Accepted Manuscript

Seasonal performance evaluation of electric air-to-water heat pump systems

Matteo Dongellini, Claudia Naldi, Gian Luca Morini

PII: S1359-4311(15)00249-5
DOI: 10.1016/j.applthermaleng.2015.03.026
Reference: ATE 6460

To appear in: Applied Thermal Engineering

Received Date: 1 December 2014

Accepted Date: 18 March 2015

Please cite this article as: M. Dongellini, C. Naldi, G.L. Morini, Seasonal performance evaluation
of electric air-to-water heat pump systems, Applied Thermal Engineering (2015), doi: 10.1016/
j.applthermaleng.2015.03.026.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

SEASONAL PERFORMANCE EVALUATION OF ELECTRIC AIR-TO-

WATER HEAT PUMP SYSTEMS

PT
Matteo DONGELLINI1,*, Claudia NALDI1 and Gian Luca MORINI1

1
Dipartimento di Ingegneria Industriale DIN,

RI
Alma Mater Studiorum Università di Bologna,

SC
Viale del Risorgimento 2, 40136 Bologna, Italy

U
AN
M
D

*Corresponding author:
TE

Tel: +39 051 2090467

fax: +39 051 2093296

EP

e-mail: matteo.dongellini@unibo.it

Co-authors mail: claudia.naldi2@unibo.it

C

gianluca.morini3@unibo.it
AC
 ACCEPTED MANUSCRIPT

ABSTRACT

A numerical model for the calculation of the seasonal performance of different kinds of electric

air-to-water heat pumps is presented. The model is based on the procedure suggested by the

European standard EN 14825 and the Italian standard UNI/TS 11300-4, which specify the

PT
guidelines for calculation of the seasonal performance of heat pumps during the heating season

RI
(SCOP), the cooling season (SEER) and for the production of domestic hot water. In order to

consider the variation of outdoor conditions the developed model employs the bin-method.

SC
Different procedures are proposed in the paper for the analysis of the seasonal performance of

mono-compressor, multi-compressor and variable speed compressor air-to-water heat pumps.

U
The numerical results show the influence of the effective operating mode of the heat pumps on
AN
the SCOP value and put in evidence the impact of the design rules on the seasonal energy

consumption of these devices. The study also highlights the importance of the correct sizing of
M

the heat pump in order to obtain high seasonal efficiency and it shows that, for a fixed thermal
D

load, inverter-driven and multi-compressor heat pumps have to be slightly oversized with respect
TE

to mono-compressor ones in order to obtain for the same building the highest SCOP values.
C EP
AC
 ACCEPTED MANUSCRIPT

1. INTRODUCTION

The Directive 28/2009/CE [1] has indicated for European countries a series of targets

concerning energy efficiency and the use of renewable energy, like a greenhouse gas

reduction of 20% and the increase of the renewable energy quote over the total energy

PT
consumption of 20%. Buildings like residential houses, offices and shops are

responsible for approximately 40% of final energy consumption and for 36% of

RI
greenhouse gas (GHG) emissions in the European Union [2]. Due to its relevance, the

SC
European Commission is focusing its action in the improvement of buildings energy

efficiency by means of the EPBD Recast Directive [3] that indicates for the Member

U
States the target of the transition to Nearly Zero Energy Buildings (NZEB) within 2020.
AN
In these buildings the reduction of energy needs through the improvement of the

building envelope thermal performances and of the heating, cooling and ventilating
M

systems is strongly pursued. Heat pumps represent a useful tool to achieve European
D

targets because, since 2009, the RES Directive [1] recognizes aero-thermal, geothermal

and hydrothermal energy as renewable energy sources.

TE

Air-to-water heat pump devices are good candidate for the replacement and/or
EP

integration of gas boilers but the correct evaluation of the seasonal performance of these

systems can be problematic for designers. In fact, the performance of air-to-water heat
C

pumps is strongly variable during the season due to the dependence on the outdoor
AC

temperature which changes continuously in time. During the whole heating or cooling

season heat pump performances (i.e. their energy consumption) are strongly influenced

by a series of external factors, like the variable building thermal load, the outdoor

climate and the kind of control system of the heat pump. This last aspect becomes

crucial because heat pumps work at nominal conditions only for a limited part of the
 ACCEPTED MANUSCRIPT

season and for this reason in the calculation of the seasonal efficiency of these devices

the behavior of heat pumps at partial loads must be carefully considered. Bettanini et al.

[4] and Henderson et al. [5] have deeply investigated the importance of a correct

evaluation of heat pump behavior at partial loads and they have concluded that the

PT
seasonal performances of these devices are strongly influenced by the capacity of the

heat pumps to maintain high values of coefficient of performance (COP) at partial

RI
loads. Nevertheless for a designer this analysis can become complicate due to the

SC
lacking of accurate data about the behavior of heat pumps far from the nominal

conditions; in fact, large part of the heat pump manufacturers still avoid to give detailed

U
information about this point on their data sheets.
AN
Many researchers have analyzed the problem of the rigorous calculation of heat pump

seasonal efficiency (SCOP) for air-to-water heat pumps [6-9]. The present work is
M

aimed to describe a mathematical model developed in order to calculate the seasonal

D

performances of different kinds of heat pump systems, like mono-compressor on-off

heat pumps, multi-compressor heat pumps and inverter-driven heat pumps. This topic
TE

has not yet been clarified completely in the open literature and in technical standards.
EP

The calculation of the heat pump seasonal performance is usually carried out through a

quasi steady-state evaluation, in which the time step is fixed equal to 1 month for heat
C

pumps linked to stable thermal reservoirs (i.e. water or ground); on the contrary, for an
AC

air-to-water heat pump, in order to take into account the variability of the outside

temperature, the bin-method is generally recommended for the evaluation of the

seasonal heat pump performances [10-13]. The numerical model presented in this paper

is able to calculate the seasonal performances of different air-to-water heat pumps

(mono-compressor, multi-compressor or inverter-driven). The model is used in order to

 ACCEPTED MANUSCRIPT

investigate in which way the seasonal performance of different kinds of air-to-water

heat pumps is influenced by the design rules adopted when the heat pump is coupled to

a building. The results shown in this paper point out that the heat pump sizing is an

important aspect which must be taken into account in order to obtain the maximum

PT
energy performance of a heating system based on an air-to-water heat pump and it is

demonstrated that the optimal sizing rules change with the kind of heat pump

RI
considered.

