Development and Running test of the ATO(Automatic Train Operation)
Equipment for Korean standard EMU
Seong-Ho HAN*, Su-Gil LEE, Tae-Ki Ahn
Rolling Stock R&D Division, Korea Railroad Research Institute
Abstract
The automatic/driverless operation, a very important technique for metro railways, is necessary for achieving higher
safety, greater reliability, and bigger transport capacity. To achieve these things, we have to build up the system design
and testing techniques for the railway system operation. These techniques are related to the onboard train control and
communication systems which include TCMS(Train Control and Monitoring System), ATO(Automatic train Operation),
ATC(Automatic train Control), and TWC(Train to wayside communication). These sub-systems should interface not
only with each other but also between the sub-systems and the signal system on the ground. In this paper, we tested the
train control system on the Seoul subway 7th line that has been developed on the basis of the standardized type of EMU
for Korean railway systems.
1. INTRODUCTION
The automatic/driverless control of the urban transit has been necessary to achieve high safety and reliability and to
minimize the errors of on-board control equipment recently.
This technique is closely related not only to the on-board control but also to the communication method of the ground
signal system. Therefore, the different systems are implemented according to several interfaces of both methods.
These techniques are related to the onboard train control and communication systems which include TCMS(Train
Control and Monitoring System), ATO(Automatic train Operation), ATC(Automatic train Control), and TWC(Train to
wayside communication).
Especially, the ATO(automatic train control) system is an onboard automatic control equipment for the limited speed
regulation and the precision station stop of train.
Because most of the train control systems were designed in some advanced countries, after the expected life span of an
EMU(25 years), there are big problems such as increase of maintenance cost, difficulty of supporting the parts, etc.
To solve these problems, we started the Korean standard project with the support of government subsidies in 1995.
This paper, one of research results from the project, shows the form of the optimal solution and how to minimize the
energy consumption of the train driving control that can be included in Automatic Train Operation (ATO) systems.
For the support proof of reliability of the development products, we set up standardized requirements and conducted
various efficiency tests, such as a component test, a combined test, a finished train test, and factory test run. In this
paper, for the automatic train control, we made use of a fixed block based on track circuit. In order to minimize the
energy consumption, we adopted the optimal control techniques based on coasting strategy using the speed band. We
applied this system to the Korean standard EMU and performed running test in Seoul subway 7th Line. With the results
of the test, we verified that the proposed algorithm could increase the energy consumption efficiency.
2. ON-BOARD TRAIN CONTROL
On-board train control of the urban transit carries out train movement control, accident prevention to ensure safety and
increases the coefficient of line utilization.
The equipments related with safe train operation are TCMS(train control and monitoring system), ATS(automatic train
stop), ATP(automatic train protection), ATC(automatic train control), and ATO(automatic train operation).
According to the train detection method and the train speed control method, on-board train control is separated into the
speed step control, the single step brake control and the moving block control as following Figure 1.
Target speed control
Distance-to-stop control
Train speed control system
Train position detection system
Fixed block
Onboard detection of Train position
(ground detection of train position)
Speed step control
speed
distance
speed
Single-step brake control
(Quasi-movable blocking control)
speed
Moving block control
distance
distance
Fig. 1 a classification of on-board train control
The speed step control used in most subway systems in Korea makes use of the fixed block of track circuit for detection
of train and the target speed control for speed control. The single step brake control makes use of the fixed block like the
speed step control and the distance-to-stop control for speed control of train. This technique, so called the quasi-moving
block system, is adopted in the Busan subway 2nd Line and Inchion subway 1st Line.
The moving block control, used in the light railway system, makes use of radio communication for train detection.
Also, the distance-to-stop control is used for speed control of train.
The speed profile of the train is generated by strategy with change point of powering, coasting, and braking mode
according to ATC limit speed and track conditions.
In powering mode, inverter and traction motor consumed electrical energy. On the other hand, energy is regenerated in
braking mode. In coasting mode, there is no energy consumption because of using inertia.
Therefore energy consumption occurs when the mode changes coating to powering.
In order to minimize the energy consumption, it is very important to use coasting mode and to decide the changing
points while considering track conditions and train operation conditions. This technique also has to consider jerk limit,
running time, running resistance, and traction/braking efforts.
Figure 2 represents a profile with optimal control for automatic train operation. In the figure, the train starts satisfying
with jerk limit of 0.8km/h/s2 and acceleration of 3.0km/h/s. After arriving target speed, the train is operated by the
constant speed control that is under the optimal strategy with speed band.
With close arrival station, the train is under precision stop control with the distance-to-stop control. To increase the
accuracy of train stop, on-board control system used precision stop maker on the ground. Also, on-board control system
use the track database in which includes various information such as distance between station, station code, civil limit
speed, gradient, curve etc.
