See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/3873272

Performance calculation for tunnel boring machine motors

Conference Paper in Conference Record - IAS Annual Meeting (IEEE Industry Applications Society) · November 2000
DOI: 10.1109/IAS.2000.883200 · Source: IEEE Xplore

CITATIONS READS

6 1,862

1 author:

P.E. Sutherland
Retired from GE and ABB
67 PUBLICATIONS 601 CITATIONS

SEE PROFILE

All content following this page was uploaded by P.E. Sutherland on 03 January 2018.

The user has requested enhancement of the downloaded file.

 Performance Calculation For Tunnel Boring
Machine Motors

Peter E. Sutherland (SM)

Power Systems Energy Consulting
GE Power Systems
Schenectady, NY 12345 USA

Abstract ● The performance of a Tunnel Boring Machine motor operating voltage of 628V, the voltage experienced by
(TBM) is dependent upon the mine power distribution system the motor was calculated to be always greater than rated.
providing adequate valtage. The voltage at the motors may be
The calculated voltage drop over the length of the cable
calculated using a standard computer load flow program. These
calculated voltages are compared to measured voltages in the was calculated to be 4.5 ‘%0when loaded. The calculated
tumel. The classical induction motor equivalent circuit is used to voltage drop in the TBM three-winding transformers was
calculate the torque output for a given voltage input. Current calculated to be 270 when loaded. This transformer was
and speed versus horsepower curves are developed from the modeled using the four-bus equivalent circuit [1]. The voltage
model and compared with motor test data. The motor modeling at the motors was boosted using taps on the TBM
results are used to adjust the simplified motor model in the load transformers, which raised the secondary voltage by 5%. The
flow program. The results of the calculation are a torque versus
total voltage drop from the surface to the motors was thus
distance curve that summarizes the performance of the TBM.
calculated to be about 1.5%.
Using thk method, the effects of changes in source voltage,
transformer taps, cable size and other factors on machine
III. INDUCTIONMOTOR MODELING
performance can be evaluated.
The motor performance was analyzed using the standard
Index Terms ● Induction Motor Performance, Tunnel
Boring Machhe, Induction Motor Modeling, Load Flow, single-cage rotor induction motor positive sequence
Voltage Drop equivalent circuit (Fig. 2) [2]. Torque-speed curves from the
model were calculated, and compared to test results frclm a
I. INTRODUCTION similar motor. The single-cage rotor model can be used to
model the running performance of an induction motor. If it is
A Tunnel Boring Machine (TBM) was used to drill a
desired to model the starting performance, then a two-cage
tunnel through rock for a distance of approximately 10 miles.
rotor model should be used. Fig. 3 shows a torque speed
The TBM was supplied through a 13.8 kV distribution cable
curve based upon the locked rotor torque and the breakdown
through the tunnel. Distribution transformers were attached
torque of the motor for which test results were obtained,
to the cable at intervals throughout the length of the tunnel to
which is compared with a curve calculated from the single-
supply lights, pumps, ventilation, etc. The electrical system
cage rotor model. The speed at which the breakdown torque
of the TBM consists of two 1900 kVA 13.8 kV I 660 Volt
appears was not known. The discrepancy in the locked rotor
transformers, each supplying four 600 Volt 422 horsepower
torque is because this is a single-cage rotor model. The
motors. The transformers also supply a number of lower
curves are similar in the operating region between the
voltage pumps.
breakdown torque and synchronous speed.
The purpose of this analysis is to estimate the performance
The parameters used in the motor model are listed in Table
of the TBM motors during the boring process to determine
1. These were determined from the motor data by matching
whether adequate voltage was supplied to the TBM. The
the model parameters with the motor data as shown in Table
analysis is to be based upon (a) motor nameplate data and (b)
2. The data used as a starting point is listed in the second
test results provided by the manufacturer for a similar motor.
column. The model is then optimized to match the motor
The voltage at the motor is to be calculated and measurement
characteristics as shown in the third column. The first stl:p is
results provided analyzed. From these results, an estimated
to calculate the speed-torque curve, and selected motor
performance curve for motor torque output versus voltage and
characteristics, from the motor model. Then, the m[odel
distance in the tunnel is found.
parameters are adjusted one by one until the model falls in
II. VOLTAGEDROP STUDY place for the portion of the speed-torque curve that is to be
matched.
The voltage drop study was performed using a computer
load flow program. This showed predicted voltages at the
TBM motors in the range of 630-660V, when loaded (Fig. 1).
This agrees with the measured motor voltages. With a rated