SC
2. MATHEMATICAL MODEL

U
In this section the numerical model for the evaluation of the seasonal performances of
AN
electric air-to-water heat pumps, based on a vapor compression cycle, coupled with a

residential building is described. The European standard EN 14825 [14] and the Italian
M

standard UNI/TS 11300-4 [15] suggest to use the bin-method in order to calculate the
D

heat pump seasonal performances, in which bin represents the number of hours in

correspondence of which the outdoor temperature drops within a fixed temperature

TE

interval having a broadening of 1 K and centered on an integer value of temperature.

EP

While the standard EN 14825 splits Europe in three winter climates (i.e. Colder,

Average and Warmer) and directly provides the bin seasonal values for each climate,
C

the standard UNI/TS 11300-4 suggests for the heating season a bin calculation method
AC

based on a normal outdoor temperature distribution, obtained starting from the local

data of monthly average outdoor temperature, outdoor design temperature and monthly

average daily solar radiation on a horizontal plane (data available for Italy from

standards UNI 10349-1 [16] and UNI EN 12831 [17]). On the other hand, it is possible
 ACCEPTED MANUSCRIPT

to derive the bin distribution of a specific location by using the hourly outside

temperature of the Test Reference Year (TRY) for the fixed location.

2.1 Bin distribution and building energy signature

PT
In the present paper the bin-method defined by the UNI/TS 11300-4 is used to evaluate

the hourly bin distribution for Bologna (44.47°N, 11.43°E) during the conventional

RI
heating season, which starts on October 15th and ends on April 15th of each year. The

SC
bin distribution shown in Fig. 1 has been obtained by considering that the outdoor

design temperature (Tdes) for Bologna is -5°C, as indicated by UNI EN 12831. By

U
observing the distribution shown in Fig. 1 it is possible to note that the minimum
AN
outdoor temperature which occurs in Bologna is equal to -4°C and the mode of this

distribution is equal to 6°C.

M

FIGURE 1 HERE
D

For the analysis of the energy interaction between a building and the coupled heat pump

it is convenient to introduce the Building Energy Signature (BES), which is defined as

TE

the thermal power required by the building at the outlet of the generation sub-system
EP

(heat pump+ back-up system) as a function of the outdoor temperature (Text). The use of

the BES is suggested by the standards EN 14825 and UNI/TS 11300-4 and the
C

methodology for the determination of the BES is introduced by Fels et al [18] and
AC

described in the standard EN 15603 [19].

The outdoor temperature in correspondence of which the building heating demand

becomes zero is called Heating Limit External Temperature (HLET). When a building

can be characterized by means of a linear BES curve, the value of HLET and of the

design load (Pdes) in correspondence of the outdoor design temperature are the only
 ACCEPTED MANUSCRIPT

parameters which are mandatory to know in order to draw the BES. In Fig. 2 a linear

BES is represented with a dashed red line drawn by considering a design thermal load of

the building of 40 kW (Pdes) when the outdoor temperature is equal to the outdoor

design temperature (for Bologna Tdes=-5°C) and a value of HLET equal to 16°C.

PT
The heating power required by the building (Pb) as a function of the outdoor

temperature can be written as follows:

RI
 HLET − Text (i) 
Pb (i ) = Pdes   (1)
 HLET − Tdes 

SC
where Pb(i) is the thermal power required by the building in the i-th bin.

U
The building thermal energy demand Qb is evaluated in correspondence of each i-th bin
AN
through Eq. (2):

Qb (i ) = Pb (i ) tbin (i ) (2)
M

where tbin (i) is the duration of the i-th bin.

D

FIGURE 2 HERE
TE

2.2 Heat pump characterization

EP

In the same (Text, P) chart reported in Fig. 2 it is possible to draw the characteristic

curve of an air source heat pump by considering that the thermal power delivered by the
C

heat pump (PHP) depends on the outdoor temperature, for a fixed value of the hot water
AC

temperature (Tw) produced by the heat pump. The characteristic curve associated to the

heat pump can be obtained by the technical data sheets given by the heat pump

manufacturer. However, these characteristic curves change if different kinds of electric

air-to-water heat pumps are considered, like mono-compressor ON-OFF Heat Pumps

(ON-OFF HPs), Multi-Compressor Heat Pumps (MCHPs) and Inverter-Driven Heat

 ACCEPTED MANUSCRIPT

Pumps (IDHPs). In fact, for a mono-compressor heat pump (ON-OFF HP) the thermal

power delivered by the heat pump is a function of the temperature of the two sources

(air: Text; water: Tw) between which the heat pump works (PHP=f(Text,Tw)); on the

contrary, for MCHPs the thermal power delivered by the heat pump depends also on the

PT
number of compressors switched on (n) (PHP=f(Text,Tw,n)) and for IDHPs the delivered

power is a function of the inverter frequency (Ф) (PHP=f(Text,Tw,Φ)). The model

RI
presented in this paper considers a fixed value of the hot water temperature (Tw)

SC
produced by the heat pump: this situation is really met when the heat pump is coupled

to a radiant floor heating loop which works with a constant inlet water temperature

U
during the whole heating season (typically 35-40°C). This means that mono-compressor
AN
heat pumps are represented by a single curve in the chart (Text, P); on the contrary,

MCHPs are represented by N curves (with N equal to the number of the heat pump
M

compressors) and IDHPs by a family of curves obtained by varying the inverter

D

frequency between the maximum (Фmax) and minimum (Фmin) value. In Fig. 2a a typical
TE

characteristic curve of a mono compressor ON-OFF heat pump is shown together with

the BES defined before; in this case BES and heat pump characteristic curve have only a
EP

common point, called balance point, in correspondence of which the heating power

delivered by the heat pump equals the heating demand of the building. The outdoor
C

temperature linked to the balance point is called bivalent temperature (Tbiv). When the
AC

outdoor temperature is lower than Tbiv the building heating demand cannot be satisfied

by the heat pump and an integration system must be activated (i.e. electric heaters); on

the contrary, when the outdoor temperature is higher than Tbiv the delivered heat pump

heating power exceeds the building thermal request and on-off cycles need to start in

order to match the energy demand.

