Intelligent Energy Management

Abstract

Intelligent energy management is the basis for economical and low-emission operation of decentral energy systems. This presentation gives a brief overview on the basics of an intelligent energy management system (iEMS) and explains some of its aspects (e.g., operational strategies, modeling and simulation) using concrete examples. An industrial-scale real-world laboratory was set up at Fraunhofer IISB, where, among other things, many aspects of an iEMS were implemented and demonstrated. In the framework ToSyCo an approach for the control of energy systems was developed. Two levels have been defined, which divide the operational strategy into a global level (storage schedules, load forecasts, etc.) and a local level (for real-time control, security, etc.).

Full text

© Fraunhofer
INTELLIGENT ENERGY MANAGEMENT
Intelligent Energy Systems / Energy Technologies
Dr. Christopher Lange, Dr. Richard Öchsner, Johannes Geiling, Alexandra Rueß
public
6/13/2022
1

© Fraunhofer
CONTENT
◼Motivation ►
◼Basics ►
◼Real-world laboratory ►
◼Energy system optimization ►
◼TotalSystemControl ►
◼Summary ►
◼Publications ►
◼Contact ►
public
2

© Fraunhofer
MOTIVATION
◼Motivation ►
◼Basics ►
◼Real-world laboratory ►
◼Energy system optimization ►
◼TotalSystemControl ►
◼Summary ►
◼Publications ►
◼Contact ►
public
3

© Fraunhofer
Motivation
◼Typical energy infrastructures in industry consists of different energy sectors
◼Electrical (AC und DC)
◼Heating and Cooling
◼Gas (e.g., natural gas, hydrogen)
◼Other (e.g., compressed air, vacuum etc.)
→The sectors are coupled via different plants (e.g., heat pump, CHP, chiller)
◼Intelligent energy management →Consideration of all relevant couplings between the components
and combination of different intelligent operational strategies
◼Peak shaving
◼Efficiency increase
◼Self-consumption optimization
public
Main goals:
Reduction of CO2emissions and energy costs
cf. [IISB19]
4

© Fraunhofer
BASICS
◼Motivation ►
◼Basics ►
◼Real-world laboratory ►
◼Energy system optimization ►
◼TotalSystemControl ►
◼Summary ►
◼Publications ►
◼Contact ►
public
5

© Fraunhofer
Basics
Load profiles
public
◼Load profile: time course of the purchased electrical power, other name: load curve
◼Load profiles are composed of
◼Base load: Purchase that is always present,
depending on seasonal and calendar variables.
◼Peak load: Short-term high purchase that
differs significantly from the base load
◼Low load: Purchase, which is closer to the base load than to the peak load
◼High load: Purchase, which is closer to the peak load than to the base load
◼There are different definitions in literature, e.g.:
J. L. Mathieu, P. N. Price, S. Kiliccote et al. „Quantifying Changes in Building Electricity Use, With Application
to Demand Response“. IEEE Transactions on Smart Grid 2.3 (2011), S. 507–518. DOI: 10.1109/TSG.2011.2145010.
P
t
peak load
high load
low load
base laod
cf. [Lan20], [Lan21a]
6

© Fraunhofer
Basics
Peak shaving
public
◼Initial situation: Load profile with the 15 min mean values of the power purchased from the energy
supplier
◼Definition of a purchase limit
(maximum allowed mean power
per 15 min period)
◼Limit can be constant or time variable
◼Load peak becomes visible
◼Highest load peak in the billing period
is relevant for electricity costs
◼Maximum 15 min average value
should be reduced to purchase limit
◼1 min values may exceed this
Load profile
power
time
Load profile with 15 min mean values
15 min
Load peak
Purchase limit (15 min)
cf. [Lan20], [Lan21a]
7

© Fraunhofer
Basics
Efficiency increase
◼Shift operation points to time periods with higher efficiency →energy storages needed
◼Example: Chiller and recooling plant
◼Chiller’s efficiency (COP*) depends on thermal power
◼Recooling plant’s efficiency depends on ambient temperature and relative humidity
*coefficient of performance: COP = Qth / Pel
COP* of chiller
based on
cooling load
Cooling capacity of
recooling plant based
on ambient
temperature and
relative humidity
-15
0
15
30
0
25
50
75
100
0
175
350
525
700
Air temperature in °C
Fan stage 1 (dry)
Relative
humidity in %
Recooling capacity in kW
50
100
150
200
250
300
350
-15
0
15
30
0
25
50
75
100
0
175
350
525
700
Air temperature in °C
Fan stage 1 (wet)
Relative
humidity in %
Recooling capacity in kW
100
200
300
400
500
600
public cf. [Pul19a], [Pul19b]
8

