Artificial Intelligence

Applications in Power
Systems
Om P. Malik

IEEE SA & NC Sections, May 7/8, 2018

 What is Artificial Intelligence?

Artificial Intelligence (AI) has recently emerged as a science

even though it may still be considered in its early stages of
development.
Depending on the goals and methods employed in
research, its definition varies. As a broad description, it
may be described as the science of making machines do
things that would require intelligence if done by humans.
AI applications are now being considered in a very wide
variety of disciplines, ranging from humanities to natural and
applied sciences. In the context of power systems, application
of artificial neural networks (ANNs) and fuzzy logic is
commonly referred to in the literature as AI applications in
power systems.

Over the past 25 years or so, feasibility of the application of

AI for a variety of topics in power systems has been explored
by a number of investigators. Topics explored vary from load
forecast to real-time control and protection, and even
maintenance.
Artificial Neural Networks
Natural Nerve Cell
Artificial Nerve Cell
 Networks Based on Artificial Nerve Cell
Model
- Multi-layer feed-forward
perceptron
- Recurrent
- Radial basis function
- Adaline
- Bayesian
- Hopfield
- Boltzman
- Kohonen
- Generalized Regression network
 Types of Neuron Models
• Artificial Neuron Cell Model
• Multiplicative neuron
Reacts to product of activation of pairs of
synapses
• Generalized neuron
Contains both summation and aggregation
functions with sigmoid and Gaussian activation
functions
 Simple and Generalized Neuron Models
Aggregation
Thresholding
Function Function

Bias s
s_ b ias
B ia s

1 f1
  2 Out pu t Opk
O u t p pk
ut O
Inputs,  f2
Input Output Xi
 DEFERENT ANN MODELS

Input  hidden  output

 ANN layer layer layer
(Conventional ANN)

  ANN Input hidden  output

layer layer layer

Input  output
  ANN  hidden
layer layer
layer

Input output
  ANN hidden
layer layer
layer

10
 Training of Neural Networks
Neural networks need to be trained. Based on the
type ofnetwork, it may be:
• Supervised learning
• Unsupervised learning
• Competitive
Although most networks are trained off-line using
available data, in some cases the weights can be up-
dated on-line in real-time to track the system operating
conditions.
 Neural Network Controllers

Copying an existing controller with a network. Inverse plant modeling using a network.

Back propagating through a forward model of the plant.

 Bayesian Networks
A Bayesian network (BN), also known as a Bayesian belief network, is a
graphical model for probabilistic relationships among a set of variables.
They have a qualitative component represented by the network
structure and a quantitative component represented by the assignment
of the conditional probability (CP) distributions to the nodes of the
network.
BNs can learn from observations. Learning of BNs can be parameter
learning and structure learning. With parameter learning, the structure
of the BN is given and only the CP parameters are learned. With
structure learning, the BN structure itself is learned. Bayesian learning
calculates the probability of each of the hypotheses given the data.
Insulation Deterioration Estimation of a Transformer
Using a Bayesian Network
Insulation LoL estimation by BN versus other methods for unit #64
Classical Direct Torque Control of an
Induction
Motor
Udc

Switching Table
abc to

3 2
4 1
5 6
+
Sector calculator
-
+ IP +
-
-
Torque and Flux Estimation
ANN Based DTC of an Induction Motor
Udc

ANN
abc to

N
3 2
4 1
5 6
+
Sector calculator
-
+ +
SMC
-
-
Torque and Flux
Estimation
 PI-DTC versus ANN-SMC-DTC
180

160

140 DTC-ANN-SMC
Reference
Rotor Speed (rad/sec)

120 DTC
X: 0.1507 160
100 Y: 149.2
150 158
X: 0.3459
80 Y: 149.2
156
100
60
154
40 50
152
20
0 150
0 0 0.2 0.4 0. 0 0.5 1 1.5 2
6
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time (sec)
Block Diagram of an Adaptive Controller
Controller Structure with MLFF NNs
 Neuro-Adaptive PSS

Response to a three phase

to ground fault, p=0.7 pu,
pf=0.62

Table 1:Dynamic Stability Margin* for Different Stabilizers.

