Industrial electronics and

Instrumentation
1 – Introductory concepts
• Introduction
– Course educational objectives and motivations
– Key problems in power systems and power electronics

• AC power transfer criteria and issues

– Basic definitions
– Maximum active power transfer
– Power factor correction
– AC power transfers between active sources
– Non-sinusoidal steady-state regime
1
Introduction

2
 Course objectives
• GOAL: provide students with key concepts in power electronics
& power systems, using measurement science as a means to
create a connection between theory and practice.

Instrumentatiion and
measurement

Mechatronics
Electrical power
systems Power electronics

Industrial electronics

3
Role of the course within the
mechatronics

This course

Source: Wikipedia
4
 Applications

Hybrid & E-mobility Smart grids Lighting systems

Industrial automation Heating Ventilating and

Robotics Air Conditioning (HVAC)
and monitoring
5
 Syllabus of the course
Theoretical lectures (~40 hours)
• Introductory concepts
• Power systems basics
• Power electronic devices
• Power converters (DC/DC, AC/DC, DC/AC)
• Data acquisition systems for industrial applications
• Industrial metrology basics
Practical lectures (~20 hours)
• Labview programming
• Safety issues in electrical systems
• Lab experiments on three-phase systems, power electronic
devices and power converters
6
 References for the course
Theory
• P. Kundur, Power System Stability and Control, Mcgraw Hill.
• N. Mohan, T.M. Undeland, W. P. Robbins, Power Electronics:
Converters, Applications and Design, John Wiley & Sons Inc.
• French College of Metrology, Industrial Metrology, ISTE
• Slides of the course
Lab sessions
• NI tutorials: http://www.ni.com/academic/students/learn/

• Tutorials for experiments based on NI Elvis II and Opal-RT

– http://sine.ni.com/nips/cds/view/p/lang/en/nid/213668
– http://sine.ni.com/nips/cds/view/p/lang/it/nid/213096

7
 Power systems & electronic issues
• Mission: Transfer, process and control the flow of
electrical power both within and between systems in a
way which is optimally suited for user loads
Primary
Goals

Avoid or mitigate major faults as Maximize efficiency

well as safety hazards for users – Excessive (unacceptable) cost of
and staff wasted energy
– Reducing equipment size,
– Reliability/protection issues
weight and cost
– Power quality issues (e.g. harmonic
– Heat dissipation issues
distortion)
– Energy conservation policies for
green-house gas emission and
sustainability
8
 The EU scenario

+20% efficiency
+20% renewable sources
- 20% fossil fuel consumption

9
 Goals of EED 2012/27/EU

Research efforts and major changes in allowed devices, e.g.

– LED-based lighting systems
– Fluorescent lamps at 20 kHz
– Variable-speed motor drives (for pumps and HVAC systems)
10
 Types of power systems
Power systems (and its components) can be roughly classified as follows:

Direct Current (DC) Alternating Current (AC)

Constant (regulated) Adjustable Single-phase Three-phase

magnitude magnitude

Constant frequency, Adjustable frequency

adjustable magnitude and magnitude
• Power converters: systems based on power electronics devices and
controlled by embedded processing platforms to perform one of the
following transformations for adapting generators to loads: DC/DC, AC/DC,
DC/AC.
11
 Overview of the electricity grid
• Electrical power is transferred at
High Voltage to reduce the ohmic
losses due to currents
• As the power is distributed, the
maximum rated power is decreased
and the voltage is progressively
reduced by multiple transformers to
make it more suitable for final users
• Most of electricity still generated at
the transmission level, but the
generation due to Distributed
Energy Resources (DERs) is
growing (e.g. wind and solar farms)
• In future, significant bidirectional
power flows between “prosumers”
are also envisioned at the
distribution level
Source: Wikipedia
12
The «smart» grid

13
 Impact of Pluggable Electrical
Vehicles (PEV) on the power grid
Main features of some fully electric vehicles (cars and light duty vehicles)
already present in the market.

