Communications, Cyber-Physical Security, …smartgridcenter.tamu.edu/sgc/doc/Communications_Cyber-Physical...Communications, Cyber-Physical Security, Sensors, Embedded Systems Chair:

Communications, Cyber-Physical Security, Sensors, Embedded

Systems

Chair: Mr. Igor Alvarado, National Instruments Corp.

Co-Chair: Dr. P.R. Kumar, TAMU

Room 2500

Research Topics

• Cyber-Physical Systems for the Smart Grid – Communications – Control – Sensors/Actuators – Embedded Systems – Cyber-Security – Testbeds – Co-design, Co-Simulation tools – Education/Training/Certification

• Integration with other services/technologies and humans: – Data Analytics – Networking – Visualization, – Smart Grid – Microgrid integration, etc. – HMIs, Haptics, AR, VR

Cyber-Physical Systems (CPS), Smart Grids and Microgrids

• Building blocks

– (Embedded) Control

• Embedded Systems

• Sensors/Actuators

– Communications

– Computation

– Cyber-security Advances in CPS research can be accelerated by close collaborations between academic disciplines in computation, communication, control, and other engineering and computer science disciplines, coupled with grand challenge applications (such a Smart Grids and networked microgrids). NSF

Image courtesy of S. Stein, et al, University of Oslo

CPS Definition

• Cyber–physical systems (CPS) are the next generation of engineered systems, that require tight integration of computing, communication, and control technologies to achieve stability, performance, reliability, robustness, and efficiency in dealing with physical systems of many application domains.

• CPSs are typically required to adapt to various changes in internal and external factors: – adaptation by switching between operation modes

– hybrid automaton, linear hybrid automata, etc.

– Real-time scheduling

Source: P.R. Kumar, et al, TAMU

Multi-Personality, Software-Defined Hardware: Reconfigurable Intelligent Electronic Devices (IEDs)

Inverter, Battery, Capacitor Control

Sectionalizer Control

Power Quality Analyzer

Metering

Transformer Monitoring

Recloser Control

Demand Response

Substation Automation

Phasor Measurement Unit

Alarm Event Recorder

Microgrid Control

Testbeds for Rapid Prototyping, Validation, Training….

Sensing and Actuation

(Embedded) Control Communications

Computation

Communications

What are the key challenges?

• Embedded Systems – RTOSes, Firmware, Modularity, Reconfigurability, Reliability, MTTR, Tech

Specs, Open Source SW/HW, Virtualization

• Control – Classic vs. Other (ANNs, Fuzzy, Genetic Algorithms, etc.), Model-based vs.

Measurements-based, Adaptive vs. Fixed, Predictive

• Sensors/Actuators – Resolution (ADC/DAC), Sampling Speed (S/s), connectivity, calibration, bus-

powered, new interfaces, etc.

• Communications: – Wired/Wireless/Networked/Self-healing/Protocols, Topologies, etc.

• Computation: – Programming tool, Solvers, AI/Machine Learning, Cognition – Real-time vs. Off-line, Centralized vs. Decentralized

• Cyber-security – GPS Spoofing, High-Impact/Low-Frequency Risks, Resilience to attacks

What research topics would you propose ? With what priority?

• Embedded Systems

• Control

• Sensors/Actuators/Perception

• Communications

• Computation/Cognition

• Cyber-security

Embedded Systems

Processor(s), RTOS Reconfigurable HW (FPGA)

PT

12

0V

inp

ut

CT

5A

inp

ut

DIO

GP

S

ENET

Po

rt

Exp

ansi

on

LEA

Inp

ut

IRIG

Co

m P

ort

ENET

Po

rt

RS-

23

2 P

ort

RS-

48

5 P

ort

AN

SI D

evic

e A

lgo

rith

ms

Dat

a R

edu

ctio

n

Res

amp

ling

Dig

ital

/Rel

ay C

on

tro

l

HM

I In

terf

ace

PM

U

C3

7.1

18

:20

11

Web

Ser

vice

s

PQ

A

61

00

0-4

-30

GP

S/Ti

me

Syn

c.

Cu

sto

m P

roto

cols

Logg

ing

CO

MTR

AD

E

IEC

61

85

0

DN

P3

USB

Po

rt

Modular I/O and connectivity

Memory

Comm. Bus

SW

HW

Example of an Embedded Controller with I/O Modules

CPU + FPGA + Memory Comms. Ports

I/O Modules

I/O Chassis

Cyber-Security

• Cyber-security needs to be integrated with system theory to guarantee resilience of the grid.