© Fraunhofer
Basics
Self-consumption optimization
◼Optimization of self-consumption (e.g., energy of RES*)
◼Charge storage, if surplus locally generated energy is available
◼Discharge storage, if demand exceeds production
public *renewable energy sources
public grid Battery energy
storage system
discharging
import
Consumer
(e.g., factory)
photovoltaic plant
chargingexport
own-consumption
charging
KPIs
Self-supply rate SSR:
𝑆𝑆𝑅 = 1 − 𝐸𝑖𝑚𝑝𝑜𝑟𝑡
𝐸𝑑𝑒𝑚𝑎𝑛𝑑
Own-consumption rate OCR
𝑂𝐶𝑅 = 1 − 𝐸𝑒𝑥𝑝𝑜𝑟𝑡
𝐸𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛
9

© Fraunhofer
REAL-WORLD LAB
◼Motivation ►
◼Basics ►
◼Real-world laboratory ►
◼Energy system optimization ►
◼TotalSystemControl ►
◼Summary ►
◼Publications ►
◼Contact ►
public
10

© Fraunhofer
public
DC microgrid
Heat Hydrogen
Cold
Real-world lab at Fraunhofer IISB as research and demonstration platform for
sector coupling using complete institute
AC microgrid
cf. [Och19], [Mar19]
11

© Fraunhofer
Real-world lab at Fraunhofer IISB
Energy storages
◼Energy storage systems at Fraunhofer IISB
◼Electrical energy storages (stationary and mobile Lithium-Ion, Redox-Flow)
◼Thermal energy storages (hot- and cold-water storages)
◼Hydrogen storages (pressure, chemical)
public
Redox-flow container
with IISB’s control system
Hot water
storages
Cold water
storage
Mobile Lithium-Ion Battery Chemical hydrogen storage
system based on LOHC
cf. [Och19], [Mar19]
©Kurt Fuchs /
Fraunhofer IISB
© Fraunhofer IISB © Fraunhofer IISB © Fraunhofer IISB © Kurt Fuchs /Fraunhofer IISB
12

© Fraunhofer
Real-world lab at Fraunhofer IISB
CPH, heat pump and cold storage
◼Application CHP* combined with heat storage
◼Increase of energy self-generation (heating, electrical power)
◼Utilization for peak shaving with algorithms and control
developed at IISB
◼Application heat pump
◼Utilization of residual heat in exhaust air systems
◼Load shedding during electrical load peaks
◼Application cold storage
◼Increasing efficiency
◼Increase of utilization of free cooling
◼Peak load reduction
public *combined heat and power plant cf. [Lan21b], [Mar19]
©Kurt Fuchs / Fraunhofer IISB
©Kurt Fuchs / Fraunhofer IISB
13

© Fraunhofer
Real-world lab at Fraunhofer IISB
Hydrogen technology –system integration
◼Fuel cell, electrolyzer, hydrogen storage
◼Development and investigation of hydrogen systems
◼Characterization of fuel cells and electrolyzers
◼Development of component models and simulations
◼Model based development of operational strategies
Hydrogen test bench:
Gas mixture and conditionings system
Measurement data of a fuel cell
experiment
LOHC container for investigation on long-
term storage of electrical energy
050 100 150 200 250 300
0
10
20
30
40
50
60
Spannung [V], Wirkungsgrad [%]
Strom [A]
0
2000
4000
6000
8000
10000
12000
Leistung [W]
Spannung

Leistung
public cf. [Ste18]
© Kurt Fuchs / Fraunhofer IISB © Kurt Fuchs / Fraunhofer IISB
14

© Fraunhofer
Real-world lab at Fraunhofer IISB
Application platform for hydrogen
public
Fuel cell
Electrolyser
Hydrogen
Pressure storage
Battery
e-motor drive drain
Drive drain
Vehicles
Stationary applications
Mobile applications
Public grid
Hydrogen
DC grid
AC grid
Waste gas
LOHC Reactor
uncharged LOHC
charged LOHC
Photovoltaic
LOHC
Test bench
Application platform: control algorithms based on simulations
Source: Fraunhofer IISB
cf. [Gei21]
15

© Fraunhofer
ENERGY SYSTEM OPTIMIZATION
◼Motivation ►
◼Basics ►
◼Real-world laboratory ►
◼Energy system optimization ►
◼TotalSystemControl ►
◼Summary ►
◼Publications ►
◼Contact ►
public
16

