the temperature of the water
entering into the heat pumps is constant.
Air Side
Water Side
Figure 5 Diagram of top level model of the ground coupled heat pump subsystem
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
129

The load curves of each building are showed in the upper
part of Figure 7. From this figure, we can see that there
are two different load styles: commercial building and
residential building. The commercial building with the
highest cooling load is Rosedale (Com5). From 12 am to
7 am the cooling load is zero, then it begins to increase
to the highest point of 22.5 kW at 4 pm. After that it
begins to decrease and drops to zero at 9 pm. The
residence on the second floor of the Pilsbury building has
the highest cooling load (Res1). From 12 am to 6 am the
cooling load keeps at highest level (12.7 kW). After that
it begins to drop and decrease to the lowest level (2.8
kW) at 10 am. From 4 pm it begins to increase and raises
to the same highest level as before at 12 am.
Figure 7 Building cooling loads (upper) and the total heat
pump power (lower)
The lower part of the Figure 7 shows the simulation
results of total heat pumps power consumption. From the
figure we can see the lowest energy consumption, about
5 kW, is at 7 am. It increases to about 25 kW at 4 pm
when the total cooling load is the highest. Subsequently
it decreases to about 5 kW at 10 pm when the total
cooling load is the lowest.
From Figure 7 we can see that the total energy
consumption of the heat pumps changes according to the
change of total cooling load.
Analysis of numerical performance
The simulation of a Modelica model typically proceeds
as follows (Jorissen, et al. 2015): First, the state variables
are initialized based on the initial equations and start
values. Then continuous time integration starts and
results are saved at intermediate time intervals. At
certain points in time, time or state events may occur,
which need to be handled by the integrator.
Totally, there are 21,635 variables in the Modelica
model. We used Dymola 2015 FD01 installed in a Dell
Workstation computer equipped with an Intel(R)
Xeon(R) CPU E5-2609. It took 716 seconds to simulate
a 24 hours physical process. Jacobian-evaluations are
used by the numerical solver. The number of Jacobian-
evaluations is 2,997. The minimum integration stepsize
is 8.91 × 10-6 s and the maximum integration stepsize is
41.9 s. The slow computing speed is mainly due to the
fast dynamics in the hydronic loop, which is needed for
the study of controability and stability. However, if one
only cares about the energy consumption, then it is
possible to speed up the computing speed by
representing the hydronic loop using models with less
dynamics.
CONCLUSION AND FUTURE WORK
This paper reports our preliminary work of developing a
virtual NZEC testbed for a net zero energy community
based on a real system. The Historic Green Village,
which is an existing NZEC, is used as reference for
testbed. As the first step in the development of the virtual
NZEC testbed, we modeled the ground coupled heat
pump subsystem of an existing net zero energy
community with the proposed testbed. Simulation results
show that the model is running, but the computing speed
is slow when the hydronic dynamics is considered.
In future work, the rest of the subsystems of the NZEC
will be modeled. We will also investigate how to
improve the numeric performance of the models
according to different application needs.
ACKNOWLEDGMENT
The first author conducted the research at the University
of Miami as a visiting scholar with the support of the
China Scholarship Council (No.201506050038) and
National Natural Science Foundation of China (Grant
No. 61473050). We also thank Mr. Tom Stockebrand,
PE (retired) for his valuable comments on this paper and
Liz and Mike Thrasher, the owners of the Historic Green
Village.
This work emerged from the Annex 60 project, an
international project conducted under the umbrella of the
International Energy Agency (IEA) within the Energy in
Buildings and Communities (EBC) Programme. Annex
60 will develop and demonstrate new generation
computational tools for building and community energy
systems based on Modelica, Functional Mockup
Interface and BIM standards.
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
130

