Académique Documents
Professionnel Documents
Culture Documents
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/309751099
CITATIONS READS
0 25
4 authors, including:
All content following this page was uploaded by Nima Alibabaei on 15 November 2016.
The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Applied Energy 185 (2017) 2943
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
h i g h l i g h t s
a r t i c l e i n f o a b s t r a c t
Article history: Based on their structure, residential houses/buildings (RHs) can offer excellent opportunities for manag-
Received 20 July 2016 ing their internal energy demand and subsequently lowering their energy cost. Demand management and
Received in revised form 22 September energy cost saving can be achieved by taking advantage of RHs/buildings capabilities in storing thermal
2016
energy. Thermal energy can be stored utilizing intelligent Strategy Planning Models (SPMs) which are
Accepted 20 October 2016
applied in the heating, ventilating and air conditioning (HVAC) system as one of the largest energy con-
sumer in RHs buildings. This study discusses the development of three different strategy planning models
including Smart Dual Fuel Switching System (SDFSS), Load Shifting (LSH), and LSHSDFSS, a combination
Keywords:
Advanced residential HVAC controller
of load shifting and fuel switching SPMs. In order to facilitate the implementation of the developed SPMs
Strategy planning models on the HVAC system of the house used in this case study, an advanced controller was designed by con-
Load shifting necting both TRNSYS-Matlab programs. The HVAC system energy demand as well as the corresponding
Fuel switching system saving on the HVAC system energy cost are analyzed in-depth numerically using each of the strategy
Residential houses planning models during both the heating and cooling seasons. Simulation results showed that in the
heating season, the operating/energy cost of HVAC system decreased significantly (23.8%) by implement-
ing SDFSS-SPM. LSHSDFSS-SPM reduced the HVAC system operating cost by 15.8%. In the cooling season,
LSH-SPM reduced the HVAC system operating cost by 6.63%.
2016 Elsevier Ltd. All rights reserved.
1. Introduction [4] shown in simulation result, RHs enable to improve the manage-
ment of energy network. The network energy saving is improved
Residential houses/buildings (RHs) must be seen as significant significantly using the distribution locational marginal price (D-
elements of a larger, dynamic network of energy system. Therefore, LMP) model developed in [4]. What makes the RHs roles promi-
a network of energy system is significantly affected by the behav- nent in managing the energy network, is their energy structure.
ior of RHs [1,2]. For example, based on Pagani and Aiello [3] energy They are capable of storing thermal energy that results in manag-
model simulation result, disruption in demand management can ing their energy demand. Arteconi et al. [5] have investigated the
be detrimental to energy systems. In contrast, as Siano and Sarno load shifting potentials of thermal energy storage (TES) in a resi-
dential building. The simulation result showed energy cost saving
using on and off peak tariffs. Thermal energy storage and energy
Corresponding author. demand management can be achieved by employing smart Strat-
E-mail addresses: nima.alibabaei@ryerson.ca (N. Alibabaei), alanfung@ryerson. egy Planning Models (SPMs) in the HVAC system as one of the lar-
ca (A.S. Fung), kraahemi@ryerson.ca (K. Raahemifar), arash.moghimi@mail. gest energy consumers in RHs/buildings. In Vakiloroaya et al. [6]
utoronto.ca (A. Moghimi).
http://dx.doi.org/10.1016/j.apenergy.2016.10.062
0306-2619/ 2016 Elsevier Ltd. All rights reserved.
30 N. Alibabaei et al. / Applied Energy 185 (2017) 2943
Nomenclature
ASH Archetype Sustainable Houses LSHSDFSS load shifting and smart dual fuel switching system
AHU Air Handling Unit OEB Ontario Energy Board
ASHP Air Source Heat Pump OT outdoor temperatures
BCS Best Case Scenario RH residential house
COP Coefficient of Performance SDFSS Smart Dual Fuel Switching System
DEC Daily Energy Cost SPM strategy planning model
DHW Domestic Hot Water TC time constant
DSO Distribution System Operators TOU time-of-use
HVAC heating, ventilating, and air conditioning TRCA Toronto and Region Conservation Authority
LEED Leadership in Energy and Environmental Design ZT zone temperature
LSH load shifting
review paper, different demand management and energy saving Furthermore, matching and tuning SPMs with the real time
SPMs for typical residential HVAC systems are described in detail dynamic characteristics of the process model (house thermal
and compared. Huang et al. [7] showed the advantage of different model) has been poorly noted before.