0-7803-6404-X/00/$10.00 (C) 2000

 Voltage Profile of Tunnel Boring

1MO
A

0.940
A

0.920
Ommmmlm
I%g

Calculated voltage profile of tunnel boring operatton. Jumps m secondary voltage are at Iocatrons where tap changes occurred.
Measured voltages are taken from readings at TBM.

rp + jxp

t
la,(Primary)
I (Secondary)
‘al

Val

RC jxm

n .
Fig. 2. Induction motor positive sequence equivalent circuit.

o 100% JW0720W O1O’3O1263144O le201eco

SIE@ (mm

FIE. 3. Tmque vs. speed curves for TBM motor (a) as calculated from motor data sheet, and (b) calculated from starting and
breakdown torques for similar motor tested by manufacturer.

0-7803-6404-X/00/$10.00 (C) 2000

 IV. TORQUE vs. VOLTAGE CALCULATION TABLE lB
WIPEDANCE PARAMETERS FOR MOTOR MODEL
From the torque-speed curves, motor torque output at
differing operating voltages was calculated (Figs. 6-8). This
I lm~edances for Model I
showed a variation of about 1.5 ‘%o in torque output over the !X”d I 0.251 IDU I
voltage range of 60CI to 700V. It is commonly stated that Ixm I 8.081DU I
torque varies with the square of the applied voltage. lr~ I 0.0301DU I
However, this refers to the starting torque and the breakdown Irs I 0.02251rxJ I
torque [3], not necessarily the torque when the motor is Irc I 36,4hxI I
driving a particular load with its own speed-torque
Xp 0.126]Pu
characteristic. The torque-voltage characteristic under load is
Xs 0.1261pu d
the path of the intersections of the motor and load speed-
torque curves, as shown in Figs. 6 and 7. These results were
separated out and plotted as a torque vs. voltage curve in Fig.
8. This was combined with the voltage drop study results TABLE 2
DERIVATION OF IMPEDANCE PARAMETERS
(Fig. 1) to calculate torque vs. distance (Fig. 9). This showed
FOR MOTOR MODEL
a variation of less than 1YO in torque output over the length of
the tunnel. Para-meter Starting Data Optimize to match -
Motor Parameter
TABLE 1A X!,d
LRNFLA Breakdown torque (TE~
MOTOR DATA
Ixm I 1 lPower Factor (pf)
I Motor Data I
V base
FLA
628 v
335 A
II II 11
——
Pf
I
I
Sbase 364 INA
I
rp Motor
Zbase
Sync Speed
1.082 Ohms
1800 rpm
resistance by I
Speed 1780 rpm
LRA 1207 A
HP 422 HP
Tbase 1425 Lb-Ft. Ixs 10.5 * X“d I I
Trated 1261 Lb-Ft.

Current vs. Horsepower

160% , 1

+cdc-660
+c6Jc-600
+test-660
+test-600

o% 25% 50% 75% 100% 125%

‘A Rated Horsepower
–.i

Fig. 4. Current vs. horsepower for calculated and measured “performance data” at 600 and 660V.

0-7803-6404-X/00/$10.00 (C) 2000

 Spsscl vs. Horsepower

1820-

18fXl I

1780- +calc-660

+calc-600
s
L 1780- -A-test-660
K
+test-6LXI
1740- y ;D

1720-

17(XI r
o% 25% WA 75% 1CQ% 125%

%Ratsd Horsepower

Fig. 5. Speed vs. horsepower for calculated and measured “performance data” at 600 and 660V

ml.
I ---- m

L
. . . . . . 626
-..
—.. —
f333

0 16r)333E40 za3xolm12eo 144a Imlem

W(w

Fig. 6. Variation in torque with voltage. A typical load torque curve is also shown

Torque-Speed vs. Volts (detail)

200%
-%
-.. ‘%
~...~
150% ~ ---- .:\\
-.. :-””.mk –-–-600
~ +-----
> :.x. ...---628
~ ,Ooye
~ . .._.. _.. —.. —-”.:~5~—”-– --—-- -––660
*2. x.
$ -G> - E–--–Load
50% ‘ <=.