 ACCEPTED MANUSCRIPT

In Fig. 2b the same graph is drawn for MCHPs and IDHPs; the two characteristic curves

drawn in Fig. 2b correspond to the maximum and the minimum heating power

deliverable by:

• a MCHP when the maximum number of compressors are switched on (N/N) or

PT
when only 1 compressor over N is switched on (1/N);

• an IDHP when the inverter frequency is set to its maximum value (Фmax) or to its

RI
minimum value (Фmin).

SC
In the cases of MCHPs and IDHPs Fig. 2b shows that it is possible to define, in addition

to the bivalent temperature (Tbiv), a secondary bivalent temperature (Tbiv,2), which is the

U
maximum outdoor temperature that the heat pump can manage without starting on-off
AN
cycles.

In Fig. 2 the heat pump characteristic curves are stopped in correspondence of the value
M

of Text for which the heat pump is switched off (Temperature Operative Limit, TOL):
D

this parameter is generally declared by the heat pump manufacturer.

In addition to the knowledge of the characteristic curves of the heat pump drawn in Fig.
TE

2, for a complete characterization of the heat pump it is mandatory to know the value of
EP

the air-to-water heat pump COP at declared capacity (COPDC), in correspondence of

several values of the outdoor air temperature and of specific indoor heat exchanger
C

inlet/outlet water temperature (Tw). COPDC input data in correspondence of the

AC

activation of each compressor (MCHPs) or in correspondence of the maximum, the

minimum and at least three additional intermediate inverter frequencies (IDHPs) are

also needed.
 ACCEPTED MANUSCRIPT

A simple mathematical model of the heat pump can be built by interpolating the heat

pump thermal power (PHP) and COPDC data given by the manufacturer as a function of

the outdoor temperature using second-order polynomial functions.

For mono-compressor heat pumps (ON-OFF HPs) it is possible to write:

PT
PHP (i) = a1 (Tw ) Text (i) + b1 (Tw ) Text 2 (i) + c1 (Tw )
(3)
COPDC (i) = a2 (Tw ) Text (i) + b2 (Tw ) Text 2 (i) + c2 (Tw )

RI
In this way PHP and COPDC become functions of the i-th bin considered.

SC
The six coefficients a1, a2, b1, b2, c1, c2 are functions of the hot water temperature for a

mono-compressor ON-OFF HP and their values fully define the characteristics of the

U
heat pump.
AN
For multi-compressor heat pumps (MCHPs) the situation is more complicated because

the heat pump thermal power and COP change if the number of activated compressors
M

changes; this means that the six coefficients a1, a2, b1, b2, c1, c2 depend also on the

number of compressors switched on. If n is the number of activated compressors and N

D

is the total number of the heat pump compressors, one can write:
TE

PHP,n / N (i ) = a1,n / N (Tw ) Text (i) + b1,n / N (Tw ) Text 2 (i ) + c1,n / N (Tw )
with n=[1, N] (4)
COPDC ,n / N (i ) = a2,n / N (Tw ) Text (i) + b2,n / N (Tw ) Text 2 (i ) + c2, n / N (Tw )
EP

where the notation n/N means that the corresponding quantity is evaluated by
C

considering n compressors switched on among the N compressors of the heat pump. In

AC

this way, a MCHP with N compressors can be completely characterized by means of

6xN coefficients.

The situation is similar for the characterization of an inverter-driven heat pump (IDHP);

in this case the heat pump power and COP change with the inverter frequency, as well

as with the hot water temperature. By fixing M values of frequency from Фmin to Фmax,
 ACCEPTED MANUSCRIPT

declared by the manufacturer, one can calculate the values of the heat pump thermal

power and COP by knowing 6xM coefficients:

PHP ,Φ j (i ) = a1,Φ j (Tw ) Text (i ) + b1,Φ j (Tw ) Text 2 (i ) + c1,Φ j (Tw )

with j=[1, M] (5)
COPDC ,Φ j (i ) = a2,Φ j ( Tw ) Text (i ) + b2,Φ j (Tw ) Text 2 (i ) + c2,Φ j (Tw )

PT
The values of the coefficients recalled in Eqs. (3-5) have to be obtained by interpolating

the data given by the heat pump manufacturer.

RI
SC
2.3 Energy calculations

Once selected the building and the heat pump, the capacity ratio CR for each bin can be

U
calculated by using Eq. (1) and Eqs. (3-5):
AN
CR(i) = Pb (i) / PHP (i) (6)

For MCHPs and IDHPs the value of CR is calculated by considering PHP as the heat
M

pump capacity with all the compressors activated (i.e. PHP,N/N(i)) or in correspondence
D

of the maximum value of the inverter frequency (i.e. PHP,Фmax(i)).

TE

If CR turns out to be greater than 1, as in the case of bins with a value of Text lower than

Tbiv, CR is set equal to 1.

EP

The thermal energy delivered in the i-th bin by the heat pump is calculated as follow:

QHP (i) = PHP (i) tbin (i) CR(i) (7)

C
AC

The heat pump COP at declared capacity, COPDC, is evaluated for each bin by means of

Eq. (3-5); the actual value of the COP, COPPL, which takes into account the losses

linked to the on-off condition, is obtained as follows:

COPPL (i) = COPDC (i) fCOP (i) (8)

where fCOP is the COP correction factor for on-off condition:

 ACCEPTED MANUSCRIPT

CR(i )
fCOP (i ) = (9)
1 − Cc + Cc CR (i )

Cc is the degradation coefficient ([14, 15]) of the heat pump, a value experimentally

quantified by the manufacturer. In absence of specific indications, [14] and [15] suggest

to use a value of Cc equal to 0.9.

PT
For MCHPs and IDHPs the value of the COPPL is evaluated by using the value of the

RI
COPDC obtained in correspondence of the activation of only one compressor (COPPL,1/N)

or in correspondence of the minimum inverter frequency (COPPL,Фmin).