energy saving SPMs such as load shifting in a HVAC predictive The main contribution of this study is in developing novel strat-
energy model. The effects of different energy conservative SPMs egy planning models including: (1) Smart Dual Fuel Switching Sys-
such as daily optimal deadband and daily optimal set point are tem (SDFSS), (2) Load Shifting (LSH), and (3) LSHSDFSS model as
investigated in [8] using EnergyPlus program. Kim et al. [9] have the combination of fuel switching and load shifting strategy plan-
developed a daylighting meta-model that has been integrated into ning models. In SDFSS-SPM, a smart controller is developed to
a HVAC system to take the maximum advantage of daylighting for select the least expensive hot air supplier (between electrical Air
preparing thermal comfort. This meta-model offered an average of Source Heat Pump (ASHP) and natural gas mini boiler), in each
13.7% energy saving. Different heat gain reduction methodologies/ hour, by taking into consideration the houses thermal demand
SPMs have been developed in [10]. Based on this paper calculation, and Time of Use (TOU) pricing scheme during the decision making
more than 75% of building heat gain were generated by solar heat process. In novel LSH-SPM, an intelligent mechanism is used to
gain and lighting system. By managing these heat gains, more than select the best pre-heating/pre-cooling starting time based on the
45% saving on HVAC system energy cost was achieved. Christan- outdoor temperature effects and the dynamic characteristics of
toni et al. [11] used EnergyPlus simulation model to implement the case study house. As the third developed SPM, LSHSDFSS-
different demand response SPMs for shifting the building electrical SPM takes advantage of both novel load shifting and fuel switching
demand. In this research, contribution of HVAC system and build- systems to offer maximum saving on HVAC system energy cost.
ing capacitance was evaluated using demand response SPMs. In An advanced controller is developed by connecting TRNSYS-
[12] research, energy saving potentials of various set point strate- Matlab programs. This advanced controller facilitates the imple-
gies were investigated in a museum. Using these strategies led to mentation of novel SPMs on the HVAC system of case study house.
77% improvement in thermal comfort while decreased 82% of The behavior of HVAC system is numerically simulated in-depth
HVAC system energy demand. Wang et al. [13] have modeled the during winter (heating) and summer (cooling) seasons. Total sav-
influence of occupants and its essential effects on building perfor- ing on the HVAC system energy demand and cost are calculated
mance by generating mean profiles of occupancy variables in order by implementing developed SPMs in winter and summer seasons.
to increase HVAC system energy efficiency. The role of residential This paper is consisted of four sections. In Section 2, the
heat pumps and load shifting, and their contribution in network methodology and model of the work are described. This section
operational cost and CO2 emission reduction has been investigated also concludes the architecture and configuration of different oper-
in [14]. As Beizaee et al. [15] have shown, the thermal demand of a ational SPMs in detail. In Section 3, results including the effect of
house can be decreased using zonal space heating SPM. This SPM intelligent strategy planning models on the HVAC system energy
could reduce the natural gas consumption by 11.2% in only eight demand and cost in various sample days and during the whole
weeks. Chassin et al. [16] investigated the impacts of discrete- heating and cooling seasons are presented and discussed. The
time SPM on a residential house HVAC system and reported up paper is concluded in Section 4.
to 25% reduction in HVAC system energy demand. A dynamic zone
modeling system as an energy conservative SPM was developed by
Li et al. [17] to reduce the HVAC system energy cost. Different inge- 2. Model description and intelligent Strategy Planning Models
nious methods like using tokens (as a surrogate for thermal (SPMs) development
demand) [18] and demand response potentials of high-raised
building ventilation fans [19] have been used as smart SPMs to 2.1. Model description
conserve the network energy.
