The Design of Multi-Platforms Rail Intelligence Flatness Detection System

International Journal of Research in Engineering and Science (IJRES)
ISSN (Online): 2320-9364, ISSN (Print): 2320-9356
www.ijres.org Volume 5 Issue 5 ǁ May. 2017 ǁ PP. 30-35
www.ijres.org 30 | Page
The Design of Multi-Platforms Rail Intelligence Flatness
Detection System
Qian Yuan-yuan1
,Gao Fei1
,Sun Xian-kun1
,Zhu Lin-rong2
（1.School Of Electric And Electronic Engineering, Shanghai University Of Engineering Science, Shanghai
201620, China；2.Shanghai Railway Bureau,Shanghai,200000)
ABSTRACT: In this paper,we design a Multi-platforms intelligent system for flatness detection of rail
welding headbased on thedevelopment environment of Android software .The system uses a STM32 chip as
control core, a handheld smart terminal or personal computer as the carrier. The datas transmitted to intelligent
terminal or computer through the bluetooth communication technology are processed rapidly, the data curve is
drawed and the flatness characteristic parameters of the measured rail welding head is identified. The system
provides a friendly intuitive monitoring and operation interface, has the characteristics of fast, reliable, energy
saving, high accuracy, etc.
Keywords: Multi-platforms FlatnessBluetooth
I. INTRODUCTION
In recent years, the high-speed railway develops violently in China , which greatly increase the capacity
of the Chinese railway, promoted the rapid development of China's economic and social.Long rail is made from
a 100-meters one .the smoothness of the welding head is closely related to the safety and stability of train.
Therefore,during the production and daily maintenance of long rail, the flatness detection of the rail welding
head is of great importance. At present, the flatness detection of rail welding head mainly has the following
several ways: first, manual feeler inspection;second, large detection device;third, detection device based on the
MCU system .The solutions above all have shortcomes such as: inconvenience on use and operation; poor
human-computer interface leading the inconvenience on display and observe;the ability for data processing of
the inter microcomputer system is poor, and data is saved as a special format whichis not suitable for sharing.
Hardware Design
The system focuses on analysingthe variability of the environment in which the detection system is
used , and strives to adapt to the measurement taskunder a variety of different environments . In general, the
detecting system is required to be lightweight, portable and easy to use; at the same time , it is also required to
ensure a certain measurement accuracy ,shorter measurement time, provide a convenient and user-friendly
interface and ensure the energy efficient . Taking the situations aboveinto account , the control block diagram of
the system shown in Figure 1.
STM32 32-bit Flash Microcontrollers based on the ARM Cortex-M3 processor has opened a new
development degree of freedom, and has a powerful processing capability of digital signal , and the features as
low power consumption, low operating voltage, and high compatibilitymeet the system requirements. The
measurement system uses STM32 as the control core to complete the execution of the operation command and
the data collection and transmission. The hardware part uses the Festo straight line electric cylinder EGC-TB-
KF, Baumear sensor S14C as well as the thunder race stepping motor driver M353.
Samrt phone
PC
STM32
Cortex-m3
Stepper motor
Super Linear Guide
stepper
motor driver
electric eddy
current sensors
Figure1 Principleblock diagram for Measurement system

The Design Of Multi-Platforms Rail Intelligence Flatness Detection System
Design Ideas
An eddy current sensor is installed on a 1-meter long linear cylinder, andthe host computer sends out
detection command . The embedded control system arranged in the measuring device controls the stepper motor
to rotate, by this means the high-precision eddy current sensor is driven to do a straight line movementalong the
guide rail upon the slide table . At the same time, the data acquisition system collects the sensor signals high-
speedcontinuously , and the data collection is completed until the end of the trip. At this point, the system enters
the standby mode of reverse measurement, meanwhile, the embedded control system processes the datas and
sends them to the host computer to be processed and the measurement results is displayed. In the same way,
after receiving the detection command in the reverse measurement mode, the motor reverses and the sensor
moves in the reverse direction. The data acquisition system carries out high-speed continuous sampling on the
rail surface until the start of the cylinder. At this time, the system enters the forward measurement mode ,and the
datas are processed and returned to the host computer and the measurement results are displayed . At the first
time to start , in order to ensure that the detection system presents the data according to the detection direction of
the forward easurement mode,the sensor position is automatically restored to the cylinder start position .System
requirements are shown as fllow :
Detection distance: 950mm; detection accuracy: less than 50um; the detection time is less than 15s.
Working Principle of Control System
System schematic diagram (part) is shown in Figure 2
Figure2Principle diagram of the embedded control system
Working principle:
The control system is based on STM32F103C6T6 controller, and the chip integrates 12-bits A/D
converter, timer, synchronous serial port and so on. It can achieve data acquisition, stepper motor control, data
transmission and other functions. In order to improve the accuracy of A/D conversion, the voltage reference
chip REF192 is used , by collecting sensors and external reference voltage at the same time, the unknown signal
is compared with the known external referencevoltage to reduce the impactcaused by the external temperature
changes , and obtain accurate voltage datas. The electric cylinder driven by the stepper motorcarries the sensor
for scanning. The travellength of the stepper motor is determined by the number of control pulse, and the speed
is controlled by the frequency of the pulse. In order to adapt to the acceleration or deceleration curve of the
stepper motor, and scanwith precise fixed length ,the system uses a pulse generator cascaded by two timers in
the chip to form a variable frequency, variable number pulses to control the stepper motor.
Considering the battery capacity and the needs of field operations, with the purse of extending the
battery lifeas far as possible , the embedded system automatically enters the standby state when the system
standby ,and waits for the wake-upsignal to start detecting . The system also designs the one-button switch
circuit for the zero standby current , the system is energized,after the open or shutdown button is pressed ,the
controller starts and controls the solid state relay paralleledon the open, shutdown button to complete the self-
locking. With this circuit, the timer will shutdown if the system does not be operated at a long time , the circuit
is shown in Figure 3.

