Upload
phamnhu
View
232
Download
6
Embed Size (px)
Citation preview
12/18/2014
1
IEEE 519‐2014
Mark Halpin
November 2014
What Has Stayed the Same?
• Most importantly, the overall philosophy
– Users are responsible for limiting harmonic currents
– System owner/operator are responsible for managing voltage quality
– All recommended limits apply only at the PCC
• Existing recommended limits are retained
– Some new ones added
12/18/2014
2
What Has Been Changed?
• Philosophy of changes Driven by 20 years of experience with 519‐1992 and increased cooperation with IEC
• Multiple changes related to– Measurement techniques– Time varying harmonic limits– Low voltage (<1 kV) harmonic limits– Interharmonic limits– Notching and TIF/IT limits
• Also “editorial” changes to– Reduce document size– Minimize miss‐use of PCC‐based limits– Better harmonize with other standards projects
Measurements
• Recommended to use IEC 61000‐4‐7 specifications
– 200 ms (12 cycle @ 60 Hz) window gives 5 Hz resolution
0
0.2
0.4
0.6
0.8
1
1.2
1.4
X-6
0
X-5
5
X-5
0
X-4
5
X-4
0
X-3
5
X-3
0
X-2
5
X-2
0
X-1
5
X-1
0
X-5 X
X+
5
X+
10
X+
15
X+
20
X+
25
X+
30
X+
35
X+
40
X+
45
X+
50
X+
55
X+
60
Interharmonics @ 5 Hz
Harmonics @ 60 Hz
12/18/2014
3
Indices
• From IEC 61000‐4‐30
– 3 s “very short” value
– 10 min “short” value
215
1i
2i,nvs,n F
151
F
2200
1i
2i),vs,n(sh,n F
2001
F
Assessment of Limit Compliance
0
2
4
6
8
10
12
14
16
18
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69
Time (h)
TD
D (
%)
What value should be compared against the limit?
12/18/2014
4
Weekly Statistical Indices
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 91
01
11
21
31
41
51
61
71
81
92
0
TDD (%)
Fre
quen
cy
.0%
20.0%
40.0%
60.0%
80.0%
100.0%
95th or 99th
percentile
Value to be compared against limit
Changes to the Limits
• New voltage limit provision for low voltage (<1 kV)– 5% individual harmonic, 8% total harmonic distortion
• Revised current limits for general transmission systems (> 161 kV)
Maximum Harmonic Current Distortion in Percent of IL
Individual Harmonic Order (Odd Harmonics)
Isc/IL <11 11≤h< 17 17≤h< 23 23≤h< 35 35≤h TDD
<25* 1.0 0.5 0.38 0.15 0.1 1.5
25<50 2.0 1.0 0.75 0.3 0.15 2.5
≥50 3.0 1.5 1.15 0.45 0.22 3.75
12/18/2014
5
Percentile‐Based Voltage Limits
• Daily 99th percentile very short time (3 s) values should be less than 1.5 times the values given in Table …
• Weekly 95th percentile short time (10 min) values should be less than the values given in Table …
Percentile‐Based Current Limits
• Daily 99th percentile very short time (3 s) harmonic currents should be less than 2.0 times the values given in Table …
• Weekly 99th percentile short time (10 min) harmonic currents should be less than 1.5 times the values given in Table …
• Weekly 95th percentile short time (10 min) harmonic currents should be less than the values given in Table …
12/18/2014
6
Interharmonic Limits (“Recommendations”)
• Voltage‐only 0‐120 Hz limits based on flicker
0
1
2
3
4
5
6
0 510
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
Frequency (Hz)
Volta
ge (%
of N
om
inal)
V≤1kV
1 kV<V≤69 kV
69 kV<V≤161 kV
V>161 kV
V≤1kV
1 kV<V≤69 kV
69 kV<V≤161 kV
V>161 kV
allvoltages
allvoltages
Editorial Changes
• Improve definitions of all relevant terms to account for greater understanding and improved instrumentation
• Removal of “flicker curve”
• Removal of “tutorial” material (shorten document)
• Strengthen introductory material dealing with PCC‐only applicability of recommended limits
12/18/2014
7
Experience So Far
• Granted, this is limited mostly to “experiments” over the last 6‐12 months– Users with relatively stable harmonic emissions are essentially unaffected
– Users with rapidly‐changing harmonic emissions may show reduced levels in measurements