In addition to supporting the network of energy system, lower 2.1.1. Methodology
overall energy cost could be achieved for the homeowners with This study is started by running case study house TRNSYS
such intelligent SPMs. Hence, as Di Giorgio and Liberati [20] con- model for a specific day in order to generate the baseline data.
cluded, both energy consumers and local grid benefit from such The weather file used to calculate the houses thermal demand is
intelligent planning models. Regardless of the fact that numerous the metropolitan Toronto weather given in the TRNSYS library. Dif-
research with different criteria have been previously conducted ferent operational command matrices are generated by Matlab
for designing different strategy planning models for residential advanced controller (according to the baseline data) in order to
HVAC systems with the aim of reducing demand and energy cost, implement various SPMs on the houses HVAC system. TRNSYS
less attention has been paid to energy conservative SPMs that model is run again considering the generated operational com-
use combination of fuels/energies for running the HVAC system. mand matrices in simulation process. The data generated by imple-
N. Alibabaei et al. / Applied Energy 185 (2017) 2943 31
24
22
20
18
16
14
12
10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Fig. 3. Price of electricity in Ontario as of November 1st, 2015 for winter and
Fig. 1. Framework of the house energy simulation system. summer seasons.
32 N. Alibabaei et al. / Applied Energy 185 (2017) 2943
2.2.1. HVAC system energy cost reduction using smart dual fuel
switching system strategy planning model (SDFSS-SPM)
The test house (House A) has an electric two-stage variable Fig. 4. House A ASHP COP validated with outdoor temperature [27].
capacity ASHP and a natural gas mini boiler to generate hot air
through the Air Handling Unit (AHU) in order to meet the space
heating demand. This strategy planning model aims to reduce
the HVAC system energy cost at each specific hour by selecting
the least expensive hot air supplier and setting up the correspond-
ing system (ASHP or mini boiler) to meet the space heating
demand.
2.2.1.1. Estimating the Fuel Cost of ASHP. Electricity price and out-
door temperature are the two main variables affecting the ASHP
energy cost. Outdoor temperature directly affects the air source
heat pump Coefficient of Performance (COP). Fig. 4 demonstrates
the experimentally validated COP curve of House A ASHP [23,27].
After determining the electricity price and COP for a given hour,
the cost of the energy unit produced by the ASHP is calculated
using the following equation:
ASHP electricity cost to prepare one unit of thermal energy $=kW h
Electricity price !=kW h
COP 100
1
where Electricity price represents the electricity cost !=kW h at a Fig. 5. Mini boiler efficiency curve -VIESMAN Co. technical data manual [28].
given hour due to TOU pricing scheme and COP indicates the ASHP
coefficient of performance calculated based on the outdoor temper- In Eq. (2), the mini boiler efficiency is determined based on the
ature at a given hour. load percentage at a given time. The 10.3 constant is used to con-
vert natural gas energy content from 1 m3 to kW h.
2.2.1.2. Estimating the fuel cost of mini boiler. Based on OEBs natural The least expensive hot air supplier is simply chosen by compar-
gas pricing scheme, the price of natural gas is the same during the ing the expected fuel cost of mini boiler and ASHP at each particular
whole day. In this study, based on OEB prices the natural gas price hour. Based on House A HVAC system instruction, the ASHP is
is estimated to be 35 /m3. Thus, the only variable used to estimate selected as the main hot air supplier and mini boiler is just used
the cost of each unit of thermal energy (produced by the boiler) is as a backup system at House A. This prioritization is exchanged by
mini boiler efficiency. The load percentage (water flow rate circu- setting up the AHU controller relay using an imperative command
lating through the boiler) is used to determine the efficiency of signal.
the boiler. Fig. 5 shows the efficiency curve of mini boiler provided
by the manufacturer [28]. Mini boiler fuel cost (to prepare one unit
2.2.1.3. Using SDFSS as a strategy planning model. The operational
of thermal energy) is calculated by the following equation:
boundaries of ASHP and mini boiler are calculated using Eqs. (1)
Mini boiler natural gas cost to prepare one unit of thermal energy $=kW h and (2). Fig. 6 presents these boundaries classified based on off-,
Natural Gas price $=m3 mid-, and peak hours. In Fig. 6, the maximum and minimum tem-
peratures are selected based on the hottest and coldest tempera-
Efficiency 10:3
2 tures of Toronto as recognized by Environment Canada [29,30].
