HYPERGRYD. This p ojec has ecei ed unding om he Eu opean Union’s Ho izon 2020 esea ch and
inno a ion p og amme unde g an ag eemen No 101036656
WP5 – TRL5 demons a ion in li ing labs and
i ual labs in LEC
Task 5.3 Demons a ion and alida ion
ac ions o RES-based Enabling echnologies
D5.8 Ou comes o HYPERGRYD
demons a ion, lessons lea n and
guidelines o eplica ion
Re . A es(2025)2636601 - 02/04/2025
DX.X Xxxxxxxx 2
DISCLAIMER
The opinion s a ed in his epo e lec s he opinion o he au ho s and no he opinion o he
Eu opean Commission.
All in ellec ual p ope y igh s a e owned by HYPERGRYD conso ium membe s and a e p o ec ed by
he applicable laws. Rep oduc ion is no au ho ised wi hou p io w i en ag eemen .
The comme cial use o any in o ma ion con ained in his documen may equi e a license om he
owne o ha in o ma ion.
ACKNOWLEDGEMENT
This p ojec has ecei ed unding om he Eu opean Union’s Ho izon 2020 esea ch and inno a ion
p og amme unde g an ag eemen Nº 101036656.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
3
P ojec
P ojec Ac onym
HYPERGRYD
P ojec Ti le
Hyb id coupled ne wo ks o he mal-elec ic in eg a ed Sma Ene gy Dis ic s
G an Ag eemen
numbe
101036656
Call iden i ie
H2020-LC-GD-2020
Topic iden i ie
LC-GD-2-1-2020
Inno a i e land-based and o sho e enewable ene gy echnologies and hei
in eg a ion in o he ene gy sys em
Funding Scheme
Resea ch and Inno a ion Ac ion
P ojec du a ion
42 mon hs (F om 1 Oc obe 2021)
Coo dina o
ARCbcn
Websi e
h p://hype g yd.eu
Deli e able
Deli e able No.
5.8
Deli e able i le
Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
Desc ip ion
The p esen epo is in ended o p esen main ou comes and lessons lea n o he
demons a ion o Hype g yd solu ions implemen ed in ou Li e-in-Labs: SONNE,
ENVI, KEZO and EURAC. Ou come o he demons a ion will be a se o lessons lea n
o be used as guidelines o he ins alla ion and ope a ion o RES echnologies in
di e en DHC and Sma Hyb id G ids.
WP No.
WP5
Rela ed ask
T5.3 Demons a ion and alida ion ac ions o RES-based Enabling echnologies
T5.4 Model demons a ion and alida ion o he mul i-ca ie ene gy dynamic model
a ENVI LIL
T5.5 Model demons a ion and alida ion o he HYPERGRYD ICT se ices a SONNE
LIL
Lead Bene icia y
17, IMP PAN - Ins y u Maszyn P zeplywowych im Robe a Szewalskiego Polskiej
Akademii Nauk
Au ho (s)
Michał Gliński (IMP-PAN), Paweł Zawadzki (IMP-PAN), We onika Radziszewka (IMP-
PAN), Jö g Ve s ae e (IMP-PAN), Mau o Pipiciello (EURAC), Foulad a Mohammad
Hossein (EURAC), Manuela Binde and Ma in B uckne (SONNE)
Con ibu o (s)
Sebas ian Bykuć (IMP-PAN), Da id Ve ez (ARC)
Type
R
Dissemina ion
PU Public
Language
English – GB
Due
31/03/2025
Submission da e
31/03/2025
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
4
Ve sion
Da e
Au ho s
Desc ip ion
V.0.1
7/03/2025
Michał Gliński (IMP PAN)
Templa e o deli e able documen s and
ask planning
V.0.2
20/03/2025
Manuela Binde and Ma in
B uckne (SONNE)
Sec ion ega ding SONNE
V0.3
25/03/2025
Michał Gliński (IMP PAN)
Sec ion ega ding KEZO
V.1.0
28/03/2025
Michał Gliński (IMP-PAN), W.
Radziszewska (IMP-PAN), J.
Ve s ae e (IMP-PAN)
Ve sion o in e nal e iew (missing inpu
om EURAC)
V.1.1
31/03/2025
Mau o Pipiciello (EURAC)
Sec ion ega ding EURAC
V.2.0
31/03/2025
Michał Gliński (IMP-PAN),
Final deli e able
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
5
Table o Con en s
1 Execu i e Summa y ................................................................................................... 9
2 In oduc ion ............................................................................................................ 10
2.1 Scope ........................................................................................................................... 10
2.2 Audience ...................................................................................................................... 10
2.3 Abb e ia ions .............................................................................................................. 10
2.4 Con ibu ions o pa ne s ............................................................................................ 12
2.5 S uc u e ...................................................................................................................... 12
3 Ou comes and lessons lea n ................................................................................... 13
3.1 SONNE ......................................................................................................................... 13
3.2 KEZO ............................................................................................................................ 19
3.3 EURAC .......................................................................................................................... 26
3.3.1 Desc ip ion o he LiL and objec i es o demons a ion .................................................... 26
3.3.2 Subs a ion desc ip ion and con ols .................................................................................. 27
3.3.3 Assessmen o he con ol pe o mance ........................................................................... 29
3.3.4 Lessons lea ned .................................................................................................................. 35
3.4 ENVI ............................................................................................................................. 36
4 Guidelines o eplica ion ........................................................................................ 40
4.1 Modula HP wi h PCM s o age (ins alled in KEZO LiL) ................................................ 40
4.1.1 Pu pose 40
4.1.2 Obse ed po en ial echnical ba ie s o eplica ion ....................................................... 41
4.1.3 Replica ion equi emen s and ecommenda ions ............................................................. 43
4.2 So p ion S o age (ins alled in KEZO LiL) ...................................................................... 44
4.2.1 Pu pose .............................................................................................................................. 44
4.2.2 Obse ed po en ial echnical ba ie s o eplica ion ....................................................... 45
4.2.3 Replica ion equi emen s and ecommenda ions ............................................................. 46
5 Conclusions ............................................................................................................. 47
6 Re e ences .............................................................................................................. 52
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
6
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
7
Lis o Figu es
Figu e 1 Li e-in lab Sonnenpla z (Aus ia) ........................................................................................... 13
Figu e 2 Key pe o mance indica o s, Sonnenpla z LEM s udy simula ion scena ios and objec i es, GSY
simula ion s udy o HYPERGRYD 2023-2024 ...................................................................................... 14
Figu e 3 P oposed open-sou ce so wa e solu ion o cen alized demand-side managemen ........ 15
Figu e 4 KEZO Reseach Cen e, iew on L5 building. ........................................................................... 19
Figu e 5 Simpli y scheme o H&C sys ems in KEZO LiL ........................................................................ 20
Figu e 6 The modula hea pump connec ed o KEZO sys em. ........................................................... 21
Figu e 7 So p ion s o age connec ed wi h Mayekawa CO2 Hea Pump in KEZO Resea ch Cen e .... 22
Figu e 8 Two-day ahead p edic ion o he adia o s and an coils hea ing powe in building L5 using
GRU app oach ...................................................................................................................................... 23
Figu e 9 Tempo al a ia ion o COP hea ing: measu emen and NN .................................................. 24
Figu e 10 Reg ession cu e o he COP hea ing p edic ion using NN ................................................. 24
Figu e 11 Model-based con ol o he adso p ion s o age using he KTH edge panel ....................... 25
Figu e 1. Ene gy Exchange Lab o Eu ac Resea ch schema ic ............................................................. 26
Figu e 2. O e iew o he subs a ion's ope a ion wi hin he con ex o 4 h and 5 h GDHN. ............... 27
Figu e 3. O e iew o he ou di e en con igu a ion schemes conside ed in his s udy. The DH
ne wo k is on he le and he building on he igh side o each scheme .......................................... 28
Figu e 4. Schema ic o MPC amewo k .............................................................................................. 28
Figu e 5. Resul o compa ison he pe o mance o MPC s. RBC wi h TOU elec ici y p ice ( es 1) 31
Figu e 6. Dis ibu ion o HP ac i a ion du a ions du ing cheap and expensi e pe iods unde RBC and
MPC wi h TOU elec ici y p icing. ........................................................................................................ 32
Figu e 7. Compa ison o ene gy s o ed pe cha ging cycle and numbe o cha ging unde RBC (blue)
and MPC ( ed) wi h TOU elec ici y p icing ......................................................................................... 32
Figu e 8. Resul o compa ison he pe o mance o MPC s. RBC wi h ealis ic elec ici y p ice ( es 2)
.............................................................................................................................................................. 33
Figu e 9. Dis ibu ion o HP ac i a ion du a ions du ing cheap and expensi e pe iods unde RBC and
MPC wi h ealis ic elec ici y p icing.................................................................................................... 34
Figu e 10. Compa ison o ene gy s o ed pe cha ging cycle and numbe o cha ging unde RBC (blue)
and MPC ( ed) wi h ealis ic elec ici y p icing ................................................................................... 34
Figu e 12 Li e-in lab En i onmen Pa k (Tu in-I aly) ........................................................................... 36
Figu e 13 Main buildings wi hin En i onmen Pa k ............................................................................. 37
Figu e 14 Model o he medium and low ol age elec ic dis ibu ion sys em o ENVI in ENCO. Red
poin s ep esen nodes wi h elec ic demands ................................................................................... 38
Figu e 15 Mod Model o he he mal dis ibu ion ne wo k o En i Pa k in SAIn . Nodes desc ibe he
supply o demand maximum hyd aulic p essu e ................................................................................ 38
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
8
Lis o Tables
Table 1 Technology speci ic KPIs panel ................................................................................................ 22
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
9
1 Execu i e Summa y
The goal o HYPERGRYD p ojec is he de elopmen o a se o eplicable and scalable cos e ec i e
echnical solu ions o allow he in eg a ion o Renewable Ene gy Sou ces (RES) wi h di e en
dispa chabili y and in insic a iabili y inside The mal G ids as well as hei link wi h he Elec ical
G ids, including he de elopmen o inno a i e key componen s, in pa allel wi h inno a i e and
in eg a ed ICT se ices o med by a scalable sui e o ools o he p ope handling o he inc eased
complexi y o he sys ems om building o Local Ene gy Communi y (LEC) le els and beyond, and
accele a e he sus ainable ans o ma ion, planning and mode niza ion o Dis ic Hea ing and
Cooling (DHC) owa ds 4 h and 5 h gene a ion.