• The 200 ms window acts as a smoothing filter
• Percentiles and multipliers appear to be relatively consistent with “short time harmonic” multipliers often used with 519‐1992
Passive Mitigation of Power System Harmonics
Mark Halpin
November 2014
12/18/2014
8
Outline
• Passive Filters
– Basic resonance concepts
– Single‐tune filters
– C‐type filters
• Performance comparisons
– Sensitivities to network conditions
– Overall effectiveness
• Conclusions
Series Resonance Concept
CL
eq
XXj
C
1LjZ
LC
1r
Major concept: The impedance can become a very low value
resonant frequency, r
inductive
capacitive
12/18/2014
9
Series Resonance In Practice
HarmonicVoltages
harmoniccurrents
Effects include:1. Heating in transformer2. Fuse blowing at capacitor bank
Typical resonances:‐‐500 kVA, 12.47 kV, 5%‐‐300‐1200 kvar capacitor‐‐r=173‐346 Hz (3
rd‐6th harmonic)
Parallel Resonance
CL
CL
eq
XX
XXj
Cj
1//LjZ
LC
1r
Major concept: The impedance can become a very high value
resonant frequency, r
capacitive
inductive
12/18/2014
10
Parallel Resonance in Practice
harmonicvoltages
HarmonicCurrents
Effects include:1. Excessive voltage distortion2. Capacitor bank fuse blowing
Typical resonances‐‐500 kVA, 480 V, 5%‐‐400 kVA load, 80% pf lagging‐‐pf correction to 95% lagging (120 kvar)‐‐r=547 Hz (9
th harmonic)
Resonance Summary
• Series resonance
– Widely exploited in harmonic filters
– Can lead to (harmonic) overcurrents
• Parallel resonance
– Frequently leads to (harmonic) overvoltages
– Sometimes used in blocking filters
12/18/2014
11
Single‐Tuned Filters
• “Single tune” means a single resonant point
Classical Single‐Tuned Filter
C‐type Filter
Applications
• Classic single‐tuned filters– Common in industrial applications
• Inside facility• At the PCC• May use multiple filters, each tuned to a different frequency
– Traditionally used by utilities (declining)
• C‐type filters– Not common in industrial applications– Becoming dominant in the utility environment– Often used in conjunction with classic single‐tuned designs
• Purpose is always the same—give harmonic currents a low‐impedance path “to ground”– Results in reduced voltage distortion
12/18/2014
12
Application Considerations
• Ratings– Capacitor
• RMS voltage
• Peak voltage
• RMS current
• kVA
– Reactor currents• Peak current
• RMS current
• Losses
Filter Application Procedure
• Use frequency scan and harmonic study to determine requirements– Number of filters (estimate)
– Tuned frequency for each
– Ratings (estimate)
• Start with lowest‐frequency filter and work upward (in frequency)– Each filter has parameters than can be at least partially optimized
– Consider credible system changes
– Assess impacts of filter parameter variations (±10%, maybe more)
• Evaluate total performance vs. requirements– Consider credible system changes
– Specify required ratings (tweak design as necessary)
12/18/2014
13
Comments on Frequency Scans
• These results indicate the potential for a problem• They are extremely useful for designing filters
– Identification of high/low impedance frequencies (resonant conditions)
– Assessment of filter impacts on frequency response• Alteration of undesirable impedance characteristics• Demonstration of intentional low impedance path(s)
• They are subject to the accuracy of the models used• Complete assessments require a harmonic study
– Results subject to model accuracy and assumptions– Limit compliance– Ratings of components
Demonstration Case
• Basic harmonic situation and sensitivities– Series and parallel resonance conditions
• Mitigation using filters– Single‐tuned “industrial”– C‐type “utility”
12/18/2014
14
Normal Condition Frequency Response—LV Filter Application
(Are impedances high or low at known harmonic frequencies?)