SC
The COP correction factor for MCHPs and IDHPs is calculated with the capacity ratio

U
corresponding to the heat pump capacity with only one compressor activated or at the
AN
minimum inverter frequency. In fact, for a MCHP or IDHP the on-off condition starts

when the energy needed by the building becomes lower than the energy that the heat
M

pump would deliver with only one compressor activated or at the minimum inverter

frequency (Text>Tbin,2).
D

For a mono-compressor heat pump, the electric energy EHP,us (i) used by the heat pump
TE

in the i-th bin is:

E HP ,us (i ) = QHP (i ) / COPPL (i )

EP

(10)

For a MCHP, in order to evaluate the electric energy used by the heat pump, for each
C

bin it is mandatory to know how many compressors are activated and how long. In fact,
AC

if the thermal power required by the building is higher than the heat pump capacity

corresponding to the activation of n compressors, but is lower than the heat pump

capacity corresponding to the activation of n+1 compressors, then n+1 compressors are

activated for a certain period of time and n compressors for the remaining time, so as

the total energy delivered by the heat pump equals the building energy demand.
 ACCEPTED MANUSCRIPT

As an example, if in the i-th bin a MCHP with 2 compressors (N=2) has to provide an

amount of energy intermediate between the energy which would be delivered with 1

compressor and with 2 compressors working for the entire bin duration, then the period

of time with 2 working compressors (tbin,2/2) can be estimated as follows:

PT
Qb (i ) − PHP ,1/ 2 (i ) tbin (i )
tbin ,2/ 2 (i ) = (11)
PHP ,2/ 2 (i ) − PHP ,1/ 2 (i )

RI
Consequently, the period of time in which the MCHP works with only 1 compressor

switched on (tbin,1/2 ) is equal to:

SC
tbin ,1/ 2 (i ) = tbin (i ) − tbin ,2/2 (i ) (12)

U
Then the electric energy consumption of the MCHP is evaluated as:
AN
 QHP (i)
 if Qb (i) < PHP,1/2 (i) tbin (i)
COPPL (i)
 P (i) t (i) P (i) t (i)
M

EHP,us (i) =  HP,2/2 bin,2/2 + HP,1/2 bin,1/2 if PHP,1/2 (i) tbin (i) ≤ Qb (i) < PHP,2/2 (i) tbin (i) (13)
 COPDC,2/2 (i) COPDC,1/2 (i)
 Q (i)
 HP
if Qb (i) ≥ PHP,2/2 (i) tbin (i)
D

 COPDC,2/2 (i)
TE

For an IDHP, in order to evaluate the electric energy used by the heat pump, for each

bin it is mandatory to know the actual value of the inverter frequency (Фeff). In fact, for
EP

values of the outdoor temperature between Tbiv and Tbiv,2, Фeff is the value of the inverter

frequency at which the power supplied by the heat pump equals the thermal power
C

required by the building. Once calculated Φeff through interpolations of the

AC

manufacturer data of the heat pump power and frequencies, the actual value of the COP

(COPDC,Φeff) can be evaluated by interpolating the M values of COPDC,Φj as function of

Φeff.

Then the electric energy consumption of the IDHP is evaluated as:

 ACCEPTED MANUSCRIPT

 Q (i)
 HP if Qb (i) < PHP,Φ1 (i) tbin (i)
COPPL (i)

 QHP (i)
EHP,us (i) =  if PHP,Φ1 (i) tbin (i) ≤ Qb (i) < PHP,ΦM (i) tbin (i) (14)
COPDC,Φeff (i)
 Q (i)
 HP
if Qb (i) ≥ PHP,ΦM (i) tbin (i)
COPDC,ΦM (i)

PT
For each kind of heat pump, if the heat pump capacity in the i-th bin is lower than the

RI
building demand (Text<Tbin), the missing energy is delivered by the back-up system (i.e.

electric heaters). The energy used by the electric heaters coincides with the energy they

SC
provide.

EBU (i) = Qb (i) − QHP (i)

U
AN (15)

The mean seasonal COP of the heat pump, SCOPnet, and of the whole system composed

by electric air-to-water heat pump and electric heaters as back-up system, SCOPon, are
M

evaluated according to the following definition ([14,15]):

∑Q ∑ Q (i)
D

HP (i ) b
SCOPnet = i
SCOPon = i
(16)
∑E (i ) ∑[E (i ) + EBU (i )]
TE

HP , us HP ,us
i i

3. CASE STUDY
EP

The developed model is used in order to evaluate the influence of the best choice of the

bivalent temperature (Tbiv) in order to maximize the seasonal performance of different

C

models of air-to-water heat pumps (ON-OFF HP, MCHP, IDHP) coupled with a radiant
AC

floor heating system. Three different models of heat pumps have been selected with

similar rated performances at full load. In Table 1 the main technical characteristics

declared by the manufacturer for the selected heat pumps are quoted. As indicated in

Table 1, all the selected heat pumps are reversible and use R410A as refrigerant. The

ON-OFF HP has one fixed speed scroll compressor, while the MCHP has two identical
 ACCEPTED MANUSCRIPT

fixed speed scroll compressors, connected in parallel, with a size which is about the half

of the ON-OFF HP compressor. On the contrary, the IDHP has a variable speed scroll

compressor driven by an inverter which allows modulating its pressure head.

TABLE 1 HERE

PT
In order to highlight the performance of the three selected heat pumps, in Fig. 3a and Fig.

3b the heating capacity and the electric power input of the selected heat pumps are

RI
reported respectively as a function of the outdoor temperature at full load conditions. The

SC
technical data of the selected heat pumps are given by the manufacturer and they are

collected for a value of the hot water temperature (Tw) produced by the heat pump equal

U
to 35°C, typical inlet value of the water temperature in the water loop of a radiant floor
AN
heating system.

As shown clearly by the data of Fig. 3a and Fig. 3b, for a fixed hot water temperature
M

(Tw=35°C) the three selected heat pumps have the same heating capacity (Fig. 3a) but the
D

electric power input is quite different at the same outdoor temperature (Fig. 3b): the ON-

OFF HP has the lower electric consumption at full load for an outdoor temperature less
TE

than 10°C; on the contrary, the selected IDHP is characterized by the highest electric
EP

energy consumption within the whole range of outdoor temperature considered.

FIGURE 3 HERE
C

However, during the whole heating season, heat pumps are called to work at full load
AC

only for a limited period of time. As shown later, that duration of the period in which the

heat pump is able to work at full load depends on the kind of the heat pump considered

and, of course, by the choice of the bivalent temperature. As a consequence, for a correct

evaluation of the seasonal heat pump performance it is important to take into account the

behavior of the heat pump at partial loads. When the heat pump works at partial loads the
 ACCEPTED MANUSCRIPT

differences among ON-OFF HP, IDHP and MCHP become more and more evident

because these devices react in a different way to the reduction of the thermal loads of the

building.