HYPERGRYD also aims a de eloping eal ime managemen o bo h elec ical and he mal ene gy
lows in he coupled ene gy ne wo k complex, including he syne gies be ween hem. The e o e,
HYPERGRYD aims a h ee o e -a ching Gene al Objec i es:
• To p o e Sma Ene gy Ne wo ks as he u u e o E icien Ene gy Managemen in DHC in
syne gy wi h he Elec ical G ids in LEC/Sma Ci ies o he u u e;
• To de ine he oadmap o design and plan u u e DHC as well as he mode niza ion o he
exis ing ones in di e en clima es and RES pene a ion le els owa d 4 h-5 h gene a ion,
• To demons a e HYPERGRYD RES-based Enabling Technologies, Sma Ene gy G id Solu ions
empowe ed by new ICT ools and se ices as he key o his e olu ion.
Du ing he p ojec , he HYPERGRYD’s solu ions will be implemen ed ac oss ou Li e-In-Labs cases in
h ee ep esen a i e clima es, wi h special conside a ion o hei cos e ec i eness and po en ial
eplicabili y o inally achie e hese h ee main objec i es.
The pu pose o his deli e able in sho is: i) o p esen s main ou comes o a demons a ion o
HYPERGRYD solu ions o s akeholde s, ii) se o lessons lea n , iii) guidelines o eplica ion o h ee
enabling echnologies.
Ta ge audience o his epo is he wide g oup o s akeholde s om dis ic hea ing sec o s and H&C
indus y.
On behal o Au ho s
Michał Gliński, IMP PAN
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
16
lows in o de o op imize o ini ially plan hem, ocusing p ima ily on physical simula ions.
As a esul , he so wa e calcula es:
• he e u n empe a u e om gene a ion asse s;
• he supply empe a u e ha eaches he use s;
• he he mal losses ac oss he pipes;
• he p essu e and empe a u e d ops o each pipe (bo h on he supply and e u n side);
• he p essu e a he nodes (supply and e u n side);
• he low a es in he pipes and ne wo k-side hea exchange s;
• he ou le empe a u e a he nodes
on he ne wo k side. The low a e and empe a u e p o iles on he seconda y side o he
hea exchange s connec ed o he nodes a e no aken in o accoun .
GET p o ides he op imiza ion o exis ing o planned DHC ne wo ks as a se ice based on he «
Exe goeconomic op imiza ion ool o 4 h and 5 h gene a ion o DHC » o op imize he ope a ion o he
DHC g id om an exe goeconomic poin o iew owa ds he 4 h gene a ion o DHC, conside ing
in es men s, uel and ope a ion cos s, and he p ice index, as well as lexible hea and elec ici y a i s
and cos s o he educ ion o GHG emissions. Po en ial use cases o his se ice can be:
• The simula ion and calcula ion o he in eg a ion o decen alized p oduc ion sys ems in
o de o in eg a e RES and P2H, which is pa icula ly in e es ing in he case o a iable
a i s o p oduce s. Fo example, low a i s a special ime slo s can be used op imally
o eeding hea om a hea pump in o he DHC sys em. Va iable a i s o hea
consume s a e no common ye , bu his could be an in e es ing aspec o he u u e.
• The in es iga ion o po en ial educ ions o hea losses and pumping cos s.
• An exe gy-based analysis o educe he exe gy le el, he empe a u e le els, and he
hea losses.
• The compa ison o di e en scena ios, e.g. he cu en si ua ion and a scena io
conside ing a g id ex ension o he in eg a ion o new consume s o o decen alized
eed-in sys ems.
• The op imiza ion o he DHC g id based on a p essu e d op in o de o pinpoin
bo lenecks o loca e sui able posi ions o boos e pumps.
• The calcula ion o he e ec s o a g id ex ension o he whole DHC g id.
Some esul s om use case 1, mainly hose ela ed o he in eg a ion o lexible hea ing
sys ems in o he elec ical g id and dynamic p icing e ec s on lexible asse s, we e used as
inpu .
Mo e in o ma ion abou he ools can be ound in D3.2, D3.4, D3.6 and D3.8 o his p ojec .
The HYPERGRYD P ojec success ully demons a ed he po en ial o hese ad anced ICT ools o
op imize and in eg a e enewable ene gy sou ces wi hin he mal and elec ical g ids. The alida ion
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
17
o hese ools ac oss mul iple s akeholde s—including p i a e households, municipali ies, DHC
ope a o s, and elec ic g id ope a o s— p o ided aluable insigh s in o bo h echnical and non-
echnical aspec s o he implemen a ion o he ools and se ices p esen ed. The key indings o he
alida ion p ocess can be ound in de ail in D5.6 o his p ojec and can be summa ised as ollows:
• P i a e Households/Consume s/P osume s: Thei p ima y in e es lies in being eliably
supplied wi h hea , ho wa e , and elec ici y a easonable p ices a he han engaging wi h
echnical de ails. Al hough he esiden s in G oßschönau demons a e openness o
con ibu ing o enewable ene gy ini ia i es, bols e ed by a long-s anding cul u e o
sus ainabili y, ba ie s such as adminis a i e complexi y and high ini ial in es men s hinde
ull pa icipa ion in ad anced ene gy managemen sys ems. G an ing hi d-pa y access o
hea pumps, PV sys ems, o ba e y in e e s o coo dina ed ene gy managemen aced
signi ican skep icism om local s akeholde s. Conce ns include po en ial ene gy
una ailabili y and unexpec ed cos s due o echnical e o s. Many esiden s p e e e aining
ull con ol, hough olun a y ecommenda ions ins ead o di ec emo e con ol could be a
mo e accep able al e na i e. T us and anspa ency a e c ucial, as willingness o g an
access depends on clea bene i s. While skep icism emains, da a sha ing is al eady
p ac iced in exis ing local and egional ene gy communi ies, whe e pa icipan s bene i om
lowe ene gy cos s, educed g id ees, and ma ke independence. These insigh s highligh he
impo ance o clea communica ion and use -cen ic implemen a ion o ene gy managemen
solu ions.