• What do we need to:

– Make systems more secure

– Guidelines:

• Prevent, Deter, Detect, Respond

• Prior/During/After, Adaptability to scenarios

• Are current guidelines from NIST, ISA, NERC, IEEE (other) enough?

Are CPS Testbeds important for your research?

• Testbeds can help to create bridges between theory and practice, design and implementation

• How important are they to test: – sensors/actuators – embedded controllers – communication systems – computing platforms and algorithms – security schemes/technologies/approaching – Smart Grid/Microgrid topologies – Systems-level performance, integration (Smart

Grids/Microgrids/Hybrids)

Source: SmartAmerica Challenge, NIST

Example Architecture w/ Testbeds

Education/Training: A System

Design Approach

• How important is Education, Training, Certification and Workforce Development?

• What are the key areas to focus on?

• Should Research initiatives in Energy CPS put more emphasis on education, training, certification, etc.?

Education/Training: A System Design Approach

Intro to

Engineering

Intro to

Controls

(Power)

Circuits

Signal

Processing/

Computation

Sensors &

Actuators

Advanced

Control

Embedded

Systems

Design

RF/Comms.

Smart Grid/

Microgrids

Integration with other services/technologies and humans

• Data Analytics

• Networking

• Visualization,

• Smart Grid – Microgrid integration

• HMIs, Haptics, AR, VR

Grant Proposals & Collaborations

Academia & Nat’l Labs

Synergies

Points-of-Engagement

Overall Plan

Prioritized Project list

Project

Ranking & Selection

Thrust Areas

Industry

•Project Summaries: •Director or PI , Co-PIs •Problem Statement •Current State of Practice & Research •Approach and Method •Industry Sector Impacted •Deliverables •Project Plan: tasks, project duration, type, budget….

Conclusions

– Points discussed: • Embedded Systems

• Communications

• Control

• Sensors/Actuators

• Cyber-Security

• Education/Training/Certification as part of Research Grant Proposals

– Not discussed (due to time constraints) but mentioned: • Testbeds

• Co-design, Co-Simulation tools

Conclusions (Cont’d)

• Studying the benefits (e.g. performance, resilience, etc.) of using more intelligent devices in the field, closer to the I/O signals

• How to tackle what is really an interdisciplinary problem?

• The use of reconfigurable “general purpose” hardware with software tools that can be used by “domain experts”

• How to estimate the states of the system, classify them, etc., specially for large, complex systems with thousands of control loops, variables…. in real-time

Conclusions (Cont’d)

• The use (and security) of different kinds of networks (e.g. sensor networks, controller network, etc.)

• As the Grid becomes more “digital”, how to control/keep track of software modifications, versions, patches, etc.

• How to extract information from more data, from more sensors, from controllers… in the network

• Extremely large distributed control system: how to provide stability, robustness...

Conclusions (Cont’d)

• Impact of distributed, decentralized control system in minimizing communications

• Broad time-scales, time-delays involved • System-level design tools that cover different

design components (mechanical, EM, electrical, thermal, communications, computation, control, etc).

• Hybrid system validation, verification, model-checking of hybrid (discrete/continuous) designs

• The use of High-Performance Computing tools as part of the CPS

Conclusions (Cont’d)

• Cyber-security of single networks vs. coupled networks, resilience to cyber-attacks

• Transitioning from legacy systems to new controller/communications technologies

• Predictability: MPC and other adaptive control approaches that can improve performance predictability under changing operating conditions; reliability via redundancy or other means.

• Education: – UG: adapting 4-year program to a “system design” focus; new

books; new content; course compression; more depth needed – Grad: more depth needed; “system design” approach with an

interdisciplinary approach

Backup Slides

667 MHz Dual-Core ARM Cortex-A9 processor

28K Logic Cells (Artix-7)

80 DSP slices, 16 DMA channels

92 Billion calculations per second

Xilinx

ZYNQ

New Embedded Technologies

Source: Xilinx

Resilience Construct

Source: NERC

Response to Severe Events

Source: NERC

Source: NERC

Source: NERC

Communications

Facilities

Source: NERC

Source: IEEE