© Fraunhofer
Energy system optimization
Simulation
◼Advanced energy system simulation
◼Data analysis methods and tools
◼Model library for energy-related plants
and components, e.g., energy storages,
generators like heat pump, chiller, CHP,
PV, aqua-thermal plant etc.
◼Dimensioning algorithms for
applications like peak shaving and
increase of self-supply etc.
→Non-invasive optimization of energy systems
considering all relevant energy sectors
→Scenario-based study of adjustments and
extensions
public
Free Loadprofile-Analysis-
Tool, available online:
https://www.proenergie-
bayern.de/de/veroeffentli
chungen.html
Simulation tool for the energy
building infrastructure based on
a component library as well as
intelligent operational strategies
(project: ProEnergie –Bayern)
cf. https://www.energy-seeds.org/
17

© Fraunhofer
Energy system optimization
Data-based modelling of energy system components
◼Model library for energy system components
◼Data-based approach and use of AI
◼Training of model with historical measurement data
◼Automated process for easy transfer to other applications
public
Train and
optimize model
data Preprocess and splitting of data
Trained model
Validate model
Training data
Validation data
Comparison of measurement and two distinct models
Power
Measurement
model SVR_1
model SVR_2
cf. https://www.energy-seeds.org/
18

© Fraunhofer
Energy system optimization
Automation
◼Implementation of innovative plant concepts for energy systems
◼Realization of demonstrators and prototypes
◼Implementation of control and measurement functions according to IEC 61131
◼Plant characterization and functionality validation
Heat pump system for heat
recovery from exhaust air
Efficient connection to cooling
network via transmission station
Large cold storage as intelligent
component in the cooling network
public
Overview screen for visualization of a
CHP plant (HMI)
cf. https://www.energy-seeds.org/
© Fraunhofer IISB © Fraunhofer IISB © Fraunhofer IISB© Fraunhofer IISB
19

© Fraunhofer
TOSYCO
◼Motivation ►
◼Basics ►
◼Real-world laboratory ►
◼Energy system optimization ►
◼TotalSystemControl ►
◼Summary ►
◼Publications ►
◼Contact ►
public
20

© Fraunhofer
TotalSystemControl
Framework for an intelligent energy management system (iEMS)
public
Global level
✓Load prognosis
✓Superior operating strategy
✓Schedules for storage operation
✓Peak shaving
Local level (PLC*)
✓Safety functions
✓Basic functionality, autonomous operational strategy
✓Physical inputs and outputs
✓Operating unit for the plant
*programmable logic controller
Electrical Energy
Storage
(BESS)
Cold Thermal
Energy Storage
(CTES)
Combined Heat
and Power Plant
(CHP)
Heat Pump
System
(HPS)
TotalSystemControl (ToSyCo)
communication, data archiving, storage scheduler,
load management, peak shaving, prognosis
Weather
Prognosis
(external)
Hydrogen
Storage System
(HSS)
Redox-Flow
Battery
(RFB)
Global level
Local level
cf. [Lan19]
21

© Fraunhofer
TotalSystemControl
Operational strategies
◼Example: Extension of a heat/electricity-controlled CHP unit to include peak shaving functionality
◼Division of the thermal energy storage into four virtual zones, reserved zone for peak shaving
◼Intelligent operational strategy (including FSM) for retrofitting existing CHP systems
◼Battery system for increasing the dynamics (e.g., start-up process of the CHP)
public
Measurement result for
peak shaving with CHP
and battery. 18 %
reduction of peak
*finite state machine
SOCmax
SOCps
SOCmin
Peak active 0 1 0 1
CHP on 1 1 0 0
charge discharge
Full
Peak shaving
operation
Normal and
peak shaving
operation
Empty
= starting
points
cf. [Lan21a]
22

© Fraunhofer
TotalSystemControl
Load forecasting
◼Methods
◼Time series models (ARIMA)
◼Artificial Neural Networks
◼Machine learning algorithms
◼Standalone implementation
◼Automated process
◼Integration in predictive control algorithms
◼Accuracy:
◼Electrical load: MAPE 5%
◼Thermal load: MAPE 10 %
public
Parameter determination
Prognosis algorithm
Historical data Data preprocessing
prognosis data
Load forecast
Cooling load
Forecast
Thermal power in kWEl. power in kW
Production plant
Forecast of cooling load (top), which is strongly shaped by the operation
of a production plant (bottom)
cf. [Mar19]
23

© Fraunhofer
SUMMARY, PUBLICATIONS, CONTACT
◼Motivation ►
◼Basics ►
◼Real-world laboratory ►
◼Energy system optimization ►
◼TotalSystemControl ►
◼Summary ►
◼Publications ►
◼Contact ►
public
24