Also, on he subjec o lexible a i s o elec ici y and hea , local s akeholde s emained
cau ious, while mos in e na ional pa icipan s a he wo kshop ound lexible a i s
ele an . Many local consume s may s uggle o adap o hou ly p ice changes, p e e ing
simple day/nigh a i s. Howe e , some p osume s ha e al eady begun adjus ing hei
beha io , using PV elec ici y o EV cha ging o appliances when a ailable. Success ul
adop ion o dynamic hea ing models would equi e g adual lea ning and a ge ed educa ion
e o s o inc ease accep ance.
Thus, simpli ied implemen a ion, inancial incen i es, and anspa en communica ion o
bene i s a e p e equisi es o engagemen . These insigh s emphasize he impo ance o
aligning echnical ad ancemen s wi h use needs and p e e ences o success ul in eg a ion
in o local ene gy sys ems.
• Municipali ies: Municipali ies such as G oßschönau ecognize he bene i s o op imizing
elec ici y and hea supply, in eg a ing RES, and ad ancing clima e goals. Howe e , inancial
cons ain s pose signi ican challenges. In es men s in in as uc u e mus balance economic
e u ns wi h public p io i ies. Municipali ies equi e obus inancial models and echnical
suppo o scale enewable ene gy ini ia i es while add essing compe ing local needs. Clea
policy amewo ks and unding mechanisms ailo ed o municipal budge s could enable
g ea e adop ion o inno a ions. Rega ding DHC, municipali ies a ely ac as planne s o
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
18
ope a o s o DHC sys ems di ec ly in Aus ia. I is mo e common o a coope a i e o be
es ablished o hese pu poses.
• DHC Ope a o s: DHC ope a o s ecognize he alue o inno a i e ools o op imizing exis ing
and planned ne wo ks o enhancing hem, whe eby hey p e e using a se ice a he han
using a ool and ely on pe sonal con ac wi h he se ice-p o ide . Key p io i ies include he
educ ion o empe a u e losses and he op imiza ion o summe ope a ions, bo h o which
con ibu e o g ea e ope a ional e iciency and lowe ene gy cos s. Howe e , b oade
adop ion o he ools equi es add essing echnical and da a- ela ed challenges, ensu ing he
cos -e iciency o solu ions, and designing use - iendly in e aces. Cos -e iciency is
emphasized as a c i ical conside a ion, no only o op imizing ope a ions, minimizing
in as uc u e in es men s, and jus i ying ne wo k expansions, bu also o inco po a ing
decen alized eede s and echnologies in o exis ing sys ems and implemen ing dynamic
p icing s a egies.
• Elec ic G id Ope a o s: Al hough he HYPERGRYD ools align wi h he u u e needs o
elec ic g ids, di ec eedback om egional elec ic g id ope a o s was limi ed due o
collabo a ion challenges and egula o y complexi ies. Regional s akeholde s epo ed g id
o e loads and es ic ions on eed-in capaci ies because o he apid expansion o PV sys ems
o e he las yea . While ools like hose de eloped by KTH equi e eal- ime elec ici y
p icing o e ec i e pee - o-pee u ili y ading, dynamic a i s a e s ill in hei in ancy in he
egion. Technical equi emen s like he widesp ead ins alla ion o sma me e s a e almos
comple ed. Regula o y amewo ks ha accele a e he adop ion o lexible a i s and
incen i ize collabo a i e e o s wi h ene gy communi ies will be c i ical o unlocking he ull
po en ial o such ools.
Conce ning use case 2, he g id ope a o may no be he p ima y a ge o such ools,
because i o en delega es g id planning o in e na ionally enowned companies wi h
es ablished expe ise. These companies ypically ely on hei yea s o expe ience a he han
using ools such as SAIn .
• Resea che s and enginee ing o ices on he o he hand, p e e he usage o ools o e
se ices. The mos no able enginee ing o ice in he G oßschönau a ea, which also planned
he DHC ne wo k o G oßschönu, is “innoVAT GmbH”. This o ice has many yea s o
expe ience and s a ed planning g ids be o e specialized so wa e was a ailable. Du ing a si e
isi , he enginee o his company could desc ibe nume ous de ails abou he ne wo k design
based pu ely on expe ience. This is why hey a e s ill e using o use a so wa e and con inue
planning g ids o g id ex ensions only, elying on he 2D and 3D compu e -aided design
so wa e applica ion « Au oCAD » and Mic oso Excel o calcula ions. The alida ion o he
SAIn ool by enginee ing o ices would equi e mo e de ailed aining on he ool. Language
ba ie s migh also pose a p oblem, as Ge man-language ools end o be p e e ed.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
19
In summa y, he alida ed ools can guide sus ainable ene gy ansi ion s a egies and suppo
mode niza ion owa ds he 4 h and 5 h-gene a ion o DHC sys ems. The in eg a ion o lexible asse s,
such as hea pumps and decen alized RES, combined wi h ad anced ICT solu ions, has demons a ed
clea po en ial o imp o e ene gy e iciency, educe cos s, and enhance g id s abili y. Fu u e e o s
should add ess egula o y ba ie s, use educa ion, and cos -sha ing mechanisms o ensu e wide
adop ion and long- e m success.
3.2 KEZO
The KEZO Li e-in lab is loca ed in KEZO Resea ch Cen e in Jabłonna nea Wa saw. KEZO is one o he
mos mode n esea ch complexes dealing wi h he use o enewable ene gy in Poland. Cons uc ed
in 2015, i was mean o combine he unc ion o esea ch cen e, con e ence cen e and li ing
labo a o y. KEZO has a numbe o hea sou ces, enewable ene gy sou ces, ene gy s o age uni s o
bo h elec ici y and hea . S akeholde s a e IMP PAN and companies wo king in he ield o hea pump,
ene gy s o age, EV, PV, and o he ins i u es conduc ing expe imen al esea ch in Reseach Cen e .
Figu e 4 KEZO Reseach Cen e, iew on L5 building.
The KEZO Resea ch Cen e consis s o h ee buildings: 2 labo a o y buildings (B1, B2) and one main
building L5 ha houses o ices, addi ional labs, a con e ence oom and gues ooms. The cooling and
hea ing sys em consis s o ou main lines ha connec all buildings wi h hea and cold sinks, and
sou ces placed along he way o he main lines. Each building has i s own en ila ion, an coils,
adia o s and loo hea ing uni in which hea exchange s a e supplied om he main hea ing and
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
20
cooling lines. The e is no connec ion wi h he DHN o he ci y. KEZO possesses o e 180 kWp o PV in
a ious sys ems including on ca po s, acke s, building in eg a ed pho o ol aics, as well as a 12 kW
wind u bine and o e 60 m2 o sola collec o s. Sma elec ici y me e s ha e been ins alled in each
labo a o y o measu e and log ene gy consump ion.
Wi hin he HYPERGRYD p ojec , many modi ica ions we e made o he KEZO ene gy and H&C sys ems
o p epa a ion o connec ion he new Enabling Technologies. Fo de e mina ion H&C cha ac e is ics
o KEZO building and pe o mance maps o exis ing and new echnologies, addi ional ul asonic hea
me e s and ene gy me e s wi h MODBUS communica ion ha e been ins alled he e. Two o he new
Enabling Technologies de eloped wi hin he amewo k o Task 2.2, i.e. he modula Hea Pump wi h
PCM s o age and so p ion s o age, ha e been deli e ed o KEZO LiL o es ing. The i s s ep was o
in eg a e he echnologies and so wa e de eloped in WP2 and WP3 espec i ely wi h he exis ing
ene gy sys em in KEZO Li ing Lab and wi h he al eady ins alled selec ed echnologies o hea and
cold gene a ion. The wo ks we e ca ied ou based on he de ailed s a egy o es s and alida ion
de eloped in Task 5.1.