Fig. 6 shows that the switching point during off-peak hours is
3 C. In other words, when the outdoor temperature (during the
Table 1
off-peak hours) is lower than 3 C, each unit of thermal energy
Compering the increment rates of TOU electricity prices. produced by mini boiler is less expensive than ASHP. The switching
points at mid- and on-peak hours are 4 C and 11 C, respectively.
TOU electricity prices (/kW h)
It should be noted that these optimal switching points will change
Year Off-peak Mid-peak Peak whenever one or more of the influencing factors, such as electricity
2014 11.7 15.4 18.0 TOU prices, changes.
2015 13.3 17.8 22.6 Fig. 7 illustrates the flowchart of SDFSS-SPM used in this study
Increment rates 12.0% 13.4% 20.3%
to determine the switching points in different hours. The HVAC
N. Alibabaei et al. / Applied Energy 185 (2017) 2943 33
Peak Hours
3 0 4 11
21 34
Outdoor Temperature ( )
Start
No
Yes
If (1 i 6 or 20 i 24)
Run mini boiler && Run ASHP
Outdoor Temp. 3 C
No
Yes
If (7 i 11 or 17 i 19)
Run mini boiler && Run ASHP
Outdoor Temp. 11 C
No
Yes
If (12 i 16 )
Run mini boiler && Run ASHP
Outdoor Temp. 4 C
system Daily Energy Cost (DEC) is minimized by modifying the mentioned before in Section 2.1.5, the efficiency of SDFSS-SPM
operational command matrix generated in the baseline scenario increased by growing the annual rates of electricity prices.
when no SPM is implemented in the system based on the results
of the SDFSS strategy planning model.
2.2.2. HVAC system energy cost reduction using load shifting strategy
planning model (LSH-SPM)
2.2.1.4. Calculating the impact of SDFSS-SPM on HVAC system DFC at Buildings are complicated/complex entities. Various parameters
different outdoor temperatures. Based on the simulation results, the including, but not restricted to, building construction/material,
energy cost saving rate continuously increases as the outdoor tem- outdoor disturbances such as solar irradiance, outdoor tempera-
perature decreases. This is because ASHP COP, the most important ture, indoor disturbances (such as the heat generated by lighting,
parameter with considerable effect on energy cost saving, varies appliances and solar gains), thermal mass, equipment capacity,
with changes in outdoor temperature (OT). infiltration and occupant behavior significantly impact the result
As Fig. 4 shows, COP changes between 1.64 and 5.31. It should of various strategy planning models and particularly load shifting
be taken into consideration that the rate of reduction of COP models implemented in buildings. Since buildings constitute a sig-
directly affects the rate of energy cost saving. nificant part of the network energy system, their behavior has
Fig. 8 shows the energy cost saving by using SDFSS-SPM (based remarkable influence on the total network energy system. For
on 2014 and 2015 electricity prices) for different daily average out- example, the overload on energy network during peak hours can
door temperatures. As the figure shows, when the outdoor temper- be notably decreased by shifting the HVAC load of buildings from
ature is colder, SDFSS strategy planning model is more effective in peak to off-peak hours. To do the load shifting, thermal energy
energy cost saving. For example, when the daily average outdoor should be stored in the house during off-peak hours. Outdoor tem-
temperature is around 20 C, SDFSS-SPM brings $5.28 and $8.13 perature is the only parameter significantly affecting the thermal
of saving respectively using 2014 and 2015 electricity prices. As energy stored in a house. When it is not very cold/hot outside, ther-
34 N. Alibabaei et al. / Applied Energy 185 (2017) 2943
Fig. 8. Energy cost saving using SDFSS-SPM in House A in different outdoor temperatures.
mal energy can be stored in the house for immediate future use. temperature exceeded its maximum permitted temperature (this
However, in cold/hot and extremely cold/hot weather, this is less particular time is called saturation time in this study). The time
likely. It should also be taken into consideration that house compo- constant (TC) factor obtained by subtracting saturation point and
nent/construction, wall layers/thickness, orientation, thermal rise up times was used to calculate the pre-heating starting time.