OPEN CPSS NAPSS
Maximum Power 2.65 pu 3.35 pu 3.60 pu
Maximum Rotor Angle 1.55 rad 2.14 rad 2.36 rad

* Dynamic Stability Margin is defined as the maximum power output at which the generator loses
synchronism while input torque reference is gradually increased
ADALINE Network as an Identifier
Radial Basis Function Network
RBF-Identifier & Pole-Shifting Controller
Stability Margin Test

APSS CPSS APSS

Experimental Power System Model
 Identifier Controller

Disturbance u-vector
(t) u_vector
 vector
+ ω_vector
Plant -
PLANT
GN
Controller 

GN Unit Delay
Identifier
Σ

Learning Algorithm
GN-identifier
Output i(t+T)
 Performance of GN identifier

Results of GN identification for a 3-Phase to Ground fault at generator bus for 100 ms at P=0.7, Q=0.3 (lag).

Experimental Results of GN identification under 23 % step change in torque reference and trained on-line.
28
Performance of GNPSS and GNAPSS under three phase to ground fault for 100ms at the middle of one line in a double
circuit system at P=0.7pu and Q=0.3 pu (lag) .

Performance of GNPSS and GN based adaptive PSS when one line is removed at 0.5 sec. and re-energized at 5.5 sec and
then again same line is removed at 10.5 sec. and re-energized at 15.5 sec. at P=0.8 pu and Q=0.4 pu (leading). 29
Fuzzy Logic
General Concept
Fuzzy Logic Membership Functions
 Examples of Membership Functions distributions
.
NB NM NS ZO PS PM PB 1
1
0.5
Initial
0
0 -1.5 -1 -0.5 0 0.5 1 1.5
-1 0 1 (a) =2.5
(a) Initial 1
1 Contracted Initial
0.5

0
-1.5 -1 -0.5 0 0.5 1 1.5
0
-1 0 1 (b) =1.0
ted
(b)Contrac Expanded 1
1
0.5
Fig. 2. Linear scaling Fig. 3. Nonlinear scaling
0
-1.5 -1 -0.5 0 0.5 1 1.5
0
-1 0 1 (c) =0.6
(c) Expanded
18-May- 32/1
8 5
reference System

Inference mec anxsm

knowledge base
 Fuzzy Rules Table

We

NB NM Z PM PB

NB NB NB NB Z Z

NM NM NB NS Z Z

Z NS NS Z PS PS

PM Z PS PM PM PM

We
∆ PB Z PM PB PB PB
Conventional and Fuzzy PID Algorithm
Fuzzy Logic Self-tuning PI Algorithm
 Hybrid Micro-Grid Configuration

Wind Generator
Diesel Generator

PWind
Clutch Induction Wind Turbine
PDiesel
Diesel Synchronous Generator Dynamics
Engine Generator

Dump Load

Hydro-Pumped Storage Facility PDumpLoad

Dump Load Resistor Bank
Controller

Tunnel and Head PHydro

Pump-Turbine Synchronous
Penstock
Dynamics Generator
Water
Flow
Consumer Load
PLoad
Dump Load Frequency Control

Fuzzy IF-THEN Inference engine

Δf
ΔP
dΔf
dt

23/42
DL frequency control
Previous fuzzy frequency
control

[4]

Proposed fuzzy
controller

1 Large membership functions reduce the regulation time

2 Small membership functions reduce oscillations around settling point

24/42
 HPS Turbine Mode Controller

Simulink Fuzzy PID Controller Model

Linear Fuzzy PID Controller Non-Linear Fuzzy PID Controller

 HPS Pumping Mode Controller

Simulink Fuzzy Logic Controller Schematic

Fuzzy Logic Controller

Control Surface
 Transmission Lines

Loads
Generator

SVC

Power system diagram including SVC

Device
 Transmission Lines

Loads
Generator
Vm

SVC -
Vref
Δω / ΔPe
+
u(t)

Controller

Classical control approach for

a FACTS device
 Proposed Solution
Transmission Lines

Loads
Generator
Vm
-
SVC Vref
+
u(t)
ΔPSVC
Adaptive Controller

System
Identification

Proposed adaptive control system structure for SVC device

In a SMIB system
 Single Machine Infinite Bus System Simulation
Results
μ (e)

Consequent Parameters

Example of membership function before and after adaptation

………. Before adaptation After adaptation
Multi-machine System Simulation Results
3 6 9 7 1