14
 Impact of Pluggable Electrical
Vehicles (PEV) on the grid (cont’d)

Source: A. Perujo, B. Ciuffo, ‘’Potential Impact of Electric Vehicles on the Electric Supply System,’’ JRC-IES Technical Report

According to some projections, the increment of electrical energy consumption for

PEV recharge is expected to be small even in the case of significant penetration.
However, the peak demand of power (e.g. at nighttime) could become critical.
15
AC power transfer criteria
and issues

16
 AC power: basic definitions - 1
i(t)
e(t ) = E cos (ω0 t + ϕ e ) ⇒ E = Ee jϕe
~
Source e(t) Load Phasors
i (t ) = I cos (ω0 t + ϕ i ) ⇒ I = Ie jϕ i
~
ℑ{}
⋅ E
~
Instant power
ℜ{}
( ) ( )
p(t ) = e(t )i (t ) = E ⋅ I * + E * ⋅ I + E ⋅ I e j 2ωot + E * ⋅ I *e − j 2ωot =
1 ~ ~ ~ ~ 1 ~ ~
4 4
~ ~ ~
I
⋅
p(t)
= E ⋅ I ⋅ cos(ϕe − ϕ i ) + E ⋅ I ⋅ cos(2ωo t + ϕe + ϕ i ) =
1 1
2 2
= P + P cos(2ωo t + 2ϕ i ) − Q sin(2ωo t + 2ϕ i )
P

Mean value Oscillating

(constant) component (Displacement) power factor t
Active (real) power
1 ~ ~ 1
( )
P = ℜ E ⋅ I * = E ⋅ I cos Φ = E rms ⋅ Irms cos Φ [W]
2 2
1
Reactive (imaginary) power Q = ℑ E~ ⋅ ~
2
1
( )
I * = E ⋅ I sin Φ = E rms ⋅ Irms sin Φ [VAR]
2
17
 AC power: basic definitions - 2
Apparent power
1~ ~ 1
Complex power S = P + jQ = E ⋅ I * ⇒ | S |= E ⋅ I = E rms ⋅ Irms [VA]
2 2
Types of (linear) loads
~ General case ZL
E
ZL=R ϕe − ϕ i = 0 ⇒ cos Φ = 1 ~
I
No reactive power: S =P, Q=0 |S| Resistive + mainly
~ Q inductive load
π E
ZL=jωL ϕe − ϕi = − ⇒ cos Φ = 0 ~ P
2 I
No active power: P =0, Q=S (lagging current) P
Q Resistive + mainly
π ~ ~ |S| capacitive load
ZL=1/(jωC) ϕe − ϕi = ⇒ cos Φ = 0 I E
2
No active power: P =0, Q=-S (leading current)

18
 The max. active power transfer theorem
~ In principle, to transfer the maximum
Zs=Rs+jXs I
amount of active power from a generator to
~ ZL=RL+jXL
V a load, the impedance of the load has to be
~ the complex conjugate of the source
E
(matching condition), i.e. ZL=ZS*
Proof: the complex and active active power provided to the load are given by:
1 ~ ~∗ E 2
S = V ⋅I =
ZL
⇒ = ℜ{ } =
2 RL
[ ]
P S E
(RS + RL )2 + ( X S + X L )2
rms
2 2 Zs + ZL 2

The maximum of this quantity is achieved for ∇P = 0 which results from

dP
1) =0 XL = − XS 2 Just at most 50% of the
dX L E rms 1
Pmax = = Pa available active power of
dP 4RL 2 the generator can be
2) =0 RL = RS transferred to the load
dRL
This theorem is not very useful in power systems because:
• Loads can be nonlinear and are given (they can be hardly matched to the generator)
• In any case efficiency is too low (unacceptable when large amounts of power are
transferred and potentially dissipated)
19
 Power factor correction - 1
• In power systems, many loads have a prevailingly reactive nature
(especially inductive – due to electrical motor coils)
• In such conditions, the active power actually absorbed by the load can
be very small.
• To solve this problem, additional complementary shunt components
have to be used to maximize the power factor
Inductive load
Without C With C NOTE: Not only Φ 2 < Φ1
~ ~ ~ ~
I ~ IC V V but also the current
IL RL
~
V Φ1 ~ Φ2 magnitude after adding C
1/(jωC) ~ ~ ~ I ~
jωL
I = IL IL IC is smaller  lower ohmic
losses over the line
Q1 = S1 sin Φ1 = P tan Φ1 P (tan Φ1 − tan Φ 2 )
Q1 − Q2 = Vrms ωC
C=
2
|S1| Q1
|S2| Q2 = S 2 sin Φ 2 = P tan Φ 2 ωVrms
2