Figure3 Standby switching circuit
Software Design
The wireless communication technology between multiple-platforms structure and the embedded
system has been considered in this system , and it achieves an unified communications solutions between
variety of equipments and embedded system , also designs the program based on embedded system to achieve
hardware control, data acquisition, data communication, has realized the unity storage solutions of datas from
multiple platforms . Software architecture is shown in figure 4.
Android is a Linux-based open source operating system, is mainly used in software design for portable devices,
in this environment, the design has the following modules:
Android系统软件
检测功能数据存储
检测模块通信模块自校功能显示模块
数据格式化模
块
数据管理模块
Figure4 Software architecture of the system
Detection Module： this module is used to set up necessary marking parameters, send out the
measurement instruction to the data acquisition system through communication module, and get the testing data;
After further processing and optimization ,the data is transmitted to the display module and data formatting
module.
communication module: It is an important part of the complete set of testing system, the bridge
between PC and data acquisition system. This system follow standard wireless communication protocol, and use
the multi-threading technology to complete data communication between PC and data acquisition system.
Self correcting module: different work environment, temperature and humidity will cause different
degrees of deformation to the mechanical part, the self-calibration need to be done to eliminate the system error
when the work environment changes.
Display module: it is a public module shows the data dealed and the test datas , related judgment
results are presented to the operator by means of graphics on it .
Data storage: This moduel will format the detected datas into an unified format and store them in the
data management module.Because there are many different platforms and PC operating systems, the unified
data format which is useful in different platform and different operating system is indispensable.
Data management module: completes the founction of data storage and reload.

Programming Introduction
The example program for packet processing :
Packet processing is divided into two parts, one is to parse the received packet, and then trigger the
corresponding events;Another is to generate the corresponding packets according to the corresponding
command and the packets are sented by the bluetooth module. The flow diagram of Data reception is shown in
figure 5.
Receive data
Wait for the
header of packet
N
Is the header of
packet
Y endN
Wait for the length
of packet
Y
end
Wait for the
length of packet
Get the length of
packet
Y
Wait for the main
data
end
The length got =the
length of packet
Wait for the main
data
N
end
Wait for the header
of packet
Y
end
Figure5 Flow chart of data reception
The specific program code is shown as follows:
private void bulidPacket(String cmdName,byte[] data)
{
if(_isSendingPacket)
return;
_isSendingPacket = true;
int size = 10 + (data == null?0: data.length);
byte[] bytes = new byte[size];
bytes[0]= (byte) 0xff;
bytes[1]= 0x0;
bytes[2]= (byte) (size& 0xff);
bytes[3]= (byte) (size >> 8);
bytes[8]= (byte) cmdName.charAt(0);
bytes[9]= (byte) cmdName.charAt(1);
if(data != null)
{
for (int i =0; i < data.length; i++) {
bytes[i+10]= data[i];

}
}
_rDeviceBt.SendDate(bytes);
_isSendingPacket = false;
}
Experiment and Result Analysis
In the condition of laboratory, the experiment for multi-platform portable intelligent rail flatness detection
system was carried out about the accuracy and repeatability of datas .
Table1 Test results
Number calibration value
(mm)
first time
(mm)
second time
(mm)
third time
(mm)
1 0.02 0.023 0.019 0.017
2 0.05 0.051 0.053 0.048
3 0.1 0.104 0.101 0.099
4 0.2 0.198 0.201 0.203
5 0.35 0.351 0.353 0.348
6 0.5 0.504 0.501 0.503
7 0.65 0.648 0.649 0.651
8 0.8 0.801 0.8 0.803
9 0.9 0.901 0.902 0.899
10 1.0 1.001 1.004 1.003
During the experiment ,10 thickness ranging feelers are placed on a 00 level ruler and detected by the
system designed in this paper. the experiment is repeated three times and the results are shown in table 1. the
measurement error is within 0.01 mm and repeatability within 0.005 mm , Comply with the design
requirements . The operation interface on intelligent terminal is shown in figure 7.
Figure7 Operation interface on Terminal
II. CONCLUSION
In this paper , we design the software and hardware of the intelligent detection system for the rail and
implements multiple-platforms detecting the surface roughness of rail welding head and realizes the unified
storage of multiple-platforms data . Various flexible cases are adopted and the the follow-up maintenance costs
significantly reduced. the test of Shanghai design institute proves that the system meets the design requirements

REFERENCES
[1]. Yu, J. Q. ,&Yu, B.(2014). Development and application of STM32F series ARM Cortex M3 nuclear
micro controller.Tsinghua university press.
[2]. James,W. C. (Eds),&Ye, B. (2011). c # design pattern. Science press.
[3]. Ning, Y. P. ,&Liu ,Z. F.(2005) .The current research status of high precision straightness measurement
technology at home and abroad. Modern manufacturing engineering,6,21-30.
[4]. Wang ,Y. S. (2010). Development tour of the Android platform . Bei jing,Mechanical industry press.
Author Information
Yuanyuan Qian
Shanghai University of Engineering Science
Longteng road333, songjiang district, Shanghai
Contact e-mail:1638581957@qq.com
Fei Gao
Xiankun Sun
Linrong Zhu
Shanghai railway bureau
Temmoku east road 80,Shanghai

The Design of Multi-Platforms Rail Intelligence Flatness Detection System

More Related Content

What's hot (19)

Similar to The Design of Multi-Platforms Rail Intelligence Flatness Detection System (20)

More from IJRESJOURNAL (20)

Recently uploaded (20)

The Design of Multi-Platforms Rail Intelligence Flatness Detection System