mass, windows size and interior design have substantial impacts TC factor ensures that maximum thermal energy is stored in the
on stored thermal energy capacity and consequently the behavior house without compromising thermal comfort.
of the house in cold/hot and very cold/hot outdoor temperature. In Ontario, in winter, peak hours start at 7:00 am and end at
When designing a LSH-SPM model, the occupants thermal 11:00 am. To support the local grid as a grid-friendly house, the
comfort is the most important item that should be taken into con- HVAC system should ideally remain off during this period. This
sideration. Based on the ASHRAE Standard, to ensure thermal com- ideal HVAC operating scenario called Best Case Scenario (BCS) in
fort, indoor temperature should be kept between 2024 C in this study. To implement the LSH-SPM, first an operational com-
winter season and 2226 C in summer season during the occu- mand matrix is built to fulfill the BCS condition. To this end, the
pied/active periods [21]. corresponding elements of the matrix are set to 0 to keep the
HVAC system off during peak hours. One of the most important
2.2.2.1. Methodology. As mentioned in Section 2.1.4, 5-min is parameters preventing from the BCS is the outdoor temperature.
selected as the simulation time steps for TRNSYS program. How- If the outdoor temperature is cold or very cold, HVAC system
ever, based on the simulation result with various outdoor temper- should be turned on during peak hours to ensure the ASHRAE Stan-
atures, this time step is considered not long enough to store dard. In other words, in this case BCS cannot be implemented into
sufficient thermal energy in House A in order to exploit LSH- the system due to the strong effect of the severe outdoor temper-
SPM. In other words, with this time step, the zone (1st floor) tem- ature. To investigate, TRNSYS model is run based on a given day
perature drops quickly when the HVAC system is shut down. In weather condition. Simulation results show that if the ZT remains
addition, simulation results also show that since LSH-SPM is within the ASHRAE Standard range (in each hour), BCS is selected
intended to be used during peak hours with high electricity price, as the certified operational command matrix. However, if at a
keeping the HVAC system on for more than 15 min increases the specific time step ZT drops below the ASHRAE standard range,
ASHP energy demand, which is not ideal according to the philoso- the corresponding element of operational command matrix will
phy of load shifting strategy planning model. Therefore, in order to be set to 1 by Matlab to turn the HVAC on during that specific
improve LSH-SPM efficiency, 15 min was selected as the HVAC sys- time step. Then the TRNSYS model is re-run with modified opera-
tem operating time (time step) during the peak hours. tional command matrix to observe the effect of outdoor tempera-
ture on ZT in the next time step. This process is repeated to
cover the whole day. As the result, at the end of the process, the
2.2.2.2. Pre-heating starting time (winter season). As mentioned most appropriate operational command matrix that met the ASH-
before, there is no opportunity to shift the HVAC load from peak RAE Standard is built using this strategy planning model. Fig. 9
to off-/mid-peak hours without using pre-heating process. One of illustrates the abovementioned procedure in a simple flowchart.
the most important issues that should be taken into consideration
before pre-heating the house is pre-heating starting time. Different
parameters affect the optimal starting time of pre-heating process; 2.2.2.3. Calculating the effect of LSH-SPM on Daily Energy Cost (DEC)
outdoor temperature is the most important one, and characteris- in different outdoor temperatures. To study the impact of LSH-SPM
tics of the house and zone temperature before starting the pre- on HVAC system energy cost saving, different simulations are run
heating process are the 2nd and 3rd important parameters. based on various outdoor temperatures. Based on the simulation
If pre-heating starts earlier than its optimum time, more elec- results, when the daily average outdoor temperature is warm
tricity is consumed and consequently the rooms temperature (equal to or greater than 14 C), ASHP remains off almost the entire
exceeds the ASHRAE Standard range. If pre-heating starts later time due to the low thermal demand of the house. Hence, DEC is
than its optimum time, the thermal energy stored in the house increased by using LSH-SPM. When the daily average outdoor tem-
would not be adequate to fully support LSH-SPM during the peak perature is cold (lower than 0 C), DEC after applying LSH-SPM is
load hours. In this study, the TRNSYS model of House A was run still higher than baseline control scenario when no SPM is used.