0
0.2
0.4
0.6
0.8
1
1.20 1 2 3 4 5 6 7 8 9101112131415161718192021222324252627282930313233
Imped
ance ()
Harmonic #
System Normal
Sensitivities—Substation SC Power(equivalent impedance at LV bus)
0
0.2
0.4
0.6
0.8
1
1.2
0 1 2 3 4 5 6 7 8 910
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Imped
ance ()
Harmonic #
130 MVA
150 MVA
170 MVA
Increasing severity and frequency with fault MVA
Decreasing severity andIncreasing frequency with fault MVA
Increasing severity (lower Z) and increasing frequency with Fault MVA
These sensitivities would be considered pretty small and insignificant
12/18/2014
15
Sensitivities—Capacitor Status(equivalent impedance at LV bus)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 1 2 3 4 5 6 7 8 9101112131415161718192021222324252627282930313233
Imped
ance ()
Harmonic #
All Caps
LV Only
MV Only
No Caps
Low(er) frequency resonance not much affected by MV cap
High(er) frequency resonance significantly affected by MV cap
inductive
capacitive
Low(er) frequency resonances not much affected by things thatimpact high(er) frequency response—opposite not true!!
Sensitivities‐‐Conclusions
• Large changes in system impedances, equivalents, etc., (fault MVA) are usually needed for significant effects
• Relatively small changes in capacitor bank status (or size) can have major impacts
• Filters must function under all of the potential scenarios
12/18/2014
16
Design Approach
• Convert existing 480 V cap bank to filter bank by adding series reactor– Capacitor voltage rating often will be exceeded in the end!
– X/R ratio of reactor can have significant impact
• Losses
• Performance
– Additional resistance can be added in series if needed (losses will increase!) for performance
m4.15X
H7.40L
006908.L21
300
LC21
f
L
tune
Note: Tuned frequency normallytaken ≈5% below targetAvoid overloadParameter variation
5th Harmonic Single‐Tune Design
0
0.5
1
1.5
2
2.5
3
3.5
0 1 2 3 4 5 6 7 8 910
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Imped
ance ()
Harmonic #
X/R=100
X/R=10
X/R=1
R=0.0770 Ohm
12/18/2014
17
Filter Quality (“Q”) Factor
• The “sharpness” of the frequency response of a filter is often indicated by the filter “Q”
• The filter Q indicates
– Damping (less sharp characteristic—more damped)• Lower Q, more damping
– Losses• Lower Q, more losses
• For the previous slide
– Q=500, 50, 5, 1
R
Lf2
R
XhQ tune)60(Ltune
A Closer Look at Q
0
0.5
1
1.5
2
2.5
3
3.5
0 1 2 3 4 5 6 7 8 9
10
Impedance ()
Harmonic #
Q=500
Q=50
Q=5
Q=1
All this discussion of Q doesn’t look like a big deal…
12/18/2014
18
Performance Evaluation(480 V Bus Impedance)
0
0.2
0.4
0.6
0.8
1
1.2
0 1 2 3 4 5 6 7 8 9101112131415161718192021222324252627282930313233
Imped
ance ()
Harmonic #
No Filter
5th Filter (Q=500)
5th Filter (Q=1)
5th Filter (Q=10)
Filter Q has an obvious impact on the entire response!