Figu e 5 Simpli y scheme o H&C sys ems in KEZO LiL
This demons a ion ac i i ies ha e been ca ied ou in Task 5.3, wi h ocus on he
de e mina ion o KPIs o benchma king o he HYPERGRYD solu ions and o alida e de eloped
models and con ol s a egies.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
21
The i s es ed de ice was he modula hea pump wi h PCM s o age o hea ing and DHW
pu pose, i was de eloped by Ochsne and AIT. The second de ice is he so p ion s o age de eloped
by SORTEC in coope a ion wi h CNR. Bo h de ices we e e y ad anced, bu also e y complica ed in
bo h echnical dimension and managemen .
The modula hea pump wi h PCM s o age was supplied om BTES s o age as he lowe hea sou ce
in 5 h GDHC sys em (Fig.6). The de ice was es ed in wo modes: DHW p oduc ion and hea ing mode,
o which he hea ing e iciency and COP was calcula ed. The cha ac e is ics o cha ging and
discha ging he PCM s o age we e de e mined. Based on he ob ained esul s, he echnologic
speci ic KPIs we e calcula ed, which we e hen compa ed wi h hose alues o adi ional hea
pumps. The calcula ed KPIs a e wo se in compa ison, which is due o oo a high empe a u e on
supply side o PCM s o age. Fo a ull phase ansi ion o RT57 in he PCM s o age, he e should be
an in low empe a u e o e 60°C; such empe a u es a e on he bo de o wha modula hea pump
can p o ide wi h minimal e iciency. Fo he whole cha ging p ocess o he PCM, he o al COP is
below 1.5.
In he cons uc ion o he modula hea pump, 2 addi ional pla e hea exchange s we e used. These
we e mean o p o ec he wa e ins alla ion o KEZO agains leaks and agains seeping o he RT57
in o he ci cui . Un o una ely, hose pla e hea exchange s we e causing addi ional esis ance in
ans e ing hea which caused wo sening o he ope a ional pa ame e s o he modula hea pumps.
Fo he hea ing mode he maximum COP was a ound 2. The PCM s o age has an a e age deg ee o
compac ness o 56% and a ene gy s o age capaci y o 4.17 kWh. The ene gy s o age e iciency was
ca. 85%.
Figu e 6 The modula hea pump connec ed o KEZO sys em.
The so p ion s o age was supplied wi h wa e o 80-85°C om he high empe a u e CO2 hea pump
(Fig.7). Fo p ope ope a ion o his de ice, he e has o be an addi ional ci cui o consuming hea ,
i.e. he a e age empe a u e ci cui MT, wi h a supply empe a u e o 30-35°C om e.g. hea ing
sys em, and low empe a u e ci cui which would ake cold wa e a 12-16°C o cooling he buildings.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
22
The s o age was es ed in wo modes: 1) hea s o age, 2) p oduc ion o hea and cold simul aneously.
The ene gy s o age capaci y is 8 kWh and he maximum ene gy s o age e iciency was abou 67.3%.
The ime o cha ging he es ed s o age is 1 h 46 min wi h a cha ge powe o 4.7 kW. A ull discha ge
akes 1h 29 min wi h 5.4 kW o discha ge powe . The s o age is based on so p ion and deso p ion
p ocesses, so i can be used as a mid- e m s o age.
Figu e 7 So p ion s o age connec ed wi h Mayekawa CO2 Hea Pump in KEZO Resea ch Cen e
Technology speci ic KPIs o he modula hea pump wi h PCM s o age and so p ion s o age
we e de e mined based on es s ca ied ou in KEZO LiL wi hin Task 5.3. These KPIs a e lis ed in Table
6, he las column con ains he esul s o he calcula ion based expe imen al da a.
Table 1 Technology speci ic KPIs panel
Technology/
Tool
KPI
Index
Symbol
Resul s
Modula hea pump
COP hea ing
KET 4.1
COPhea
1.92
COP cooling
KET 4.2
COPcool
NA
COP DHW wi h PCM losses
KET 4.3
COPDWH
1.44
COP DHW wi hou PCM losses
KET 4.4
COPPCM
1.69
PCM s o age
Deg ee o compac ness
KET 5.1
ΦPCM
56%
Ene gy s o age capaci y
KET 5.2
ESCPCM
4.17 kWh
Ene gy s o age densi y
KET 5.3
ESDPCM
50.1 kWh/m3
Ene gy s o age e iciency
KET 5.4
ηPCM
85.5%
DHW cha ging cycles pe day
KET 5.5
DHWcycles
NA
Cha ge/discha ge ime
KET 5.6
chPCM,
disPCM
1 h 6 min
15 min
So p ion s o age
Ope a ing empe a u e le els
KET 6.1
TchSOR,
TdisSOR
85°C
13°C
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
23
Ene gy s o age capaci y
KET 6.2
ESCSOR
8 kWh
Ene gy s o age e iciency
KET 6.3
ηSOR
67.3%
Cha ge/discha ge powe
KET 6.4
Qchnom,SOR,
Qdisnom,SOR
4.7 kW
5.4 kW
Cha ge/discha ge ime
KET 6.5
chSOR, disSOR
1 h 46 min
1h 29 min
The nex demons a ion ac i i ies we e pe o med es s o he exis ing and he new enabling
echnologies in a ie y con igu a ion acco ding o de ined Use Cases:
• UC1. Simula ion o building, ho el and e ia y loads in small dis ic s.
Using his o ical da a collec ed ia he KTH edge panel, p edic ion models we e de eloped o es ima e
he hea ing and cooling loads o he KEZO buildings. These models enable he iden i ica ion o he
key pa ame e s in luencing hea ing and cooling demands wi hin he esea ch acili y, while also
acili a ing he ene gy low assessmen wi hin he local sec o coupling sys em. In his deli e able,
he e will be a pa icula ocus on he hea ing demand since mos o he da a was collec ed in he
win e season. Below, is a wo-day ahead p edic ion o he hea ing load co esponding o he
adia o s and an coils in he Building L5 using Ga ed Recu en Uni , which is a ype o ecu en
neu al ne wo k (RNN).
Figu e 8 Two-day ahead p edic ion o he adia o s and an coils hea ing powe in building L5 using GRU app oach
Below, he p edic ion pe o mance o he NN model o hea ing COP p edic ion a e displayed, along
wi h he p edic ion e o me ics.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
24
Figu e 9 Tempo al a ia ion o COP hea ing: measu emen and NN
Figu e 10 Reg ession cu e o he COP hea ing p edic ion using NN
• UC2. DHC wi h coo dina ed ope a ion o hea pumps, so p ion cooling, s o ages, and PV
p oduc ion.
Fo his use case, he ene gy lexibili y a ailable wi hin he KEZO emula ed DHC sys em, enhanced by
he newly ins alled ha dwa e, will be le e aged o op imize ene gy low managemen . In he
demons a ion phase, he so p ion ene gy s o age se es as he p ima y lexibili y mechanism o
egula ing ene gy lows ela ed o hea ing and cooling demands. Since his s o age sys em is powe ed
by a CO₂ hea pump du ing he es ing phase, i s sec o coupling po en ial is also conside ed. To his
end, he abili y o pe o m load shi ing and sha ing h ough he cha ging and discha ging cycles o he
adso p ion s o age will be analyzed and simula ed using load p edic ion da a om UC1.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
25
Figu e 11 Model-based con ol o he adso p ion s o age using he KTH edge panel
• UC3. Real da a-d i en simula ion o mul i-ca ie DHC.
Based on da a collec ed ia he edge panel, i was possible o de elop da a-d i en models o he CO2
hea pump and o he so p ion s o age. La e , hose models, along wi h he hea ing/cooling demands
and PV p oduc ion, a e used o he simula ion o elec ic and he mal powe lows wi hin he coupled
local g ids a KEZO.
Based on he indings om UC1 and UC2, UC3 was success ully alida ed by in eg a ing load p edic ion,
sys em models, and con ol s a egies in o a comp ehensi e da a-d i en simula ion o he KEZO
emula ed sec o coupling ne wo k. The p ima y objec i e o his simula ion was o assess he accu acy
o he de eloped model by analyzing p edic ion e o s, e alua ing simula ion compu ing ime—
e lec ing he e iciency o ad anced ene gy managemen algo i hms—and de e mining he op imal
cos s associa ed wi h sa is ying hea ing and cooling demands h ough op imal con ol s a egies.