More in detail, if an ON-OFF HP is considered, at large outdoor temperature, when the

PT
building loads are lower, the heat pump heating capacity can be reduced only by starting

a series of on-off cycles of the compressor. During these unsteady working conditions

RI
the heat pump loses efficiency: for each start and stop there are a series of energy losses

SC
which can be accounted for by means of the Cc coefficient of Eq. (9). The value

suggested for this coefficient by the Italian and European standards [14-15], in absence

U
of specific values given by the manufacturers, is equal to 0.9 but this value in general
AN
tends to underestimate the real energy losses due to the ON-OFF cycles, as observed by

Madonna et al. in [20].

M

On the contrary, MCHP presents a more fine control system in order to reduce the
D

heating power at larger outdoor temperatures. In the case of the selected MCHP, this

device has 2 compressors of the same size connected in parallel, littler than the mono-
TE

compressor of the ON-OFF HP, connected to a single refrigeration loop. When the
EP

building thermal load decreases, the controller unit turns off one compressor and

consequently decreases also the refrigerant mass flow rate within the refrigeration loop.
C

In Fig. 4 the COP trend of the selected MCHP is shown as a function of the outdoor air
AC

temperature and of the number of activated compressors (1/2: red line; 2/2: black line).

FIGURE 4 HERE

Fig. 4 demonstrates that MCHP performance improves when the number of activated

compressors is reduced; in fact, when one compressor is turned off, the refrigerant mass

flow rate in the refrigeration loop is reduced but, since the evaporator and the condenser
 ACCEPTED MANUSCRIPT

are fixed, the total heat transfer area of these heat exchangers is still the same. In these

conditions the thermal efficiency of the evaporator and of the condenser slightly

increases. It is possible to conclude that MCHP is able to work with larger COP values

with respect to an ON-OFF HP at partial load because the starting of the ON-OFF cycles

PT
is strongly delayed and it starts when the outdoor temperature overcomes the value of the

secondary bivalent temperature (Tbiv,2) of the system heat pump-building.

RI
If an IDHP is used the thermal power delivered by the heat pump can be adapted to the

SC
needs of the building in a continuous way by changing the compressor rotating speed by

means of an electronic inverter. In the case of the selected IDHP, the inverter frequency

U
can be varied from a minimum of 30 Hz (minimum compressor rotating speed) to a
AN
maximum of 120 Hz, corresponding to the full load working condition (maximum

compressor rotating speed). When the building thermal load decreases the inverter
M

reduces the compressor rotating speed: in this way the refrigerant mass flow rate within
D

the refrigeration loop and consequently the heat pump heating capacity decrease. Also in

this case, until the outdoor temperature is lower than the secondary bivalent temperature
TE

(Tbiv,2) of the system, the heat pump avoids ON-OFF cycles by following exactly the
EP

building thermal load and in this way efficiency losses at partial loads are strongly

reduced. In addition, when the IDHP compressor operates with a reduced rotating speed
C

its efficiency initially increases. This behavior is illustrated by Fig. 5 in which the trend
AC

of the COP of the selected IDHP is shown as a function of the inverter frequency for a

fixed hot water temperature equal to 35°C (Tw). Each COP curve is obtained for a fixed

value of the outdoor temperature between -18°C and 12°C. Fig. 5 shows that, at full load,

the heat pump works at the lowest COP value but this value reaches a maximum when

the inverter frequency is halved with respect to the maximum frequency (120 Hz). Also
 ACCEPTED MANUSCRIPT

in this case, when the outdoor temperature is larger than the secondary bivalent

temperature the IDHP compressor rotates at its minimum rotating speed and the on-off

cycles start.

It is possible to summarize the above considerations by observing that MCHPs and

PT
IDHPs are able to work with a larger COP with respect to an ON-OFF mono-compressor

HP because with MCHP and IDHP it becomes possible to delay the starting of the on-off

RI
compressor cycles and in this way to reduce the energy losses at partial loads. More the

SC
secondary bivalent temperature is larger with respect to the bivalent temperature, more

the COP at partial loads can be improved by adopting a MCHP or an IDHP with respect

U
to a mono-compressor ON-OFF HP.
AN
FIGURE 5 HERE

The values assumed by Tbiv and Tbiv,2 depends not only by the heat pump but also by the
M

thermal loads of the building; for this reason, in order to highlight the role played by the
D

thermal loads of the building for a fixed size of heat pump, the selected heat pumps have

been coupled to 11 different buildings located in Bologna, each one characterized by a

TE

different thermal load. In this way, for each case different bivalent and secondary bivalent
EP

temperature are obtained. BES is used to univocally identify the building thermal needs

(see Section 2.1). All the BES have been obtained by considering a Heating Limit
C

External Temperature (HLET) equal to 16°C. As evidenced in Table 2 the bivalent

AC

temperature ranges between about -5.00°C (# 1) and about 6.5°C (#11); this means that

the selected heat pumps are able to cover completely the thermal needs of building #1 in

Bologna where the design outside temperature (Tdes) is equal to -5°C; on the contrary, the

other buildings (#2 - #11) are characterized by thermal loads that the selected heat pumps

are able to cover only partially. In these cases, an electrical heater is considered as back-
 ACCEPTED MANUSCRIPT

up system. As described before, each simulation has been made by considering the

selected heat pump coupled to a radiant floor heating system working with hot water

having a fixed temperature equal to 35°C (Tw), value independent from the outside

temperature. In Table 2, the values of the secondary bivalent temperature are also

PT
summarized for the 11 building considered.

TABLE 2 HERE

RI
By analyzing different combinations heat pump-building by means of the model

SC
described in Section 2, it is possible to analyze the influence of the bivalent and

secondary bivalent temperature (in the case of MCHP and IDHP) on the seasonal energy

U
performance of the system (SCOPon , SCOPnet), as well as the energy delivered by the
AN
heat pumps during the season and the energy delivered by the electric back-up system

when the heat pump thermal capacity is lower than the thermal load needed by the
M

building. As an example, in the next Figures the results obtained for the building #7 are
D

reported. The typical bin distribution of Bologna is shown in Fig. 1 and it is also reported

in Fig. 6a, 6b and 6c together with the BES of building #7 (red line) and the heating
TE

capacity of the selected ON-OFF HP (Fig. 6a), MCHP (Fig. 6b) and IDHP (Fig. 6c)
EP

respectively.