5th harmonic currents produce much less 5th
voltage after filter
Performance Evaluation(LV Filter Impact on MV System at Cap Bank)
0
5
10
15
20
25
0 1 2 3 4 5 6 7 8 9101112131415161718192021222324252627282930313233
Imped
ance ()
Harmonic #
No Filter
High Q (500)
Low Q (10)
Lower Q: Not as much filtering at 5th harmonicmuch less amplification at higher frequencies
5th harmonic currents produce much less 5th
voltage after filter
12/18/2014
19
Filtering on 12 kV Network
• Discussion so far based on filtering on customer‐side (LV)
– Presumably associated with limit compliance
• If all network users are in compliance (currents), excessive voltage distortion may still exist
– Strong resonances can create large (noncompliant)voltage effects from small (within compliance) currents
– Solution is to filter on MV (utility) side
– Filter designs must account for LV filter presence (or not)
Same Approach for Filter Design
367.10X
mH5.27L
235.10L21
300
LC21
f
L
tune
Q=100R=0.5184 Q=10R=5.1835
Note: Tuned frequency normallytaken ≈5% below targetAvoid overloadParameter variation
12/18/2014
20
12 kV Filter Performance
0
2
4
6
8
10
12
14
16
0 1 2 3 4 5 6 7 8 9101112131415161718192021222324252627282930313233
Imped
ance ()
Harmonic #
No Filter
High Q (100)
Low Q (10)
Filter eliminates 5th resonance, but creates new ones that could be as bad (or worse).Best solution probably to split 600 kvar into 2x300 kvar and make two filters—5th and 7th
The C‐type Filter
• Tuning (selection of parameters) is more difficult than for single tuned filters
• Starting from an existing cap bank Ctotal– Step 1 Choose L to tune filter frequency as for single‐tuned designs (based on Ctotal)
– Step 2 Divide existing capacitance into two parts
• C2 chosen so that L and C2 are series resonant (Z=0) at the power frequency
• C1 determined from “Ctotal‐C2” (C in series combines as parallel)
– Step 3 Pick R to provide desired high(er) frequency damping
12/18/2014
21
C‐type Filter Example
• Will a 12 kV C‐type perform better than the conventional single‐tuned design?
• Existing 600 kvar bankCtotal=10.235F– L=10.367 (27.5 mH) for ftune=300 Hz (from ST design)
– For 60 Hz “bypass” tuning, C2=255.85 F• C1=10.66 F
– Select R for desired damping
• Note Q defined differently
Lf2R
XhR
Qtune)60(Ltune
C‐type vs. ST Filter Performance
0
2
4
6
8
10
12
0 1 2 3 4 5 6 7 8 9
10
Imped
ance ()
Harmonic #
No Filter
ST Q=100
ST Q=10
CT Q=5
CT Q=10
12/18/2014
22
12 kV Filter Sensitivities(LV Cap/Filter Off‐line)
0
1
2
3
4
5
6
7
8
0 1 2 3 4 5 6 7 8 910
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Imped
ance ()
Harmonic #
ST Q=10 (No LV)
CT Q=15 (No LV)
ST Q=10
CT Q=15
CT Q=0.5 (no LV)
The real advantage of the C‐type is control of HF response
Comments on Comparisons
• Both filter types are effective at the tuned frequency• C‐type has very low power frequency losses
– Single‐tuned filter has resistive losses proportional to cap bank reactive current squared
• Low Q single tuned designs are helpful to reduce secondary resonances created by filter additions– Alternative is to add secondary filters
• Low Q C‐type designs provide good damping of secondary resonances by default– Much less likely to encounter “secondary” problems
• C‐type designs make poor utilization of existing cap banks– Consider using one bank for var compensation with a separate
filter installation
12/18/2014
23
Passive Filter Conclusions
• Two main types exist—both work– Single tuned
• Main advantages: Simplicity, up‐front cost
• Main disadvantages: losses, can create secondary problems
– C‐type• Main advantages: Low losses, HF response
• Main disadvantage: up‐front cost, poor utilization of existing cap banks
• Frequency scans are a great tool for filter design– A harmonic study is required to determine necessary ratings