at various outdoor temperatures to find the optimum starting time. Fig. 10 shows the energy cost savings for different daily average
To this end, the HVAC set-point was first set to 20.7 C (the heating outdoor temperatures after using LSH-SPM. These energy cost sav-
set-point temperature selected by TRNSYS model developers). ings were calculated based on 2014 and 2015 electricity prices. As
Then, when the HVAC system was turned on (this particular time Fig. 10 shows, LSH-SPM is beneficial when the daily average out-
is called rise-up time in this study), it remained on until the zone door temperature changes between 0 C and 12 C. Like SDFSS-
N. Alibabaei et al. / Applied Energy 185 (2017) 2943 35
Matlab Controller (LSH SPM) planning model will not be advantageous. Hence, the only strategy
planning model that could be implemented during summer season
BCS
Operaonal
is LSH-SPM. In summer season, pre-cooling method should be used
Command Matrix to cool down the zone temperature during mid-peak hours. Based
on summer season TOU pricing scheme, peak hours start from
11:00 am and end at 17:00 pm. Since in summer season, pre-
Run TRNSYS cooling process should be implemented during mid-peak hours,
Program the impact of LSH-SPM on energy cost saving is not as strong as
it is in winter when pre-heating process takes place during off-
peak hours with lower electricity prices.
The methodologies described in Section 2.2.2 are also used for
Check if Zone Temperature Ensure
pre-cooling the house in summer season. In summer season,
the ASHRAE Standard 23.5 C is selected as the zone temperature set-point. Based on
Yes ASHRAE Standard [21], during summer season zone temperature
can only change between 22 C and 26 C. These boundaries are
No set in Matlab program in order to manage and control the imple-
mentation of LSH-SPM. In other words, zone temperature is
reduced to 22 C before peak hours start (i.e., at 10:00 am) to store
cooling energy in the house and subsequently to keep the HVAC
Revise
system off during peak hours.
Operaonal
Command Matrix Figs. 11 and 12 show the zone temperature and operational
command matrix, respectively, before implementing LSH-SPM.
Figs. 13 and 14 show the zone temperature (ZT) and operational
command matrix, respectively, at the first iteration of LSH-SPM. As
Cered the figure shows, first the zone temperature was reduced to 22 C
Operaonal (for pre-cooling) and then the HVAC system was kept off during
Command the peak hours to reduce the energy cost.
Matrix
When the zone temperature during peak hours is higher than
the upper range permitted by the ASHRAE Standard (temperature
Fig. 9. The operational mechanism of LSH-SPM model.
violation), the Matlab controller will turn on the HVAC system in
that particular time step in the next iteration. Since the impact of
LSH-SPM on saving money during summer were not significant,
the simulations were not run for the cooling season and hence
the figure shown in Section 2.2.2.3 (Fig. 10) were not generated.
Zone Temperature (
The weather file used to simulate the thermal demand is the 3.1.2. Methodology
metropolitan Toronto weather provided in the TRNSYS library. The process system and strategy planning models described in
Fig. 16 depicts the hourly outdoor temperature during the heating Section 2 are used in this section for analyzing the behavior of
season. HVAC system in daily basis during winter season. 2015 electricity
Based on Fig. 16, the minimum outdoor temperature for prices presented in Section 2.1.5 are used in energy cost
Toronto is 22.27 C and occurs on January 12th (273 h). calculations.
N. Alibabaei et al. / Applied Energy 185 (2017) 2943 37
Generate Final
Run Implement Implement Operaonal
TRNSYS LSH SPM SDFSS SPM Command
Matrix
Fig. 18. HVAC system Daily Energy Cost (DEC) and energy cost saving based on each
strategy planning model - December 1st.
Fig. 17. Outdoor temperature on December 1st TRNSYS metropolitan weather data.
38 N. Alibabaei et al. / Applied Energy 185 (2017) 2943
Fig. 19. Outdoor temperature on January 1st TRNSYS metropolitan weather data.
Fig. 21. Outdoor temperature on February 1st TRNSYS metropolitan weather data.
the fact that LSH-SPM has the potential to reduce the operating
cost of HVAC system (when the daily average outdoor temperature
changes between 0 C and 12 C), the HVAC system operating cost
increased by a total of 9.11% when this strategy planning model
was implemented for the entire heating season. The operating cost
of HVAC system decreased significantly with implementation of
SDFSS-SPM in the heating season. This fuel switching strategy
planning model reduced the HVAC system operation cost by 23.8%.