To achie e his, KTH conduc ed digi al simula ions o he KEZO LiL's a ge ed HVAC sys em unde
a ying ou doo (wea he ) and indoo (occupan ) condi ions. The ocus was no on simula ing he KEZO
emula ed DHC ne wo k as a whole, bu a he on e alua ing he impac o op imally con olling he
inno a i e ha dwa e echnologies de eloped wi hin HYPERGRYD, pa icula ly he so p ion s o age, on
he ene gy low wi hin he HVAC sys em o Building 5 and he ene gy lows esul ing om ha . This
app oach enables a deepe unde s anding o he in e ac ion be ween he op imally d i en ene gy
s o age solu ions and sec o -coupled ene gy dis ibu ion.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
32
Figu e 17. Dis ibu ion o HP ac i a ion du a ions du ing cheap and expensi e pe iods unde RBC and MPC wi h TOU
elec ici y p icing.
Figu e 18. Compa ison o ene gy s o ed pe cha ging cycle and numbe o cha ging unde RBC (blue) and MPC ( ed) wi h
TOU elec ici y p icing
The desc ibed MPC decisions educe dependency on expensi e pe iods and enhance cos sa ings,
allowing o he same load o be sa is ied as he RBC. Al hough he o e all elec ici y consump ion
inc eases by 2.3 kWh (2.4%) and he sys em COP dec eases by 0.3 (5.8%), mainly due o he HP
ope a ing a lowe COP du ing o e cha ges, he ope a ing cos (elec ici y) signi ican ly educed om
18.9 € (RBC) o 10.44 € (MPC), ep esen ing sa ings o abou 44.79%.
Tes 2: Pe o mance assessmen wi h ealis ic elec ici y p ice
The es 2 condi ions a e he same as hose in es 1, wi h one single di e ence: while in es 1, he
elec ici y p ice conside ed is ic i ious, he e i is a ealis ic one. I a ies be ween 0.11 and 0.27
€/kWh, acco ding o he a ia ion in he “PUN Index GME” (sho o P ezzo Unico Nazionale, o
Na ional Single P ice) o e h ee days in Decembe 2024 (11-14). The “PUN Index GME” is he
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
33
e e ence index o he I alian elec ici y ma ke , i.e. he e e ence index o elec ici y aded in he
Day-Ahead Ma ke (MGP).
Figu e 19 p esen s he esul s o Tes 2, which yield simila ou comes o hose o he p e ious es s.
He e, he a e age elec ici y p ice (0.19 €/kWh) se es as he h eshold conside ed in he analysis o
di e en ia e be ween expensi e pe iods (p ices highe han his h eshold) and cheap ones (p ices
lowe han his h eshold, g ey a eas o he ou h plo ). The MPC s a egy ac i a es he HP mo e
equen ly du ing pe iods o low elec ici y p ices, aking ad an age o lowe cos s. Du ing hese
imes, MPC ex ensi ely uses he HP o maximize TES cha ging. In con as , RBC shows he highes
le els o HP ope a ion du ing peak elec ici y p ice pe iods, closely ollowing he load pa e n. This
di e ence is due o MPC’s abili y o p edic u u e ends, including elec ici y p ices and load peaks,
enabling i o make mo e in o med and op imal decisions. An example is be ween 2:00 and 4:00 on
day 1, when he elec ici y p ice was low, he MPC s a egy p edic ed an upcoming he mal load peak
in he ea ly mo ning. Consequen ly, he ank se poin was inc eased o i s maximum alue, allowing
he sys em o supe cha ge he ank in an icipa ion o he demand. A ound midday, he elec ici y
p ice d opped again, p omp ing he MPC o cha ge he ank u he . Howe e , i is wo h no ing ha
he ank's he mal capaci y also played a signi ican ole in he decision-making p ocess. Fo ins ance,
he MPC ac i a ed he HP a 18:00 on he i s day despi e he high elec ici y p ice due o a he mal
load peak and insu icien s a e o cha ge in he ank o mee he demand.
Figu e 19. Resul o compa ison he pe o mance o MPC s. RBC wi h ealis ic elec ici y p ice ( es 2)
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
34
The o al du a ions o HP ac i a ion we e simila in he wo cases, wi h 1199 minu es o RBC and
1207 minu es o MPC. Figu e 20 shows ha unde RBC, he HP ac i a ions we e almos e enly
dis ibu ed be ween cheap (48.5%) and expensi e (51.5%) pe iods, indica ing no esponsi eness o
p ice a ia ions. The sys em ope a ed solely based on he mal demand wi hou conside ing elec ici y
p ices. In con as , he MPC wi h a a iable p icing s a egy signi ican ly shi ed ope a ions owa d
cheape pe iods. App oxima ely 67.19% o he HP ac i a ions occu ed du ing low-cos pe iods, while
32.81% occu ed du ing high-cos imes. Figu e 21 gi es an insigh in o he numbe o cha ges and
he ene gy s o ed pe cha ge. As in he p e ious es , he numbe o cha ges inc eased e en mo e,
eaching 37 sho e and mo e equen ac i a ion pe iods. I should be no ed ha no explici e m
was included in he objec i e unc ion o penalize o con ol he numbe o HP ac i a ions, which
likely con ibu ed o his beha iou . This app oach enables he hea pump o ope a e mo e lexibly,
adap ing e ec i ely o he a iable condi ions o he elec ici y p ice and load by s o ing ene gy in
smalle inc emen s, mos ly be ween 2 and 10 kWh. O e all, he MPC s a egy esul ed in a 6.86%
educ ion in ope a ing cos s, om 18.07 € o 16.83 €.
Figu e 20. Dis ibu ion o HP ac i a ion du a ions du ing cheap and expensi e pe iods unde RBC and MPC wi h ealis ic
elec ici y p icing.
Figu e 21. Compa ison o ene gy s o ed pe cha ging cycle and numbe o cha ging unde RBC (blue) and MPC ( ed) wi h
ealis ic elec ici y p icing
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
35
3.3.4 Lessons lea n
MPC consis en ly ou pe o med RBC by lowe ing elec ici y cos s and achie ing no able sa ings. The
analysis shows ha MPC e ec i ely shi ed TES cha ging o pe iods wi h lowe p ices while mee ing
load demands. Howe e , he inc eased ac i a ion cycles sugges a dynamic esponse o changing
condi ions. To imp o e his, i may be help ul o add a e m o he objec i e unc ion o educe hese
cycles.
The mos c i ical insigh s de i ed om he esea ch ac i i y on he desc ibed DH subs a ion and
ad anced con ols a e:
• MPC e ec i ely u ilizes dynamic p icing and load shi ing by scheduling hea pump ope a ions
du ing pe iods when elec ici y cos s a e lowe and op imizing TES cha ging s a egies. This
app oach no only le e ages dynamic elec ici y p icing o educe ope a ing cos s bu also shi s
ene gy consump ion o o -peak imes. As a esul , i enhances o e all sys em e iciency and
imp o es demand-side managemen .
• Con inuous model aining ensu es sys ems adap o changing condi ions and emain accu a e,
main aining op imal pe o mance o e ime.
• Accu a e he mal load o ecas ing is c i ical o e ec i e MPC. Employing ANN o de elop a ROM
p o ides p ecise and adap able o ecas s, enhancing sys em esponsi eness.
• E iciency imp o emen s h ough a ious subs a ion schemes: By explo ing di e en subs a ion
schemes, i is possible o enhance he sys em's e iciency. Implemen ing empe a u e-based
ansi ions and combining hys e esis me hods acili a ed smoo he and mo e e ec i e changes
be ween ope a ional schemes, ul ima ely leading o imp o ed subs a ion e iciency.
Al hough his s udy has demons a ed signi ican bene i s om using MPC in decen alized
subs a ions, he e is s ill po en ial o u he imp o emen . The ollowing a e sugges ions o
enhancing he sys em and di ec ions o u u e esea ch based on ou indings:
• Longe i y cons ain s o equipmen : While MPC educed elec ici y cos s, i may also esul in a
highe HP swi ching a e, which can lead o inc eased wea on equipmen o e ime. This could
be inco po a ed as a cons ain in he MPC's decision-making p ocess.