Speed [km/h]
The fixed speed of TWC
70
The limit speed of ATC
Coasting control
Constant speed control
Precision Stop Control
The speed band
60
50
The start acceleration control
Distance[km]
Start station
TWC Loop
Mini bond
P
1
P
3
P
2
Arrival
station
P
4
3.5m
21m
108.5m
546 m
Fig. 2 A profile with optimal control for ATO.
3. RESULTS OF RUNNING TEST
The running test was conducted on partial sections (on 7th Line, (PORAMAE°ÍCHONGDAM) of the Seoul subway 7th
Line operated by SMRT. Table 1 shows the detail information of the Seoul 7th Line.
Table 1 Information of the Seoul subway 7 Line
Name of stations
SPEED[km/h]
PORAMAE ↔ SHINDAEBANG SAMGORI
SHINDAEBANG SAMGORI ↔
CHANGSUNGBAEGI
CHANGSUNGBAEGI ↔ SANGDO
SANGDO ↔ SOONGSIL UNIV.
SOONGSIL UNIV. ↔ NAMSONG
NAMSONG ↔ ISU
ISU ↔ NAEBANG
NAEBANG ↔ EXPRESS BUS TERMINAL
EXPRESS BUS TERMINAL ↔ BANPO
BANPO ↔ NONHYON
NONHYON ↔ HAKDONG
HAKDONG ↔ KANGNAMGU OFFICE
KANGNAMGU OFFICE ↔ CHONGDAM
Up(West→East)
Num.
Dist.
Of
[m]
Section
784
3
Down(East→West)
Dist.
[m]
Num. Of
Section
793
6
1,152
5
1,158
4
941
908
1,973
1,011
1,042
2,204
916
846
994
899
1,108
3
5
8
5
7
8
4
5
7
4
5
938
906
1,976
1,021
1,030
2,206
919
845
994
898
1,105
3
4
5
6
5
7
4
5
6
5
6
Figure 3, 4, 5 show the results of the running test that include speed profiles operated with the automatic train
operation control between NAMSONG and SOONGSIL UNIV., between SOONGSIL UNIV. and SANGDO, and
between PORAMAE and SHINDAEBANG SAMGORI.
Figure 6 shows a view of the running test on the Seoul subway 7th Line with automatic train operation mode.
Fig. 3 A speed profile of Seoul subway 7 line
(NAMSONG -> SOONGSIL UNIV.)
SOONGSIL UNIV. -> SANGDO
DISTANCE: 906[m]
2000.5.
80
SPEED(km/h)
70
60
50
40
RUNNING SPEED
TARGET SPEED
30
20
10
0
0
100
200
300
400
500
600
700
800
900
1000
RUNNING DISTANCE[m]
Fig. 4 A speed profile of Seoul subway 7 line
(SOONGSIL UNIV. -> SANGDO)
PORMAE -> SHINDAEBANG SAMGORI DISTANCE: 784[
2000 5 30
80
The Speed[km/h]
70
60
50
40
The Running Spee
The Target Speed
30
20
10
0
0
100
200
300
Th R
400
i
500
Di t
600
700
800
900
[ ]
Fig. 5 The speed profile of Seoul subway 7 line
(PORAMAE -> SHINDAEBANG SAMGORI)
Fig. 6 The running test view of automatic train control in Seoul subway 7 Line
4. CONCLUSIONS
This paper introduced a development technique of the on-board train control that plays an important role in safety and
the running performance of EMU. We implemented the automatic train operation with optimal control. We performed the
running test in Seoul subway 7 Line and proved that this technique is available for urban transit system.
5. REFERENCES
[1] K.J. Gemmeke, "Experiences with the implementation of the train communication network", Computers in
Railways•≥, Vol.1, pp253_267, 1994.
[2] Motoo Kusakari, Masaharu Sakuma, Eisuke Isobe, Tadashi Takaoka, "On-Board Train Information Control Network
Systems", Hitachi Review Vol.40, No.4 pp303_308, 1991.
[3] G. Legoff, P.Sainton, "Using Synchronous Language for Signalling".
[4] Tadashi Takaoka, "Onboard Train Control and Communication Systems for High-Efficiency Railway Systems",
Hitachi Review, Vol 46, No.2, 1997.
[5] Peter Pudney and Phil Howlett, "Optimal Driving Strategies for a Train Journey with Speed Limits, J. Austral. Math.
Soc. Ser. B36, pp38-39, 1994.
[6] P.G.Howlett and P.J.Pudney, energy-Efficient Train Control, Springer, pp215-228, 1995.
[7] Chang.C.S and Sim.S.S, "Optimising train movements through coast control using genetic algorithms", IEE proc.
Electr. Power. Vol.144, No.1, pp65-73,1997.
[8] Shiuh-Jer Huang, “ Fuzzy Control of Automatic Train Operation System“, International Journal of Modelling and
Simulation, Vol 17, No2, 1997