‘h%
o% .
1710 1728 1746 1764 1782 1800

S@ssd (RPM)

Fig. 7. Detail of torque curves of Fig. 6.

0-7803-6404-X/00/$10.00 (C) 2000

 Torque vs. Voltsge

101.0%

g ‘m”5%

% 100.0%
$

+ 88.5%

800 620 640 660 680 700

Voltage

Fig. 8. Variation of torque with voltage when load torque is proportional to speed squared. This curve was calculated at each voltage by finding the point
where the motor curve crossed the load curve, as shown in Fig. 7.

Variation of Torque with Distance

‘“’’”’”~

99.5%

t
99.0% ~
o 2000 4QO0 60C0 6CC0 10000

Ring

Fig. 9. Variation of torque with distance, combining Figs. 1 and 8. This curve was calculated at each voltage by finding the point where the motor cume
crossed the load curve, as shown in Fig. 7.

Current vs. Voltsge

—
with Constsnt HP Output

1i 0.0% Model
~ f 05.0”/.
:

0
~
!W%
$ 100,0% ) i

95,00/o1

90.0%

85.0%~
t
%3’.%,.,,

6 ‘“ ““”\,%

f
E --W---Const Z
-; Const kW+jl
--M--’ Const kVA

‘con’”

600 610 620 630 640 650 660 670 680 690 700
Voltage

Fig. 10. Comparison of induction motor model and common power flow motor models

The motor model was compared with the available models Of these models, the constant kVA and the constant
in the load flow program. These are: kW + jI are the most accurate models for running induction
1. Constant kVA motors. Fig. 10 shows a comparison of the calculated motor
2. Constant kW + j~ current with each of these models and with the single-cage
3. Constant Z rotor model of the induction motor. The constant kVA model
4. Constant I is seen to be the most accurate for this motor, bul. the

0-7803-6404-X/00/$10.00 (C) 2000

 difference between it and the constant kW + jI model is BIOGRAPHY
very small.
Peter E. Sutherland (M ’83, SM ’97)
These calculations were made assuming a fixed load received the A.S. Degree in Electrical
torque-speed characteristic curve. However, in actual Engineering Technology (’79) and the B.S.
practice, the load characteristics are constantly changing as degree in Electrical Engineering (’83) fTom
the boring proceeds through rock with varying characteristics. the University of Maine at Orono. In 1986
These results are those of theoretical calculations based upon he received the M.SC.E degree in Electrical
idealized characteristics. However, they do compare fairly Engineering from the University of New Brunswick (Canada).
well with what data was available from actual operations.
He has worked as a Test Engineer and a Design Engineer
V. CONCLUSIONS for Accutest Corp. of Chelmsford, Mass., a manufacturer of
Using the previous results it is possible to plot the variation automatic test equipment for the semiconductor industry. For
of torque with distance in the tunnel, as shown in Fig. 9. a short time he worked as a Planning Engineer for an Electric
These results show that the motor supplies close to rated Utility Company. In 1987 he joined General Electric
torque even under calculated worst case voltage drop Company, and has been employed as an Engineer in the GE
conditions. Industrial Power Systems Engineering Operation in
Schenectady, N.Y., as an Instructor in the GE Training and
REFERENCES Development Center, and as a Power Systems Engineer in
[1] Oommen, M.P. Kohler, J.L., “Effect of three-winding transformer Albany, N.Y. His current position is Senior Engineer in the
models on the analysis and protection of Mine Power Systems,” IEEE GE Power Systems Energy Consulting Department, where he
Trans. Ind. Appl., Vol. 35, No. 3, May/June 1999, pp. 670-674. performs a variety of power system studies for industrial
[2] Clarke, E., Circuit araitysis of A-C Power Systems, New York: Wiley,
1950, Vol. II, pp.210-222.
clients.
[3] Bennett, A.H., “The impact that voltage and frequency variations have
Mr. Sutherland is currently enrolled as a part-time student
on AC induction motor performance and life in accordance with
NEMA MG-1 standards,” Conference record of 1999 IEEE Pulp and in the Ph.D. program in Electric Power Engineering at
Paper Industry Technical Conference, pp. 16-26. Rensselaer Polytechnic Institute, Troy, N.Y. Mr. Sutherland
is a member of Eta Kappa Nu and Tau Beta Pi. He is a
Registered Professional Engineer in Maine and in New York.

0-7803-6404-X/00/$10.00 (C) 2000

View publication stats