• Explo e ad anced machine lea ning echniques o model aining, including ein o cemen
lea ning, o enhance model adap abili y and p edic ion accu acy ac oss a ious condi ions.
• Op imize he sizing and ope a ion o TES: Analyse he e ec s o a ious TES sizes and
con igu a ions on sys em pe o mance o enhance TES design o speci ic applica ions and
imp o e ene gy s o age u iliza ion.
• In eg a e enewable ene gy sou ces: In es iga e he in eg a ion o enewable ene gy sou ces,
including sola he mal and pho o ol aic (PV) panels, along wi h ba e y s o age, in o he MPC
amewo k o enhance sus ainabili y and educe dependence on g id elec ici y.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
36
• Expand he s udy o inco po a e cooling applica ions: This esea ch p ima ily concen a ed on
hea ing applica ions. Fu u e in es iga ions could b oaden he ocus o include cooling ope a ions
wi hin he dis ic hea ing ne wo k. Implemen ing model p edic i e con ol (MPC) in cooling
sys ems could enhance pe o mance and e iciency h oughou he yea , o e ing comp ehensi e
solu ions o bo h hea ing and cooling equi emen s.
• Expand he analysis o encompass he en i e hea ing season and conside a ious clima es: The
s udy ocused on h ee ypical win e days o a speci ic clima e o expe ience he MPC ope a ion
and i s ad an ages. By ex ending he analysis o he en i e hea ing season and conside ing a ious
clima es, i is possible o in es iga e he ope a ion o he MPC unde di e en load condi ions
and achie e seasonal cos sa ings.
• Expand he s udy by conside ing a iable DH condi ions: he MPC esul ed in he e ec i e
u ilisa ion o elec ici y dynamic p icing and load shi ing in cases o cons an condi ions a he
DHN side. Howe e , bo h DH he mal ene gy cos s and empe a u es can a y h oughou he
day and season, which a ec s he o e all sys em e iciency and ope a ing cos s. Fu u e
in es iga ions could conside hese aspec s and hei co ela ion wi h a ious subs a ion
ope a ing schemes by inco po a ing hem in o he MPC decision-making p ocess.
3.4 ENVI
The ENVI Li e-In Lab, loca ed a En i onmen Pa k in Tu in, se es as an inno a ion hub dedica ed o
ecological ansi ion and sus ainable de elopmen . En i onmen Pa k is a echnology pa k in Tu in
dedica ed o sus ainable inno a ion and ecological ansi ion. I hos s companies, esea ch cen e s,
and labo a o ies ocused on enewable ene gy, ene gy e iciency, and sus ainable mobili y. As pa
o he HYPERGRYD p ojec , i s ene gy moni o ing in as uc u e suppo s da a collec ion on
elec ici y, hea , and wa e consump ion o ICT ool de elopmen and alida ion.
Figu e 22 Li e-in lab En i onmen Pa k (Tu in-I aly)
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
37
In pa icula , he complex consis s o 10 buildings, including 5 o ice buildings, 4 labo a o ies, and a
can een, all in e connec ed by an in e nal dis ibu ion ne wo k o elec ici y, hea ing, and cooling.
Figu e 23 Main buildings wi hin En i onmen Pa k
Wi hin he HYPERGRYD p ojec , he lab has been u ilized o alida e a mul i-ca ie ene gy model
aimed a in eg a ing he mal and elec ical ne wo ks wi h enewable ene gy sou ces. The sys em
includes a Dis ic Hea ing and Cooling (DHC) ne wo k, hea pumps, pho o ol aic ins alla ions, a
hyd opowe sys em, and all c ucial componen s o es ing he e ec i eness o he de eloped
solu ions. This demons a ion is pa o Task 5.4, which ocuses on alida ing he mul i-ca ie ene gy
dynamic model a ENVI.
The me hodological app oach was s uc u ed a ound se e al key ac i i ies o ensu e a obus
alida ion p ocess. Task 5.4 played a cen al ole in coo dina ing he demons a ion, while Task 3.5
con ibu ed by de eloping he modeling ool applied in he ENVI LiL.
The p ocess began wi h eal- ime da a acquisi ion om he SCADA sys em, allowing con inuous
moni o ing o ene gy sys em pe o mance. The in eg a ion o a digi al win, de eloped using ENCO’s
SAIn ool, enabled he simula ion o a ious ope a ional scena ios, p o iding de ailed insigh s in o
he in e ac ions be ween he mal, elec ical, and enewable ene gy sou ces.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
38
Figu e 24 Model o he medium and low ol age elec ic dis ibu ion sys em o ENVI in ENCO. Red poin s ep esen nodes
wi h elec ic demands
Figu e 25 Mod Model o he he mal dis ibu ion ne wo k o En i Pa k in SAIn . Nodes desc ibe he supply o demand
maximum hyd aulic p essu e
Ac i e s akeholde in ol emen , including eedback om ENVI ne wo k ope a o s and echnology
p o ide s, was essen ial o e ining he model’s usabili y and adap ing i o eal-wo ld condi ions.
The alida ion p ocess in ol ed mul iple ne wo k con igu a ions, s ess- es ing di e en ope a ional
s a egies unde a ying ene gy demand and supply scena ios. The eedback loop wi h WP4 -
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
39
HYPERGRYD Digi al Twin Pla o m as a Se ice ensu ed con inuous imp o emen s o he modeling
ool.
The alida ion a ENVI LiL ocused on h ee main use cases, each add essing c i ical challenges in
mul i-ca ie ene gy ne wo k managemen :
• Islanded Mode Ope a ion: his scena io es ed he abili y o ENVI o unc ion au onomously,
minimizing eliance on he ex e nal g id. The esul s demons a ed ha a combina ion o
hyd opowe and pho o ol aics signi ican ly con ibu ed o sel -su iciency while ensu ing g id
s abili y.
• T ansi ion o 4 h-gene a ion DHC ne wo k: he implemen a ion o e e sible hea pumps and
op imized he mal dis ibu ion s a egies esul ed in inc eased e iciency and a educ ion in
ene gy losses. Lowe supply empe a u es imp o ed o e all sys em pe o mance.
• In eg a ion o mul iple ene gy ca ie s: he demons a ion e alua ed he combined ope a ion o
CHP and he mal s o age, enhancing lexibili y and enabling mo e e ec i e load balancing
be ween elec ici y and hea ne wo ks. The esul s con i med ha a mul i-ca ie app oach
allows o be e demand-side managemen and g id op imiza ion.
The demons a ion a he ENVI Li e-In Lab success ully alida ed mul i-ca ie ene gy models,
con i ming hei po en ial o seamless in eg a ion o enewable ene gy sou ces wi h he mal and
elec ical ne wo ks. The indings unde sco e he impo ance o ad anced ICT ools, pa icula ly digi al
wins, in op imizing hyb id ene gy g ids. Collabo a ion among s akeholde s, d i en by he s uc u ed
ac i i ies o Task 5.4 and Task 3.5, p o ed essen ial in ensu ing echnical iabili y, economic easibili y,
and o e all sys em e iciency in mode n Sma Hyb id G ids.
Addi ionally, as emphasized in deli e able D5.5, he a ailabili y and u iliza ion o accu a e da a play a
decisi e ole in enhancing sys em pe o mance, making eal- ime da a managemen an indispensable
elemen o u u e ene gy communi ies.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
40
4 Guidelines o eplica ion
4.1 Modula HP wi h PCM s o age (ins alled in KEZO LiL)
4.1.1 Pu pose
The PCM-in eg a ed hea pump sys em is a demons a o o a new eme ging echnology aimed a
small houses and apa men s. The s udies ha e demons a ed ha coupling hea pumps wi h PCM
s o age can lead o signi ican ene gy sa ings and imp o ed sys em pe o mance, ou esea ch, o
ins ance, indica es ha in eg a ing PCM s o age wi h hea pumps can enhance he Seasonal
Coe icien o Pe o mance (SCOP) by up o 13%, depending on he building's hea ing demand. The
echnology demons a o shows he echnology is easible and bene icial a he scale o households,
wi h he ollowing ad an ages and disad an ages:
Ad an ages
• The modula hea pump is designed o egula e powe and hea ing capaci y in a wide ange, made
possible hanks o he modula aspec which allows o disconnec indi idual hea pump modules.