G3 G1

2
Load 2

G2 8

Load 3

5 4

G5
G4
Load 1

Schematic model of a multi-machine power system with an SVC

device installed at the middle of the tie-line
Sensorless Control of a Switched Reluctance
Machine
Fuzzy Logic Controller for SRM
 Speed Tracking
SRM x 10
-6
b

a
350 0
10
300 0

Speed error (rad/s)

250
Speed (rad/s)

-2
200
-4
150
100 -6

50 Wmes -8
Wref
0 -10
 (s)

0 5
10 15
T

m
e
 Estimated and Real Load Torque SRM
a b
12 0.05

10

Torque error (N.m)

8
Torque (N.m)

6
0
4

0 Tl
Tl-obs
-2 -0.05
0 5 10 15 0 5 10 15
Permanent Magnet Synchronous Generator WECS
 Field Oriented Control of Stator Side
Converter of PMSG
• D-Q components of the stator reference voltages, that
ultimately control the rectifier firing angle, are generated
by two PI controllers with d-q components of the stator
currents as inputs.
• Conventional PI controllers are replaced by trained ANFIS
with d-q axes stator currents error and integral of error
as inputs.
• Applied to a 1.5 MW wind turbine system with PMSG
 Six sector phase plane
Δώ

Sector F

Sector A
R
Sector E θ

Δω
O

Sector B
Sector D

Sector C
Min Axis
Fig. 6 Fuzzy sets for input variable
Two Area Power System for LFC
Frequency variation of area- 1 in a Two Area Thermal
System without Reheat unit when disturbance in both areas
0

-0.005

-0.01 Both PI
PI in Area #1 and PFC in Area #2

-0.015
df1

-0.02

-0.025

-0.03

-0.035
0 10 20 30 40 50
Time (sec)
Frequency variation of area- 2 in a Two Area
Thermal
System without Reheat unit when disturbance in area
-1
- 0
-0.002

-0.004
PI PID
-0.006 Fuzzy
Polar Fuzzy
df2

-0.008

-0.01

-0.012

-0.014
-0.016
0 10 20 30 40 50
Time (sec)
 Short Term Load Forecast
• Statistical methods
• AI based methods employing both neural networks and fuzzy logic
- Neural networks need to be trained
- Using heuristic optimization techniques, e.g. GA, that employ
random search and fuzzy rules to guide search, performance can be
improved.
- A generalized neural network (GNN) with four wavelet components
of the historic load data as input and fuzzy logic guided random
search GA as a learning tool for the GNN is used for short term load
forecast.
- RMS error with:
back propagation training – 0.0610 GAF training – 0.0486
 GNN
Input

Dc1cd
hfodified GA outPut
 Short Term Load Forecast with FL and GNN

1 1
actual actual
predicted predicted
0.9 0.9

0.8
0.8

0.7
0.7

0.6
0.6

Load (kW)
Load (kW)

0.5
0.5
0.4

0.4
0.3

0.3
0.2

0.2 0.1

0.1
0
0 20 40 60 80 100 120
0 Time(Hours)
0
20 40 60 80 100 120
Time(Hours)
 Self-Tuning Load Forecast using GNN-W-GAF
1
actual
0.9 predicted

0.8

0.7

0.6
Load (kW)

0.5

0.4

0.3

0.2

0.1

0
0 20 40 60 80 100 120
Time(Hours)
Supervisory Control of a Cogeneration Plant
Generator Fuzzy Set-Point Control
S.0 Negatlve zeo

-4 -3 -2 -1 0 1 2 3 4

N SP
zse zse
sN¥P
Fuzzy Logic Self-Tuning PI Controller
 Fuzzy Adaptive Control PSS

AVR
&
Exciter
TL Grid
u(t) Mamdani FLC +
^
z-1 y(t)
y(k+1) _

RLS identifier
APSS

RLS identifier and a self-learning Mamdani fuzzy logic controller.

Results

0.1 p.u. step increase in torque and 3 phase to ground fault at the middle of one
return to initial condition transmission line and successful re-closure
(power 0.30 p.u., 0.9 pf lead) -adaptive Mamdani fuzzy logic PSS (AMFLPSS
----fixed centers FLPSS
(power 0.9 p.u., 0.9 pf lag)
Adaptive Neuro-Fuzzy Inference System
 General Schematic of ANFIS