Φ 1 Φ 2 Q2 For the same amount of wanted active power, the total

P apparent power is smaller.
20
 Power factor correction - 2
• Example: If P =3.3 kW, cos Φ1 = 0.8, cos Φ2 = 0.96, f0=50 Hz and Vrms= 220 V,
then it results that C ≈ 100 μF
Capacitive load (less common)
~ Without L With L
I ~ ~ P
IC RL IL
~ ~ ~ ~ Φ1 Φ 2 Q2
~
V jωL
I = IC IC IL
~ |S2|
1/(jωC) Φ1 I
Φ2 |S1| Q1
~ ~
V V

Q1 = S1 sin Φ1 = P tan Φ1 2
Vrms
2
Vrms
Q1 − Q2 = L=
Q2 = S2 sin Φ 2 = P tan Φ 2 ωL ωP (tan Φ1 − tan Φ 2 )
NOTE: Both in the case of inductive and capacitive load, a perfect power
factor correction (i.e. cos Φ= 1) is not possible because of the tolerances of
the available components and because of the parasitic losses.
21
 AC power transfer between “active”
sources with significant line impedance
Thevenin’s equivalent Goal: Define the criteria to control active
circuits and reactive power flows between nodes
Node S Node R
~ ~ S and R.
ES = ES e jϕS R+jX ER = ER e jϕR
Simplifying assumptions: R≈0 and ϕR=0
~ ~ ~
I ~ ES − ER ES cos δ − ER + jES sin δ
I = =
jX jX

1 ~ ~*
SR = PR + jQR = ER I
2

ERrms ESrms ERrms ESrms cos δ − ER2rms

SS = PS + jQS SR = PR + jQR PR = sin δ QR =
X X
ℑ{}
⋅ ~ jXI
Similarly
ES
~ ϕS = ϕ R + δ ERrms ESrms ES2rms − ERrms ESrms cos δ
ER PS = sin δ QS =
ϕR ℜ{}
⋅ X X
22
 Case studies
ERrms ESrms − ER2rms
QR =
X
• Case a: δ = 0 and ERrms ≠ ESrms  PR = PS = 0 and
ES2rms − ERrms ESrms
QS =
X
If ESrms >ERrms QS>0, QR>0  QS+QR>0: Reactive power flows from S to R
If ESrms <ERrms QS<0, QR<0  QS+QR<0: Reactive power flows from R to S

REMARK 1: Q transfer depends mainly on the voltage magnitude at the ends

of the line

• Case b: δ ≠ 0 and ERrms = ESrms= Erms  PS = PR =

2
Erms E 2
sin δ , QS = −QR = rms
(1 − cos δ )
X X
If δ >0 Active power flows from S to R
QS + QR = 0 No reactive power
If δ <0 Reactive power flows from R to S transfer

REMARK 2: P transfer depends mainly on the angle difference between

voltage phasors at the end of the line (current lead vs. current lag)
23
 Case studies (cont’d)
• Case c: δ ≠ 0 and ERrms ≠ ESrms  We have both active and reactive power
transfers from S to R (or vice versa) depending on both the magnitude and
phase differences between the voltage phasors at the ends of the line (see
power flow analysis in Chapter 2 for the general case).
NOTES
ES2 + ER2 − 2ERrms ESrms cos δ
1. In all cases above: QLoss = QS − QR = rms rms
= XI rms
2

X
Part of the reactive power is stored in the line. This is sometimes referred to as
«reactive power loss»

2. If the line resistance R is no longer negligible, we also have a ohmic power

loss PLoss = RI2rms. Since PLoss and QLoss can be rewritten respectively as
PR2 + QR2 PR2 + QR2
QLoss =X and PLoss =R
ER2rms ER2rms
to transfer a given amount of active power PR with minimal losses, then QR
should be decreased as much as possible.
REMARK 3: Node voltage magnitudes at the ends of the line should not
deviate significantly from nominal values to keep QR low.
24
 Non-sinusoidal steady-state
• In power systems most of loads are nonlinear, because they include switching
electronic devices which are supposed to increase power conversion efficiency.
• As a result of such nonlinearities, the current waveforms are periodic, but distorted
(i.e. no longer sinusoidal) and this behavior affects also V and I waveforms in other
branches of the circuit, even if the terminal load is linear.
• CONCLUSION: V and I waveforms include a significant number of harmonics