• Regula ion o e iciency is acili a ed by he possibili y o swi ch each comp esso on and o ,
which elimina es he need o use expensi e solu ions, such as e.g. equency in e e s.
• Sho - e m hea s o age based on phase change ma e ial RT57 (PCM) can s o e he hea o DHW,
allowing i o be used o peak-sha ing and peak-shi ing when he e is high hea wa e demand,
e.g. in he e enings.
• Combining he modula hea pump wi h PCM s o age gi es mo e possibili ies o physically i he
de ice in a limi ed space hanks o i s compac o ma : wi h a base o 70 by 70 cm, he de ice has
a heigh o 2.3m. In his con igu a ion he equi emen is a oom o heigh min. 2.5m.
• Elec ically, he modula hea pump uses 3 phases and is capable o equally loading phases.
• PCM s o age is smalle han wa e TTES s o age, so i s main cus ome g oup a e la s, small
houses and o ices, which ha e hei own hea ing sys em and DHW p oduc ion.
• Inc eased eliabili y and edundancy: pa allel connec ion o he HP modules allows each module
o wo k independen ly. I one uni would ge damaged he o he s can wo k and ensu e
con inuous hea ing o cooling.
• Scalabili y and lexibili y: Adding addi ional modules in his pa allel con igu a ion is ela i ely
easy, which allows o adjus he sys em o changing hea ing o cooling equi emen s wi hou he
need o do big changes o he in as uc u e.
• Inc eased e iciency du ing pa ial loads: Wo king wi h many smalle uni s migh be mo e
e icien in case o pa ial load. The uni s migh swi ch on o o depending on he needs,
main aining op imal e iciency and educing he ene gy usage.
D awbacks and ba ie s
• Inc eased ini ial in es men : he pu chase o he sys em is mo e cos ly conside ing he equipmen
and ins alla ion.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
41
• Complexi y o sys em managemen : coo dina ing he ope a ion o mul iple uni s needs ad anced
managemen sys ems o ensu e equalised dis ibu ion o loads and p e en sho cycles.
• Spa ial equi emen : each addi ional module akes some physical space. While each uni is a he
compac , loca ions wi h e y limi ed space may es ic he numbe o modules. In addi ion, while
compac , he en i e cons uc ion is s ill a he bulky and po en ially p oblema ic o physically i
h ough doo s, e c. This could be mi iga ed, h ough e.g. on-si e assembly, bu he economic
easibili y o his would need o be e i ied.
• Regula ion issues and sa e y: R290 (p opane) is classi ied as a lammable cooling agen , which
equi es ollowing s ic sa e y no ms. Fo example, in ou doo uses he olume o he R290 agen
is limi ed o 152 g ams pe 1 cooling ci cui o minimize he dange o combus ion.
• To summa ise, he modula hea pumps in pa allel con igu a ion ha e many ad an ages, such as
inc eased esilience, scalabili y and e iciency, bu hey also ha e some challenges like highe
ini ial cos s, inc eased complexi y o he sys em, space limi a ion and ele a ed sa e y
equi emen s. Ca e ul planning and design a e equi ed o maximise he ad an ages, while a he
same ime educing he p oblems.
4.1.2 Obse ed po en ial echnical ba ie s o eplica ion
Du ing he es ing phases o he demons a o , se e al echnical aspec s and issues, some o which
need o be conside ed while o he s may be a ba ie o eplica ion and need speci ic a en ion o
p ac ical implemen a ions, we e obse ed. These obse ed aspec s and issues a e
• PCM ma e ial and i s cha ac e is ics. An app op ia e PCM ma e ial should be selec ed depending
on he hea pump and ecipien pa ame e s, because he RT 57 used he e has a phase ansi ion
empe a u e o app ox. 57 °C, which de ines ha he ci cui has o be supplied wi h wa e o 62-
65 °C. This is oo high as a pa ame e o a modula hea pump, causing ine ec i e condense
cooling and hus high condensa ion p essu e. This leads o o e hea ing o he e ige a ion
comp esso s. Too high condensa ion p essu e signi ican ly educes he e iciency o he hea
pump COP.
• Hea exchange in he PCM s o age. I is necessa y o selec and design a sui able hea exchange
in he PCM s o age made o a ma e ial o he han aluminium as he aluminium pla e hea
exchange poses challenges a he s age o p oduc ion and ope a ion. The p ocess o welding
pla es in he hea exchange is ime-consuming, expensi e, and equi es a good specialis in his
ield. Thick aluminium shee s we e used, which cause a signi ican inc ease in he weigh o
aluminium and a limi a ion/ educ ion o he amoun o PCM ma e ial in he exchange . Du ing
ope a ion, he aluminium pla es may also leak and PCM may leak in o he wa e . The
compa ibili y o he PCM ma e ial wi h he ma e ials used in he hea exchange is also impo an .
The exchange mus be cheap a he p oduc ion s age, du able du ing ope a ion and compa ible
wi h he PCM ma e ial.
• Moni o ing he empe a u e a he supply o he PCM s o age. The PCM s o age mus main ain
an app op ia e empe a u e ange o ope a e p ope ly. I he sys em o e hea s o ope a es
ou side he desi ed empe a u e ange, he PCM may no change phase as in ended. I is
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
48
Ene gy Communi ies (LECs), which suppo s he mode niza ion o dis ic hea ing sys ems o 4 h and
5 h GDHC.
Demons a ion and alida ion o he ICT ools o he managemen o hea and ene gy ne wo ks in
he local ene gy communi ies we e ca ied ou a he Li e-in-Lab SONNE in Aus ia. The main
objec i es we e o op imize ene gy lows and in eg a e enewable ene gy sou ces (RES) in o
in eg a ed hea and elec ici y ne wo ks. The key poin s and bene i s esul ing om he
implemen a ion o ICT ools and he analyses pe o med a e as ollows:
• The use o ICT ools such as BIM-GIS, exe goeconomic ools, AI, and edge-compu ing
algo i hms enabled accu a e modeling o he DHC ne wo k and RES. Thanks o simula ions
using ICT ools, eal- ime da a om PV sys ems, and ene gy p ices, pee - o-pee (P2P)
ading es s we e conduc ed. These showed signi ican bene i s o local ma ke
pa icipan s.
• The simula ion o he local ene gy ma ke showed ha he ac i a ion o P2P ading
con ibu ed o imp o ing he elec ici y and hea cos s by 145% and also inc eased sel -
su iciency (up o 124% in summe due o PV p oduc ion). The use o in elligen ading
s a egies, such as i ual hea pumps, yielded educed cos s and signi ican ly imp o ed sel -
su iciency a es.
• The use o ICT ools enabled he analysis o a ious scena ios in elec ici y, gas and hea
ne wo ks, which con ibu e o he educ ion o hea losses, he cos s o using de ices and
acili a es he in eg a ion o dis ibu ed RES sys ems.
• Con inuous eedback om p ojec pa icipan s, including households, municipali ies, DHC
ope a o s and elec ici y g id ope a o s, has allowed o he imp o emen o HYPERGRYD
ools and se ices. Collabo a ion wi h di e en s akeholde s has e ealed impo an
echnical and non- echnical aspec s ela ed o he implemen a ion o solu ions.
• The esul s indica e ha , despi e he in e es in sus ainable ini ia i es, he e a e ba ie s
such as adminis a i e complexi y and high up on cos s ha can hinde ull pa icipa ion in
ad anced ene gy managemen sys ems. Unde s anding use needs, simpli ying
implemen a ion and anspa en communica ion o bene i s a e key o engaging local
communi ies.
• Add essing egula o y ba ie s, use educa ion and cos -sha ing mechanisms a e essen ial o
ensu e wide adop ion and long- e m success o inno a i e ene gy solu ions.
The conclusions o wo k ca ied ou wi hin Task 5.6 a ENVI Li e-In Lab in he Tu in En i onmen al
Pa k unde line he impo ance o inno a ion in sus ainable de elopmen and ecological ansi ion.