The last strategy planning model implemented on the system
was LSHSDFSS which reduced the HVAC system operating cost
by 15.8%.
Fig. 22. HVAC system Daily Energy Cost (DEC) and energy cost saving based on each
strategy planning model - February 1st.
3.1.5. Optimal strategy planning model
As described before, there is a strong relationship between the
3.1.4. Energy cost saving analysis by utilizing different SPMs in the effectiveness of each strategy planning model and the weather
heating season condition, particularly the outdoor temperature. As a result, on
In this section, the influence of each strategy planning model in cold (or extremely cold) weather condition, the optimal strategy
the heating season is analyzed. Fig. 25 depicts the DEC of HVAC planning model for minimizing the HVAC system operating cost
system with different strategy planning models. In this figure, sim- is SDFSS-SPM. When daily average outdoor temperature changes
ulation starts on January 1st and ends on December 31th, contain- between 0 C and 12 C, LSH-SPM can be selected as the optimum
ing 233 days (the cooling season days are excluded). The red1 line strategy planning model. LSHSDFSS-SPM takes advantage of both
shows the energy cost of HVAC system when no strategy planning load shifting and fuel switching. This strategy planning model
model is implemented on the system (baseline control scenario). can be used as the optimum SPM on certain days when the daily
The second line (the green line) shows the energy cost of HVAC sys- average outdoor temperature changes between 0 C and 8 C. Con-
tem with LSHSDFSS-SPM. The third line (the blue line) shows the sidering these specifications, an optimization process is executed
energy cost of HVAC system with LSH-SPM. The last line (the black to determine the best SPM, i.e., the one that generates the mini-
line) shows the energy cost of HVAC system with SDFSS-SPM. mum HVAC system daily energy cost, on each specific day.
As it can be concluded from Fig. 25, when the outdoor temper- Fig. 27 depicts the minimum daily operating cost of HVAC sys-
ature is cold and/or very cold (i.e., during winter months), the DEC tem after running optimization process for the heating season. The
of HVAC system increased using the LSH-SPM. However, during overall operating cost of HVAC system in the heating season after
this period, the DEC of HVAC system decreased significantly when utilizing the optimization process is $716.08. This optimum/mini-
using SDFSS-SPM. LSHSDFSS-SPM took the second place in reduc- mum cost is lower than the HVAC system operating cost of $743.79
ing the DEC of HVAC system during these cold days. When the out- with SDFSS-SPM (as the most effective SPM during the heating sea-
door temperature is not very cold (i.e., at the beginning and/or the son) because on some winter days, LSH or LSHSDFSS SPM gener-
end of the heating season - the middle section of the graph in ated lower energy cost for HVAC system operation due to the
Fig. 25) there was no noticeable difference in energy cost saving specific weather conditions. The implementation of the fully opti-
using different SPMs. mized HVAC SPM achieves a 26.6% energy cost reduction compared
Fig. 26 illustrates the energy cost saving with different strategy to the base case control for a typical Toronto heating season.
planning models for the heating season. The total energy cost for
the baseline scenario is $976.10. The HVAC operating cost 3.2. Impact of the load shifting strategy planning model on HVAC
increased to $1065.03 using LSH-SPM. In other words, despite system during summer (the cooling season)
Outdoor Temperature ( )
Hour
Fig. 23. The outdoor temperature on the coldest day of year TRNSYS metropolitan weather data.
Fig. 26. Energy cost saving with different strategy planning models in the heating
season.