Basic structure of a typical ANFIS with two inputs and two-rule fuzzy system
Adaptive Neuro Fuzzy Inference System
• An ANFIS is an integration of neural networks and fuzzy
inference systems to determine the parameters of the
fuzzy system.
• Automatically realize the fuzzy system by using the neural
network methods.
• Fuzzy Sugeno models are involved in the framework of
adaptive system to facilitate learning and adaptation.
• Permit combination of numerical and linguistic data.
• Requires structural and parameter learning algorithms.
The Proposed Adaptive Neuro-Identifier
• A Multilayer Perceptron (MLP) network is constructed to represent
the plant

Architecture of adaptive neuro-identifier

Adaptive Simplified Neuro-Fuzzy
Controller

Proposed control system structure

 NFC architecture

Nonlinear Function (NLF):

f  x N   signxN  x N 

xN

18-May-8
73/1
5
Control system structure

18-May-8
74/15
Online Adaptive Neuro Fuzzy Controller for Nonlinear
Functions in the Input Layer for Damping Power
System Oscillations
A fuzzy PSS is usually made adaptive by adjustment of input
membership functions (premise) and consequent parameters (CPs).

Number of controller parameters depend on the shape and

number of membership functions.
Scaling factors have received little attention in the adaptive fuzzy
PSS design
 System Configuration



18-May-8
77/15
 Simulation results Multi-machine power system.
0.05
0.05
No PSS
0.04 0.04 CPSS
No PSS
CPSS ANFPSS
0.03 0.03
ANFPSS

ω1-ω2 (rad/s)
0.02
0.02
)s 0.01
0.01 /d
1-2 (rad/s)

a 0
r
0 (
2 -0.01

-0.01
-0.02

-0.02 -0.03

-0.03 -0.04

-0.05
-0.04 0 2 4 6 8 10 12 14 16 18 20
Time (s)
-0.05
0
2 4 6 8 10 12 14 16 18 20
Time (s)
ω2-ω3 (rad/s)

0.25
0.25 No PSS
0.2 CPSS
No PSS
0.2 CPSS ANFPSS
ANFPSS 0.15
0.15
0.1
0.1

0.05

2-3 (rad/s)
0.05
2-3 (rad/s)

0 0

-0.05
-0.05
-0.1
-0.1
-0.15 Fig.14. 0.10 pu step inc-dec in torque of G3, Fig.15. 0.10 pu step inc-dec in torque of G3,
-0.15
-0.2
PSS on G3. PSS on G1, G2 and G3.
-0.2
-0.25
0 2 4 6 8 10 12 14 16 18 20
Time (s) -0.25
0 2 4 6 8 10 12 14 16 18 20
18-May-8 Time (s) 78/15
 1.5 MW VSWECS
 e d


e q

_
i s  0 + v
d
+_ i s
PI 
dq 
SVM
DC

PWM
n * 
v sd 

d
+ gen.
r
+_
i + 
PI PI 
+_ 
n r
sq
i sq v sq

i sd d i
i
sa

q sb

isc

i sq
abc

PMSG
n mech

1 P
s
 rad / sec 
rot / min
Fig.2 Field oriented control scheme with speed sensor at generator
•Applied to the 1.5 MW wind turbine system.
• The wind speed starts at 11m/s, is changed to 9 m/s after 12 s
1.2
Generator Speed with PI Controller 1.2
Active Power with PI Controller
Generator Speed with ANFIS Controller Active Power with ANFIS Controller
1
1
Generator Speed (p.u.)

0.8 0.8

Active Power (p.u.)

0.6
0.6

0.4
0.4

0.2

0.2 Time (Sec)

0
0 5 10 15 20
0
0 5 10 15 20
Time (Sec)
 Experimental Results of Applying the
ASNFC in a Real-Time System
200 km Transmission Lines

Generator speed deviation in response to a 15% step Generator speed deviation in response to a three-phase to
increase in the torque reference (P=0.80 p.u. and 0.75 p.f. ground short circuit test at the middle of a 200 km transmission
lag) line with an unsuccessful re-closure (P=0.97 p.u. and 0.93 p.f.
lag)
 Concluding Remarks
• A wide spectrum of AI applications in power systems, from
load forecast to maintenance, is being explored.
• A general survey of the type of AI applications that have
been and are being explored for application in power system
has been attempted.
• This is not an exhaustive survey and some other applications
are also being pursued.
• Actual application of AI techniques, particularly for real-time
applications, is lagging. One application that seems to have
been adopted by the utilities is neural network based load
forecast algorithms.
Thank you
Questions?

Om Malik
[email protected]