EXAMPLE
Assume an ideal diode D with zero ON resistance
and negligible threshold voltage (VJ≈0).
• If vs(t)>0, D is ON and current i(t) is due to the
parallel between RLA and RLB
v (t ) v s (t )
i (t ) = ⇒ v (t ) =
RLA + RLB R + RLB
1 + Rs LA
RLARLB RLARLB
•
If vs(t)<0, D is OFF and current i(t) flows only
v (t ) R v (t )
i(t) and v(t) are not sinusoidal through RLB
and have non-zero mean values i (t ) = ⇒ v (t ) = LB s
RLB RLB + Rs
25
 Distortion analysis
Since both the current and the voltage waveforms applied
i (t)
to a generic load are periodic with period T0=2π/ω0, from
Nonlinear the Fourier series it follows that
e(t)
( )
+∞
load
e(t ) = E0 + ∑ E k cos kω0t + ϕek
k =1

( )
+∞
i (t ) = I0 + ∑ I k cos kω0t + ϕ i k
k =1
NOTE: the integrals (computed over an integer
numbers of cycles) of all sinusoidal terms arising from +∞
the squares of the series above are equal to zero ∑E
k =0
2
k rms
+∞ +∞
E k2
∑ ∑
T0
e (t )dt =
1 k ≠1
∫
E rms = 2
E 02 + = E k2rms ⇒ THDe = NOTE: alternative
T0 0
k =1
2 k =0 E1 definitions exist, e.g. the
rms

+∞
DC component
∑I
k =0
2
k rms
sometimes can be
excluded
+∞ +∞
I k2
∑ ∑I
T0
i (t )dt = I 0 +
1 k ≠1
I rms = ∫ = ⇒ THDi =
2 2 2
k rms
T0 0
k =1
2 k =0 I1
rms

• Amplitude and phase of harmonics depends on the severity of nonlinearity. In

50 Hz/60 Hz systems, often the first 50 harmonics are considered. The odd
harmonics are often larger than the even ones.
26
 What about power?
( ) ( )
+∞
p(t ) = e(t )i (t ) = E0I0 + ∑ E k I k cos kω0t + ϕek cos kω0t + ϕ i k + other terms...
k =1

( ) ( )
+∞ +∞
1 Ek Ik
p(t )dt = E0I0 + ∑ cos ϕek − ϕ i k = ∑ E k rms I k rms cos ϕek − ϕ i k
T0
P=
T0 ∫0
k =1 2 k =0

• The average power coincides with the active power and this is given by
the sum of the DC power and the active power of all harmonics
• Dually, the reactive power is defined as

( )
+∞
Q = ∑ E k rms I k rms sin ϕek − ϕ i k
k =0

• However, the reactive power has not a proper physical meaning because in
non-sinusoidal conditions the phasorial analysis does not strictly apply and the
apparent power is
+∞ +∞
S = E rmsI rms = ∑E
k =0
2
k rms ⋅ ∑I
k =0
2
k rms > P 2 + Q2

27
 What about power? (cont’d)
S = E rmsI rms = P 2 + Q 2 + D 2

The distortion power keeps into

account the effect of distortion. Without
further definite information, the sign of
distortion power is the same as of the
total active power.
NOTE: in nonsinusoidal regime
P
PF = ≠ cos Φ1
E rmsI rms
So the true power factor is different from the
“displacement” power factor (of the fundamental)!
• In case of polyphase systems connected through N conductors (e.g. three-phase
- see next chapter), the total active and reactive power is simply given by the
sum of the active and reactive power transferred over N-1 of them, i.e.

( ) ( )
N −1 +∞ N −1 +∞
P = ∑∑ E k rms I k rms cos ϕek − ϕ i k and Q = ∑∑ E k rms I k rms sin ϕek − ϕ i k
r =1 k =0 r =1 k =0

28