REFERENCES
Athienitis, A., O'Brien, W. 2015. Modeling, Design,
and Optimization of Net-Zero Energy
Buildings. Germany, Wiley.
Bonvini, M. 2012. Efficient modeling and simulation
techniques for energy-related system-level
studies in buildings. Politecnico di Milano,
Dipartimento di Elettronica e Informazione.
Brück, D., Elmqvist, H., Mattsson, S. E., Olsson, H.
2002. Dymola for multi-engineering modeling
and simulation. In Proc. of Proceedings of
modelica.
Coninck, R. D., Baetens, R., Saelens, D., Woyte, A.,
Helsen, L. 2014. Rule-based demand-side
management of domestic hot water production
with heat pumps in zero energy
neighbourhoods. Journal of Building
Performance Simulation 7(4): 271-288.
Fabrizio, E., Seguro, F., Filippi, M. 2014. Integrated
HVAC and DHW production systems for Zero
Energy Buildings. Renewable and Sustainable
Energy Reviews 40(2014): 515-541.
Fritzson, P., Engelson, V. 1998. Modelica — A unified
object-oriented language for system modeling
and simulation. Lecture Notes in Computer
Science 1445: 67-90.
Gaiser, K., Stroeve, P. 2014. The impact of scheduling
appliances and rate structure on bill savings
for net-zero energy communities: Application
to West Village. Applied Energy 113(2014):
1586-1595.
Hachem-Vermette, C., Cubi, E., Bergerson, J. 2015.
Energy performance of a solar mixed-use
community. Sustainable Cities and
Society(2015).
Huang, S., Zuo, W. 2014. Optimization of the water-
cooled chiller plant system operation. In Proc.
of the 2014 ASHRAE/IBPSA-USA Building
Simulation Conference, Atlanta, GA, USA.
Huang, S., Zuo, W., Sohn, M. D. 2016. Amelioration of
the cooling load based chiller sequencing
control. Applied Energy 168: 204-215.
Jorissen, F., MichaelWetter, Helsen, L. 2015.
Simulation speed analysis and improvements
of Modelica models for building energy
simulation. In Proc. of the 11th International
Modelica Conference, Versailles, France.
Kilkis, S. 2015. Exergy transition planning for net-zero
districts. Energy(2015): 1-17.
Kwan, C. L., Kwan, T. J. 2011. The financials of
constructing a solar PV for net-zero energy
operations on college campuses. Utilies Policy
19(2011): 226-234.
Lu, H., Yu, Z., Alanne, K., Zhang, L., Fan, L., Xue, X.,
Martinac, I. 2014. Transition path towards
hybrid systems in China: Obtaining net-zero
exergy district using a multi-objective
optimization method. Energy and Buildings
85(2014): 524-535.
Lu, Y., Wang, S., Shan, K. 2015. Design optimization
and optimal control of grid-connected and
standalone nearly/net zero energy buildings.
Applied Energy 155(1): 463-477.
Managan, K. 2012. Net zero communities: one building
at a time, Institute for Building Efficiency,
Johnson Controls.
Marique, A.-F., Reiter, S. 2014. A simplified
framework to assess the feasibility of zero-
energy at the neighbourhood/community scale.
Energy and Buildings 82(2014): 114-122.
Mathworks (2014). Matlab. from
http://www.mathworks.com/.
Miranda, R. J., Huang, S., Barrios, G. A., Li, D., Zuo,
W. 2015. Energy efficient design for hotels in
the tropical climate using Modelica. In Proc.
of The 11th International Modelica
Conference, Versailles, France.
Moore, T. 2014. Modelling the through-life costs and
benefits of detached zero (net) energy housing
in Melbourne, Australia. Energy and Buildings
70(2014): 463-471.
Orehounig, K., Mavromatidis, G., Evins, R., Dorer, V.,
Carmeliet, J. 2014. Towards an energy
sustainable community: An energy system
analysis for a village in Switzerland. Energy
and Buildings 84(2014): 277-286.
Python (2015). from https://www.python.org/.
Ruben, B., Roel, d. C., Filip, J., Damien, P., Lieve, H.,
Dirk., S. 2015. OpenIDEAS-An open
framework for integrated district energy
simulation. In Proc. of the 14th IBPSA
Conference, Atlanta, USA.
Todorović, M. S. 2012. BPS, energy efficiency and
renewable energy sources for buildings
greening and zero energy cities planning.
Energy and Buildings 48(2012): 180-189.
U.S. Department of Energy 2015. A Common
Definition for zero energy buildings.
Wetter, M. 2004. GenOpt, generic optimization
program, User Manual, Version 2.0.0.
Technical report LBNL-54199, Berkeley, CA,
USA: Lawrence Berkeley National
Laboratory.
Wetter, M. 2009. Modelica-based modelling and
simulation to support research and
development in building energy and control
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
131

systems. Journal of Building Performance
Simulation 2(2): 143-161.
Wetter, M., Bonvini, M., Nouidui, T. S. 2015a.
Equation-based languages – A new paradigm
for building energy modeling, simulation and
optimization. Energy and Buildings(2015).
Wetter, M., Bonvini, M., Nouidui, T. S., Zuo, W.
2015b. Modelica Buildings library 2.0. In
Proc. of The 14th International Conference of
the International Building Performance
Simulation Association (Building Simulation
2015), Hyderabad, India.
Wetter, M., Haugstetter, C. 2006. MODELICA versus
TRNSYS - a comparison between an equation-
based and procedural modeling language for
building energy simulation. In Proc. of the
Second National IBPSA-USA Conference,
Cambridge, USA.
Wetter, M., Zuo, W., Nouidui, T. 2011. Recent
Developments of the Modelica "Buildings"
Library for Building Energy and Control
Systems. In Proc. of the 8th International
Modelica Conference, Dresden, Germany.
Zuo, W., Wetter, M., Li, D., Jin, M., Tian, W., Chen, Q.
2014. Coupled Simulation of Indoor
Environment, HVAC and Control System By
Using Fast Fluid Dynamics and Modelica. In
Proc. of 2014 ASHRAE/IBPSA-USA
Building Simulation Conference, Atlanta, GA.
Zuo, W., Wetter, M., Tian, W., Li, D., Jin, M., Chen, Q.
2015. Coupling Indoor Airflow, HVAC,
Control and Building Envelope Heat Transfer
in the Modelica Buildings Library. Journal of
Building Performance Simulation.
© 2016 ASHRAE (www.ashrae.org). For personal use only. Additional reproduction, distribution,
or transmission in either print or digital form is not permitted without ASHRAE's prior written permission.
132