The p ojec alida ed he e ec i eness o ene gy models and ools based on mul iple ene gy ca ie s,
which enable he in eg a ion o enewable ene gy sou ces in o he mal and elec ical ne wo ks. The
analysis ca ied ou a ENVI LiL p o ided aluable in o ma ion on h ee key ope a ional scena ios:
au onomous sys em ope a ion, ansi ion o a ou h gene a ion hea ing ne wo k and in eg a ion o
di e en ene gy ca ie s. The esul s showed ha he app op ia e combina ion and mix o PV and
hyd opowe wi h ene gy s o age con ibu es o he sys em’s sel -su iciency, while he use o hea
pumps and hea dis ibu ion op imiza ion s a egies inc eases e iciency and educes ene gy losses.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
49
Wi hin he HYPERGRYD p ojec , a Digi al Twin Pla o m was implemen ed o SONNE and ENVI LiL. I
was de eloped by IDP o imp o e he usabili y and s abili y o di e en ools de eloped by GSY, KTH,
GET and ENCO. The pla o m is a digi al win o eal hea ing and elec ical g id wi h his o ical da a,
eal- ime da a and con inuous eedback o simula ion o he LEC.
Demons a ion ac i i ies we e ca ied ou wi hin in Task 5.3 a KEZO Li e-in-Lab in Jabłonna in Poland,
ocused on de ining KPIs o benchma king new echnologies and alida ing de eloped models and
con ol s a egies. The es ed echnologies, including a modula hea pump wi h PCM s o age, and
so p ion s o age, showed many possible applica ions. The main conclusions om hese s udies a e
as ollows:
1. Modula hea pump wi h PCM s o age:
• The wa e sou ce hea pump (WSHP) should ha e he bo ehole s o age (BTES) as he lowe
hea sou ce. Such a solu ion is p oposed in he new gene a ion 5 h GDHC sys ems.
• The es s o modula HP showed a low e iciency due o: i) addi ional hea exchange s
causing hea losses; ii) high inle empe a u e a he PCM s o age (abo e 60°C) which is a
he maximum empe a u e achie able wi h he hea pump wi h R290; he maximum alue o
COP achie ed du ing hea ing mode was app oxima ely 2.0, and he o al COP du ing cha ging
mode o PCM ob ained below 1.5.
• I was shown ha he PCM s o age is cha ged wi h ela i ely low powe and o a long pe iod
o ime o app ox. 1 hou , bu i has he abili y o quickly discha ge o p o ide DHW a up o
15 l/s. The maximum ene gy s o age e iciency in he PCM eached a ound 85.5%.
2. So p ion s o age:
• The so p ion s o age mus be supplied om a high- empe a u e sou ce HT wi h a empe a u e
a ound 70-90°C (in he KEZO case he supplied empe a u e was 80-85°C. I is also necessa y o
p o ide he possibili y o he discha ge o medium- empe a u e hea MT, e.g. o he building
hea ing sys em wi h an coils o loo hea ing, and ensu e a eed o he low- empe a u e sys em
LT, e.g. in o de o cool he acili y. This is because he HT and MT ci cui s mus ope a e du ing
he cha ging o he s o age (when he so p ion p ocess occu s), while du ing discha ging phase i
is necessa y o ope a e he MT and LT ci cui s (deso p ion p ocess).
• The so p ion s o age has 2 ope a ing modes: i) as a he mal s o age, ii) simul aneous as a hea
s o age and hea and cold p oduc ion.
• The s o age capaci y was 8 kWh, and he maximum ene gy s o age e iciency eached 67.3%. I
was shown ha he ime o ully cha ging he so p ion s o age is app ox. 1.5 hou s wi h a powe
o app ox. 5kW; simila pa ame e s we e ob ained du ing he discha ging phase.
• The abili y o s o e he mal ene gy bo h sho - e m and long- e m was demons a ed.
ICT ools o manage and op imize he ope a ion o he modula hea pump o he PCM s o age and
he so p ion s o age enables g ea e lexibili y in ene gy and hea managemen . Such ac ions
signi ican ly con ibu e o educing CO2 emissions and inc easing ene gy e iciency, which suppo s
sus ainable de elopmen goals. Tes esul s and ob ained KPIs indica e a eas o u he op imiza ion,
which may lead o imp o ing he echnologies and inc easing hei e iciency in he u u e. In
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
50
summa y, he demons a ion ac i i y conduc ed a KEZO LiL p o ided aluable in o ma ion on he
pe o mance o mode n solu ions in hea ing and cooling sys ems, highligh ing bo h hei po en ial
and he challenges ha need o be o e come o maximise he bene i s o hei applica ion in
in eg a ed ene gy sys ems.
Based on he analyses ca ied ou a he Eu ac Li e-in-Lab in Bolzano, I aly, wi hin he
amewo k o Task 5.3, key in o ma ion on he pe o mance and po en ial imp o emen s o Model
P edic i e Con ol (MPC) o decen alized Dis ic Hea ing (DH) subs a ions was p o ided. The main
conclusions d awn om he analysis a e as ollows:
• MPC ou pe o med he con en ional Rule-Based Con ol (RBC) app oach by educing elec ici y
cos s and achie ing signi ican sa ings. This was achie ed by shi ing The mal Ene gy S o age
(TES) cha ges o pe iods o lowe elec ici y p ices while s ill mee ing powe demand. MPC
e ec i ely uses dynamic p icing and load shi ing o op imize he ope a ion o hea pumps and
TES. This educes ope a ing cos s and shi s ene gy consump ion o o -peak pe iods, imp o ing
o e all sys em e iciency and s eamlining demand-side managemen .
• Con inuous subs a ion componen ’s model aining ensu es he sys em adap s o changing
condi ions and emains accu a e, main aining op imal pe o mance o e ime. Addi ionally,
accu a e he mal load o ecas ing is c ucial o e ec i e MPC ope a ion. Using a i icial neu al
ne wo ks (ANN) o de elop a educed-o de model (ROM) enables accu a e and adap i e
o ecas s, he eby imp o ing sys em esponsi eness and e iciency.
• Resea ch sugges s ha explo ing di e en subs a ion schemes, such as empe a u e-based
ansi ions and he use o hys e esis me hods, imp o es sys em e iciency by acili a ing
smoo he ope a ional ansi ions. These imp o emen s con ibu e o be e subs a ion e iciency
and educed ope a ing cos s.
• The e is a need o op imize he size and ope a ion o he mal s o age o imp o e he sys em
e iciency and pe o mance. Mo eo e , in eg a ing enewable ene gy sou ces wi h he mal and
elec ical s o age in MPC can inc ease sus ainabili y and educe he dependence on g id
elec ici y.
In summa y, he s udy highligh s he signi ican bene i s o MPC in DH sys ems, especially in e ms o
cos educ ion, e iciency and adap abili y. Howe e , he e a e se e al a eas o imp o emen ,
including equipmen du abili y, enewable ene gy in eg a ion, and expansion in o cooling
applica ions, sugges ing ha u u e esea ch should ocus on hese aspec s o u he enhance sys em
pe o mance.
The HYPERGRYD p ojec has success ully demons a ed he po en ial o ad anced ICT ools
o he op imiza ion and in eg a ion o enewable ene gy sou ces in hea and elec ici y ne wo ks.
Valida ion o hese ools con ibu ed o a be e unde s anding o he challenges and oppo uni ies
ela ed o he implemen a ion o mode n ene gy solu ions. The in ol emen o s akeholde s and
con inuous exchange o in o ma ion played a key ole in imp o ing he models and adap ing hem o
eal condi ions. Collabo a ion be ween p ojec eams, as well as he use o ad anced ICT ools such
as digi al wins, we e essen ial o achie e echnical easibili y and economic iabili y o he solu ions.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
51
The conclusions om his p ojec p o ide a solid basis o u u e ac ions aimed a ans o ming ene gy
sys ems owa ds g ea e e iciency, cos educ ion and imp o ed g id s abili y.
D5.8 Ou comes o HYPEGRYD demons a ion, lessons lea n and guidelines o eplica ion
52
6 Re e ences
G id Singula i y (2024): Le e aging hea pump lexibili y in local ene gy ma ke s.
h ps://g idsingula i y.medium.com/le e aging-hea -pump- lexibili y-in-local-ene gy-ma ke s-
ac20874345a (Accessed on 23.11.2024).