3.2.2. Methodology
Fig. 24. HVAC system Daily Energy Cost (DEC) and energy cost saving based on each The process system and LSH strategy planning model described
strategy planning model on the coldest day of the year.
in Section 2.2.2 are used in this section for analyzing the behavior
is set to 23.5 C during summer (the cooling) season. The of HVAC system in the cooling season.
metropolitan Toronto weather file available at the TRNSYS library
was used to simulate the thermal demand. 3.2.3. Impact of the LSH-SPM on HVAC system demand and energy cost
Fig. 28 depicts the hourly outdoor temperature during the cool- in the cooling season
ing season. The maximum outdoor temperature was 33.96 C and Since outdoor temperature during the cooling season is always
was recorded on July 23th (1433 h). above 0 C, SDFSS and consequently LSHSDFSS strategy planning
26
24
HVAC System Daily Energy Cost ($)
22
20
18
16
14
12
10
08
06
04
02
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230235
Different Days of Heating Season
Fig. 25. DEC of HVAC system based on different strategy planning models in the heating season.
N. Alibabaei et al. / Applied Energy 185 (2017) 2943 41
14
13
12
11
10
HVAC System DEC ($)
09
08
07
06
05
04
03
02
Fig. 27. Minimum daily operating cost of HVAC system - heating season days.
Fig. 29. DEC of HVAC system before and after taking advantage of LSH-SPM in the cooling season.
42 N. Alibabaei et al. / Applied Energy 185 (2017) 2943
[14] Patteeuw D, Gregor P, Helsen L. Comparison of load shifting incentives for low- [23] Safa A, Fung A, Kumar R. Performance of two-stage variable capacity air source
energy buildings with heat pumps to attain grid flexibility benefits. Appl heat pump: field performance results and TRNSYS simulation. Energy Build
Energy 2016;167:8092. 2015;94:8090.
[15] Beizaee A, Allison D, Lomas K, Foda E, Loveday D. Measuring the potential of [24] Dembo A, Fung AS, Ng KLR, Pyrka A. The archetype sustainable house:
zonal space heating controls to reduce energy use in UK homes: the case of un- investigating its potentials to achieving the net-zero energy status based on
furbished 1930s dwellings. Energy Build 2015;92:2944. the results of a detailed energy audit. In: Proceedings of the 1st international
[16] Chassin DP, Stoustrup J, Agothoklis P, Djilali N. A new thermostat for real-time high performance buildings conference (2010) Purdue University, Paper 15.
price demand response: cost, comfort and energy impacts of discrete-time http://docs.lib.purdue.edu/ihpbc/15.
control without deadband. Appl Energy 2015;155:81625. [25] Crawley D, Hand J, Kummert M, Griffith. Contrasting the capabilities of
[17] Li J, Poulton G, Platt G, Wall j, James G. Dynamic zone modelling for HVAC building energy performance simulation programs. Build Environ 2008;43
system control. Int J Model, Identif Control 2010;9(1/2):513. (4):66173.
[18] Radhakrishnan N, Su Y, Su R, Poolla K. Token based scheduling for energy [26] Klein S, Beckman W, Mitchell J, Duffie J, Duffie N, Freeman T. TRNSYS 16 - a
management in building HVAC systems. Appl Energy 2016;173:6779. transient system simulation program (2006, June), Madison, Wisconsin, U.S.A.
[19] Griful S, Jacobsen R, Nguyen D, Sorensen G. Demand response potential of [27] Kamel R, Fung AS. Modeling, simulation and feasibility analysis of residential
ventilation systems in residential buildings. Energy Build 2016;121:110. BIPV/T+ASHP system in cold climateCanada. Energy Build 2014;82:75870.
[20] Di Giorgio A, Liberati F. Near real time load shifting control for residential [28] VIESMANN. Technical data manual, gas-fired wall-mounted condensing boiler
electricity prosumers under designed and market indexed pricing models. 12 to 285 MBH (3.5 to 83.5 kW), Vitodens 200-W Series B2HA.
Appl Energy 2014;128:11932. [29] https://en.796 wikipedia.org/wiki/Geography of Toronto [accessed on: Oct 6th,
[21] ASHRAE. ASHRAE handbook-fundamentals. Atlanta, GA: American Society of 2015].
Heating, Refrigeration and Air-Conditioning Engineers, Inc; 2009. [30] https://en.wikipedia.org/wiki/List_of_extreme_temperatures_in_Canada
[22] Zhang D, Barua R, Fung A. TRCA-BILD archetype sustainable house - overview [accessed on: Oct 6th, 2015].
of monitoring system and preliminary results for mechanical systems. ASHRAE
Trans 2011:597612.