By comparing Fig.6a, 6b and 6c it is possible to observe that these three systems are
C

characterized by the same bivalent temperature (Tbiv≈2.5°C); this is a consequence of the

AC

similar heating capacity at full load demonstrated by the three selected heat pumps (see

Fig. 3a). Fig. 6b and Fig. 6c show also the different modulation of the delivered heating

power of the MCHP and IDHP respectively: more in detail, the MCHP can modulate its

heating capacity avoiding the on-off cycles when the outside temperature is in between

Tbiv (2.78°C) and Tbiv,2 (7.73°C). Fig. 6c shows that the selected IDHP is able to supply
 ACCEPTED MANUSCRIPT

the exact building thermal load when the outside temperature ranges between Tbiv

(2.34°C) and Tbiv,2 (11.24°C).

FIGURE 6 HERE

In Fig. 7a, 7b and 7c a comparison between the energy delivered by the three selected

PT
heat pumps (QHP) and the building thermal energy (case #7) (Qb, Eq.(2)) is shown. In the

same Figure, the electric energy consumptions of the heat pumps (EHP,us) and of the

RI
back-up system (EBU) are reported. The trend of the thermal energy required by the

SC
building #7 as a function of the outdoor temperature follows the trend of the bin

distribution of Bologna with a maximum in correspondence of an outdoor temperature

U
equal to 5°C. The heat delivered by the heat pump follows a similar trend; the energy
AN
delivered by the heat pump matches perfectly the curve of the building energy demand

for outdoor temperature values larger than the bivalent temperature. When Text<Tbiv the
M

energy delivered by the heat pump becomes lower than the building energy demand. In
D

Table 3 the value of seasonal building energy demand (Qb) is quoted and compared by

the seasonal value of the thermal energy delivered by the heat pumps (QHP); the
TE

difference between these two values highlights the downsizing of the heat pump toward
EP

the building thermal request. Since these three systems are characterized by a similar

bivalent temperature, this difference is similar for the three cases; in Table 3 this
C

difference is indicated as EBU because this is the energy delivered by the back-up system.
AC

FIGURE 7 HERE

Fig. 7a, 7b and 7c report also the electric energy used by the heat pump and by the back-

up system (i.e. electric heaters). It is noticeable that the electric back-up system is

activated below the bivalent temperature and it works in parallel with the heat pump. The

energy delivered by the back-up system ranges between 5.6% and 7% of the seasonal
 ACCEPTED MANUSCRIPT

building thermal request and it becomes maximum (7%) for MCHP which is

characterized by the higher bivalent temperature with respect to the selected ON-OFF

HP and IDHP.

In Table 3 the seasonal electric energy consumptions of the three selected heat pumps

PT
(EHP,us), calculated by applying the model described in this paper, are reported. It is

evident that the selected IDHP is characterized by the minimum electric energy

RI
consumption with respect to ON-OFF HP and MCHP. The best performance of IDHP is

SC
also highlighted by the obtained values of SCOPon and of SCOPnet; the IDHP is

characterized by a value of SCOPnet which is greater than 9% with respect to the value

U
obtained adopting the ON-OFF HP and 8.5% with respect to the MCHP. In terms of
AN
SCOPon this difference becomes equal to 7% with respect to the ON-OFF HP and 10%

with respect to MCHP.

M

TABLE 3 HERE
D

The evaluation of the seasonal performance of the 11 systems considered in this work

enables to analyze the influence of a different sizing of the heat pump on the SCOP of
TE

the system; in fact, by considering all the 11 buildings, for each heat pump it is possible
EP

to assess the seasonal performances over a large range of values of the bivalent

temperature, from -4.64°C to 6.31°C. In Fig. 8 the trend of SCOPnet as a function of the
C

bivalent temperature is represented; the values of SCOPnet takes into account the behavior
AC

of the heat pumps at partial loads during the whole season.

FIGURE 8 HERE

Looking at Fig. 8, it is evident that the influence of Tbiv on the seasonal performance of

these heat pumps is very different. Both the MCHP and the ON-OFF HP present

increasing SCOPnet when the bivalent temperature grows; however, it is evident that the
 ACCEPTED MANUSCRIPT

SCOPnet of an ON-OFF HP is more sensible to the choice of a different Tbiv with respect

to a MCHP. On the contrary, the IDHP is characterized by a value of SCOPnet more stable

when Tbiv is changed; however, the data reported in Fig. 8 evidence that SCOPnet for the

IDHP decreases if the bivalent temperature increases, which is an opposite trend with

PT
respect to ON-OFF HP and MCHP. In order to explain these trends it is important to keep

in mind that both MCHP and IDHP are less influenced by the increase of the bivalent

RI
temperature for their improved capacity, if compared with a ON-OFF HP, to adapt the

SC
delivered energy to the building request by delaying the use of on-off cycles, which start

only when the outdoor temperature exceeds Tbiv,2. Moreover, the MCHP works with the

U
best efficiency when only one compressor is active, precisely when the thermal load is
AN
equal to one compressor heating capacity. For the IDHP it is important to highlight that

Tbiv,2 is generally higher than Tbiv,2 of the MCHP; in this way the decrease of the SCOPnet
M

due to the increase of the on-off cycles becomes less important at low values of the
D

outdoor temperature.

In Fig. 9 the trend of SCOPon with respect to the bivalent temperature is represented.
TE

FIGURE 9 HERE
EP

When the bivalent temperature is increased, MCHP and IDHP have the same decreasing

trend. On the contrary, the mono-compressor ON-OFF HP presents initially an increasing

C

SCOPon when Tbiv increases up to 0°C; after that, the value of SCOPon decreases when Tbiv
AC

is increased in agreement with the other heat pump models. The decreasing trend of

SCOPon when the bivalent temperature is increased is linked to the use of the electric

back-up: in fact, higher bivalent temperature means higher energy consumptions of the

back-up system, which works with lower COP values (theoretically COP=1 for an

electric heater). In the case of a mono-compressor ON-OFF HP with a low bivalent

 ACCEPTED MANUSCRIPT

temperature the total energy delivered by the electric back-up becomes negligible and the

increase of SCOPnet when the bivalent temperature increases is more significant than the

utilization of the back-up system and this explain the maximum of the SCOPon trend. This

fact means that, adopting a mono-compressor ON-OFF HP, there is a specific value of the

PT
bivalent temperature (larger than the design temperature) which maximizes the value of

SCOPon. On the contrary, when a IDHP is selected, the maximum value of SCOPon is

RI
obtained by adopting a bivalent temperature equal to the design temperature (Tbiv=Tdes).

