Development of cyber-physical system resilience

Development of cyber-physical system resilience
using AI and ML: Case study for Vaasa harbour
grid
CR-DES Project
Mike Mekkanen
17.1.2022
WP4

Outline
▸ Cyber-physical systems CPS Critical Infrastructure CI
▸ Digital Twins
▸ AI and ML to energy grid
▸ Case study “Real-time testing of a battery energy storage controller for
harbour area smart grid: A case study for Vaasa harbour grid”
▸ Conclusions and Future Work

Critical Infrastructure
▸ Defined of CI “are engineered systems that
are built from, and depend upon, the
seamless integration of computation and
physical components.”
▸ Complex, flexible, cross-domain, need for
interoperability, impact on physical world,
▸
▸ Guaranteeing the necessary of the level of
protection and resilience of CI and essential
ICT- and OT-based services; https://link.springer.com/book/10.1007/978-3-030-00024-0

Typical Critical Infrastructure Protection
▸ Identification of critical
sectors/subsectors. At this stage, the
sectors and/or subsectors that are
considered important for national
interests are identified, whose
disruption, failure or destruction
would have a significant impact on
public health, public and civil matters,
the environment, security, social and
economic well-being

▸ Identification of critical services. At the
central level critical (or vital) services of
the sector/subsector are identified and
designated for each critical sector, then
evaluated, with specific criteria, and
prioritized to give the final list of
national critical services. For each
critical service, a list of stakeholders-
operators is drawn up, from which (or
in cooperation with) a list of the most
critical goods, products and systems
supporting this service is extracted. This
process, despite its importance, is not
obvious because of the
interdependencies between the service
that the infrastructure supports and
other services within the same or a
different sector.

▸ Designation of CI. For each critical
service, the critical assets/components
that comprise the CI are identiﬁed and
designated, critical element with direct
impact that may cause due to its
malfunction or loss thereof (ﬁrst-order
effect), second-order effects are
understood either as dependencies,
where a critical element depends on
another element or as
interdependencies where two critical
elements are mutually affected at
national or even transnational levels

▸ Protection of the CI. Procedures for
protection and security are
implemented for each CI. There is no
one-size-fits-all solution for all CI
(holistic protection ); each
sector/subsector has unique
requirements and security
countermeasures that must be
developed and assessed on a regular
basis.

Digital Twins
▸ Digital twins: It is a replica of the physical system and beyond

AI and ML to energy grid
▸ The growing of the Renewable Energy Resources (RES), poses key
challenges for power system operators, since RES are characterized by
variability and intermittency, making it difficult to predict their power
output.
▸ The main approaches to settle these issues for providing stability and
flexibility are the integration of fast-acting supply, demand side
management, and energy storage services etc.
▸ This will require that energy grids to enter the digital era,
New technologies, real-time monitoring and control, as well as
cyber-security of energy assets can result in more efficient,
secure, reliable, resilient, and sustainable

▸ AI approaches have been identified as a key tool for addressing these
challenges in energy grid.
▸ AI can be used to forecast power demand and generation, optimize
maintenance and use of energy assets, understand better energy usage
patterns, as well as provide better stability and efficiency of the power
system etc.
▸ AI can also alleviate the load on humans by assisting and partially
automating the decision-making, as well as automating the scheduling
and control of the multitude of devices used

▸ Various AI techniques used for Demand-Side Response and their classification
shown un this work, reviews of over 160 papers, 40 companies and
commercial initiatives, and 21 large-scale projects.
https://www.sciencedirect.com/science/article/pii/S136403212030191X

Case study: first version offline simulation
▸ Smart Control of Battery Energy Storage System in Harbour Area Smart
Grid: A Case Study of Vaasa Harbour
Authors: Jagdesh Kumar; Hussain Sarwar Khan; Kimmo Kauhaniemi
had been IEEE EUROCON-2021 Lviv, Ukraine, July 6 - 8

Case study: first version offline simulation
▸ Smart Control of Battery Energy Storage System in Harbour Area Smart
Grid: A Case Study of Vaasa Harbour
Authors: Jagdesh Kumar; Hussain Sarwar Khan; Kimmo Kauhaniemi
had been published at the IEEE EUROCON-2021 Lviv, Ukraine, July 6 - 8

Case study: second version Software-in-the-Loop
(SIL)
▸ Real-time testing of a battery energy storage controller for harbour
area smart grid: A case study for Vaasa harbour grid
Authors: Jagdesh Kumar; Mike Mekkanen; Mazaher Karimi; Kimmo Kauhaniemi
has been accepted and selected for presentation at the 2022 58th IEEE Industrial and Commercial Power System
Conference, which is scheduled to be held on May 1-5, 2022. conference website: (
https://www.openconf.org/IAS/2022ICPS/openconf.php)
▸ Software is connected to the real-time
simulation using a communication
protocol IEC 61850 GOOSE

Case study: second version Software-in-the-Loop
(SIL)
Authors: Jagdesh Kumar; Mike Mekkanen; Mazaher Karimi; Kimmo Kauhaniemi
has been accepted and selected for presentation at the 2022 58th IEEE Industrial and Commercial Power System
Conference, which is scheduled to be held on May 1-5, 2022. LAS VEGAS, NV conference website: (
https://www.openconf.org/IAS/2022ICPS/openconf.php)
Purpose
▸ Development and optimization of
control applications
▸ Automated testing and validation of
different case studies defined by grid
operators
▸ Resiliency investigations (e.g., threats
impact at the controller/grid)

Case study: second version

Pgrid • Pdemand SOC ” 10%
SOC • 90%
Yes
No Yes Charge
battery
No No
Yes
Yes
No No
Yes
SOC • 90%
SOC ” 10%
Battery
idle mode
Charge
battery
Discharge
battery
Discharge
battery
Battery
idle mode
Subscribing GOOSE for analysing
data (PD, PG, and SOC)
Start
End
Publishing GOOSE with
commands (0,1,-1)
Compile and run BESC file
Create BESC project file with
predefined SCD
Design SCD file
Configuring SCD file for BESC
Control algorithm for BESC

Battery energy storage at SOC=25% Battery energy storage at SOC=75%

Conclusions and Future Work
▸ Developing and validating the battery energy storage controller (BESC)
performance based on the SIL Real-time testing
▸ Develop the IEC61850 communication protocol GOOSE at the OPAL-RT
real-time simulator as a subscriber and publisher from and to the (BESC)
▸ Develop the (BESC) controlling algorithm that has been implemented as
SIL
▸ The results shows that the balance of active power and local power
demand at harbour can be maintained by charging and discharging the
battery energy storage

Conclusions and Future Work
▸ Develop the HIL Real-time testing of a the use case
study
▸ Develop the multi agent hardware battery energy
storage controller (BESC)
▸ Add smart/secure functionality to the multi agent
controller by monitoring its environment and also
connecting more DERs and different loads
▸ Develop different testing scenarios (cyber attacks)
and analyze their impacts on the power network
operation by monitoring its physical parameters

Development of cyber-physical system resilience

More Related Content

Similar to Development of cyber-physical system resilience (20)

Recently uploaded (20)

Development of cyber-physical system resilience