© Fraunhofer
General approach and summary
public
Data analytics
Data science and statistics
Diagrams, key figures Algorithm developm ent
Operational strategies
Energy flow control
Use of Artificial Intelligence
Implementation and optimization
Implementation of operational strategies
Optimization of plants and parameters
Monitoring
Measuring and logging
data processing
System optimization
Efficiency & profitability
Sim ulation
Component models
Plant dimensioning
System simulation, scenarios
25

© Fraunhofer
◼Scientific papers and books
[Gei21] J. Geiling, M. Steinberger, F. Ortner, R. Seyfried, A. Nuss, F. Uhrig, C. Lange, R. Öchsner, P. Wasserscheid, M. März, P. Preuster.
„Combined dynamic operation of PEM fuel cell and continuous dehydrogenation of perhydro-dibenzyltoluene”. International
Journal of Hydrogen Energy 46.72 (2021), S. 35662 –35677. ISSN: 0360-3199. DOI: 10.1016/j.ijhydene.2021.08.034
[Lan20] C. Lange, A. Rueß, A. Nuß, R. Öchsner, M. März. „Dimensioning battery energy storage systems for peak shaving based on a
real-time control algorithm“. Applied Energy 280 (2020), 115993. ISSN: 306-2619. DOI: 10.1016/j.apenergy.2020.115993.
[Och19] R. Öchsner, A. Nuß, C. Lange, A. Rueß. „Research Platform: Decentralized Energy System for Sector Coupling“. Chemical
Engineering & Technology 42.9 (2019), S. 1886–1894. DOI: 10.1002/ceat.201800714.
[Pul19a] P. Puls, C. Lange, R. Öchsner. „Hybrid Cooling Towers in a Free‐Cooling Application: Modeling and Field Measurement
Verification“. Chemical Engineering & Technology 42.9 (2019), S. 1871 –1878. DOI: 10.1002/ceat.201800712.
[Mar19] M. März, R. Öchsner (Hrsg.). „Innovative Technologien für intelligente dezentrale Energiesysteme. Stuttgart: Fraunhofer
Verlag (2019). ISBN: 9783839614860.
public
Relevant publications (1)
26

© Fraunhofer
Relevant publications (2)
◼PhD-Thesis
[Lan21a] C. Lange. „Energiesektoren-übergreifende Lastspitzenreduktion mit elektrischen und thermischen Energiespeichern“. PhD-
Thesis, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) (2021). URN: urn:nbn:de:bvb:29-opus4-169778.
[Pul19b] P. Puls. „Simulationsgestützte Effizienzoptimierung von industriellen Kaltwassersystemen mit thermischen Speichern“. PhD-
Thesis, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) (2019). URN: urn:nbn:de:bvb:29-opus4-108054.
[Ste18] M. Steinberger. „Verstromung von wasserstoffreichen Gasgemischen mit PEM-Brennstoffzellen am Beispiel einer
Epitaxieanlage“. PhD-Thesis, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) (2018). ISBN: 978-3-8439-3887-7.
◼Presentations
[Lan21b] C. Lange. „BHKW des Jahres 2020. BHKW mit Wärmespeicher und Batterie zur Strom-/Wärmeversorgung sowie
Lastspitzenreduktion “. Presentation. BHKW 2021 –Innovative Technologien und neue Rahmenbedingungen, 09.11.2021 –
10.11.2021, Magdeburg (2021). DOI: 10.13140/RG.2.2.26423.80803.
[IISB19] IISB. "Energiesysteme neu denken“. Symposium. Presentations (2019). Available online:
https://www.iisb.fraunhofer.de/en/press_media/events/eroeffnung_bau_b.html (access: 15.03.2021).
[Lan19] C. Lange, A. Nuß, A. Rueß, R. Öchsner. „Total System Control (ToSyCo) for Peak Shaving and Efficiency Enhancement“.
Presentation. International Renewable Energy Storage Conference (IRES), 12.03.2019 –14.03.2019, Düsseldorf (2019). DOI:
10.13140/RG.2.2.13002.03520.
public
27

© Fraunhofer
Contact
Fraunhofer Institute for Integrated Systems
and Device Technology IISB
Schottkystr. 10
91058 Erlangen, Germany
www.iisb.fraunhofer.de
www.energy-seeds.org
public
Founded by:
Project website for ProEnergie
PROENERGIE -BAYERN
Effizienz- und Flexibilitätsgewinn durch
Optimierung von Betriebsstrategien der energetischen Gebäudeinfrastruktur
basierend auf prognostizierten Energiebedarfen der Produktion
Dr.
-Ing. Christopher Lange
Dr.
-Ing. Richard Öchsner
Johannes Geiling
+49 (0) 9131 761
-107
+49 (0) 9131 761
-116
+49 (0) 9131 761
-488
Send mail
to C. Lange
Send mail to R. Öchsner
Send mail to J. Geiling
28