SC
The MCHP has an optimal bivalent temperature which is in between the design

temperature and the optimal bivalent temperature for an ON-OFF HP. These results

U
highlight that the optimal sizing of a heat pump system is strongly influenced by the
AN
modulation characteristics of the adopted heat pump.
M

CONCLUSIONS
D

In this paper a mathematical model for the evaluation of the seasonal performances of

air-to-water heat pumps coupled to a radiant floor heating system is presented. The
TE

developed model uses the bin-method and the building energy signature in order to
EP

calculate the thermal energy demand during the heating season. Different heat pump

models are considered: mono-compressor heat pumps (ON-OFF HP), multi-compressor

C

heat pumps (MCHP) and variable-speed mono-compressor heat pumps (IDHP), each
AC

one characterized by a specific capacity of modulation of the delivered thermal power.

The mathematical model calculates heat pump performances (i.e. heating capacity and

COP) for each partial load working condition and then evaluates the seasonal efficiency

SCOPnet and SCOPon of the system. The proposed procedure takes into account energy

losses due to the starting of the on-off cycles. The model has been used in order to
 ACCEPTED MANUSCRIPT

evaluate the influence of the bivalent temperature and of the secondary bivalent

temperature (in the case of MCHP and IDHP) on the seasonal performances of the

system. With this aim, the thermal performances achievable by adopting specific ON-

OFF HP, MCHP and IDHP for the production of hot water with a temperature of 35°C,

PT
coupled with 11 different buildings, have been simulated. The numerical results confirm

that the modulation characteristics of the different heat pumps strongly affect the

RI
optimal sizing of the system. With a fixed building thermal load, mono-compressor ON-

SC
OFF HPs present an optimal bivalent temperature higher than multi-compressor and

inverter-driven models. Thus, down-sizing a mono-compressor heat pump and over-

U
sizing multi-compressor and inverter-driven models lead to an optimization of the
AN
system efficiency.
M

AKNOWLEDGEMENTS
D

The authors acknowledge the support provided by Galletti Group Spa (Italy) which

makes available the technical data sheets of their air-to-water heat pumps.
TE
EP

NOMENCLATURE

a, b, c Coefficients of the polynomial regression in Eqs. (3,4,5) (-)

C
AC

Cc Degradation coefficient of the heat pump (-)

COPDC Coefficient of performance at full load (-)

COPPL Coefficient of performance at partial load (-)

CR Capacity ratio (-)

 ACCEPTED MANUSCRIPT

EBU Energy delivered by the electric back-up system (kWh)

EHP,us Electric energy used by the heat pump (kWh)

fCOP Coefficient of performance correction factor for on-off cycles (-)

PT
HLET Heating limit external temperature (K)

RI
Pb Heating power required by the building (kW)

SC
Pdes Design load of the building (kW)

PHP Thermal power delivered by the heat pump (kW)

U
AN
Qb Building energy demand (kWh)
M

QHP Energy delivered by the heat pump (kWh)

SCOPon Seasonal coefficient of performance of the system (heat pump and back-up)
D

defined by Eq. (16) (-)

TE

SCOPnet Seasonal coefficient of performance of the electric heat pump defined by Eq.
EP

(16) (-)
C

tbin Duration of the i-th bin (h)

AC

tbin,n/N Period of time with n compressors switched on over N total compressors in a

generic bin (h)

Tbiv Bivalent temperature (K)

Tbiv,2 Secondary bivalent temperature (K)

 ACCEPTED MANUSCRIPT

Tdes Heating design temperature (K)

Text Outdoor air temperature (K)

TOL Heat pump Temperature Operative Limit (K)

PT
Tw Hot water temperature produced by the heat pump (K)

RI
Greek symbols

SC
Ф Inverter working frequency (Hz)

Фeff Actual value of the inverter frequency (Hz)

U
AN
REFERENCES

[1] European Parliament, Directive 2009/28/EU: on the promotion of the use of energy
M

from renewable sources (RES Directive), 23 April (2009).

D

[2] D. Cortés-Borda, A. Ruiz-Hernández, G.Guillén-Gosálbez, M. Llop, R. Guimerà, M.

TE

Sales-Pardo, Identifying strategies for mitigating the global warming impact of the

EU-25 economy using a multi-objective input-output approach, Energy Policy 77,

EP

(2015) 21-30.

[3] European Parliament, Directive 2010/31/EU: on the energy performance of

C

buildings. 19 May (2010).

AC

[4] E. Bettanini, A. Gastaldello, L. Schibuola, Simplified models to simulate part load

performances of air conditioning equipments, in: Proceedings of IBPSA 8th Int.

Conf. Building Simulation ’03, Eindhoven (2003) 107-114.

[5] H.I. Henderson, D. Parker, Y.J. Huang, Improving DOE-2’s RESYS routine: user

defined functions to provide more accurate part load energy use and humidity
 ACCEPTED MANUSCRIPT

predictions, in: Proc. of ACEEE Summer Study on Energy Efficiency in Buildings,

August 15-20, Pacific Grove, CA (2000), 112-124.

[6] E. Kinab, D. Marchio, P. Rivière, A. Zoughaib, Reversible heat pump model for

seasonal performance optimization, Energy and Buildings 42, (2010) 2269-2280.

PT
[7] H. Park, J. S. Lee, W. Kim, Y. Kim, The cooling seasonal performance factor of a

hybrid ground-source heat pump with parallel and serial configurations, Applied

RI
Energy 102, (2013) 877-884.

SC
[8] K. Klein, K. Huchtemann, D. Muller, Numerical study on hybrid heat pump systems

in existing buildings, Energy and Buildings 69, (2014) 193-201.

U
[9] C. Naldi, G. L. Morini, E. Zanchini, A method for the choice of the optimal balance-
AN
point temperature of air-to-water heat pumps for heating, Sustainable Cities and

Society 12, (2014) 85-91.

M

[10] P. Vocale, G. L. Morini, M. Spiga, Influence of outdoor air conditions on the air
D

source heat pumps performance, Energy Procedia 45, (2014) 653-662.

[11] K. Papakostas, A. Bentoulis, V. Bakas, N. Kyriakis, Estimation of ambient

TE

temperature bin data from monthly average temperatures and solar clearness index.
EP

Validation of the methodology in two Greek cities, Renewable Energy 32, (2007)

991-1005.
C

[12] P.W. Francisco, B. Davis, D. Baylon, L. Palmiter, Heat pump system performance
AC

in northern climates, ASHRAE Transactions 110, (2004) 442-451.

[13] I. Sarbu, D. Dan, C. Sebarchievici, Performances of heat pump systems as users of

renewable energy for building heating/cooling, Wseas Transactions on Heat and

Mass Transfer, (2014) 51-62.

 ACCEPTED MANUSCRIPT

[14] EN 14825:2013, Air conditioners, liquid chilling packages and heat pumps, with

electrically driven compressors, for space heating and cooling – Testing and rating

at part load conditions and calculation of seasonal performance, (2013).

[15] UNI/TS 11300-4, Energy performance of buildings – Part 4: Renewable energy

PT
and other generation systems for space heating and domestic hot water production,

(2012).

RI
[16] prUNI 10349-1, Heating and cooling of buildings – Climatic data, (2015).

SC
[17] UNI EN 12831, Heating systems in buildings – Method for calculation of the

design heat load, (2006).

U
[18] M. F. Fels , PRISM: An introduction, Energy and Buildings 9, (1986) 5-18.
AN
[19] EN 15603:2008, Energy performance of buildings – Overall energy use and

definition of energy ratings, (2008).

M

[20] F. Madonna, F. Bazzocchi, Annual performances of reversible air-to-water heat

D

pumps in small residential buildings, Energy and Buildings 65, (2013) 299-309.
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

Table captions

PT
Table 1 - Heat pumps rated performances (7°C dry bulb, 6°C wet bulb for outdoor air

RI
temperature; 40-45°C inlet/outlet water temperature) and technical

SC
characteristics.

U
Table 2 – Heating design loads of the reference buildings and their relative bivalent
AN
temperatures as a function of the selected heat pump.
M

Table 3 - Main outputs obtained for the case #7.

D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
Table 1

RI
Heat pump typology ON-OFF HP MCHP IDHP
Heating Capacity (kW) 35.6 34.2 34.9

SC
COP 3.42 3.17 3.06
Total power input 10.4 10.8 11.4
(kW)
Frequency range (Hz) 50 50 30-120

U
TOL (°C) -10 -10 -18
Compressors number 1 2 1
AN
Compressor model Maneurop SANYO Siam
SH-140-4 1: CSBN453H8H ANB52FTTMT
2: CSBN453H8H
M

Refrigerant fluid R410A R410A R410A

Refrigerant circuits 1 1 1
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
Table 2

RI
Case 1 2 3 4 5 6 7 8 9 10 11
Design load 28 31 34 37 40 45 50 55 60 67 75

SC
(kW)
ON-OFF Tbiv -4.5 -3.1 -2.0 -1.0 -0.1 1.2 2.3 3.23 4.0 5.0 5.9
HP
MCHP Tbiv -4.0 -2.6 -1.5 -0.5 0.5 1.7 2.8 3.6 4.5 5.4 6.3

U
Tbiv,2 3.1 4.0 4.8 5.5 6.1 7.0 7.7 8.3 8,9 9,5 10,1
IDHP Tbiv -4.6 -3.3 -2.1 -1.0 -0.1 1.2 2.3 3.3 4.1 5.1 6.0
AN
Tbiv,2 8.2 8.8 9.4 9.8 10.2 10.8 11.2 11.6 12.0 12.3 12.7
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
Table 3

RI
Heat pump SCOPnet SCOPon Tbiv Tbiv,2 Qb QHP EHP,us EBU
(°C) (°C) (kWh) (kWh) (kWh) (kWh)

SC
ON-OFF HP 3.80 3.29 2.30 / 92509.3 87336.5 22971.8 5172.8
MCHP 3.82 3.21 2.78 7.73 92509.3 86209.4 22542.8 6299.9
IDHP 4.15 3.53 2.34 11.24 92509.3 87353.0 21033.0 5156.3

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

Figure captions

Figure 1 - Bin distribution for the heating season (Oct. 15th - Apr. 15th) in Bologna

PT
(Italy).

Figure 2 – Typical trend of the Building Energy Signature (BES) and characteristic

RI
curve of a mono-compressor ON-OFF air source heat pump (a) and of a

SC
multi-compressor/inverter-driven air source heat pump (b).

Figure 3 – Heating power capacity (a) and total power inputs (b) of the selected heat

U
pumps as a function of the outdoor temperature (Tw=35°C).
AN
Figure 4 – COP of the selected MCHP at full load (two compressors activated: 2/2) and

with one compressor switched on (1/2) for a fixed value of the hot water
M

temperature (Tw=35°C).
D

Figure 5 – COP of the selected IDHP for different values of the outdoor temperature

and for a fixed value of the hot water temperature (Tw=35°C).

TE

Figure 6 – Superposition of bin trend, building energy signature (case #7) and the heat
EP

pump thermal power: ON OFF HP (a); MCHP (b); IDHP (c).

Figure 7 – Building heating energy (Qb) for the case #7, the thermal energy delivered by
C

the heat pump (QHP), the electric energy used by the heat pump (EHP,us) and
AC

the electric energy used by the back-up system (EBU) as a function of the

outdoor temperature: ON-OFF HP (a); MCHP (b); IDHP (c).

Figure 8 – SCOPnet as a function of the bivalent temperature for the selected heat

pumps.

Figure 9 – SCOPon as a function of the bivalent temperature for the selected heat pumps.
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
Figure 1

RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

Figure 2

PT
RI
U SC
AN
(a)
M
D
TE
EP
C
AC

(b)
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

Figure 3

PT
RI
U SC
AN
M

(a)
D
TE
EP
C
AC

(b)
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
RI
Figure 4

U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
RI
Figure 5

U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

Figure 6

PT
RI
SC
(a)

U
AN
M
D
TE

(b)
EP
C
AC

(c)
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

Figure 7

PT
RI
SC
(a)

U
AN
M
D
TE

(b)
EP
C
AC

(c)
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
RI
Figure 8

U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Matteo Dongellini, Claudia Naldi and Gian Luca Morini

PT
RI
Figure 9

U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

• A model for the prediction of seasonal performance of HPs has been developed.
• The model considers mono-compressor, multi-compressor and inverter-driven HPs.
• The procedure takes into account HPs performances at partial load.
• Optimization of heat pump sizing depending on its control system.

PT
RI
U SC
AN
M
D
TE
C EP
AC