Journal of Physics:

Conference Series

PAPER • OPEN ACCESS You may also like

- Hybrid and electric vehicle tribology: a
Simulation and application of the electric power review
Hemanth G, Suresha B and
steering system simulation platform based on Ananthapadmanabha

- The analysis of diagnostics possibilities of

Simulink the Dual- Drive electric power steering
system using diagnostics scanner and
computer method
To cite this article: Senqi Gao et al 2022 J. Phys.: Conf. Ser. 2402 012007 W Szczypiski-Sala and K Dobaj

- The manufacturing method of the

composite worm gear housing using the
cold injection
A Baier, K Herbus and P Ociepka
View the article online for updates and enhancements.

This content was downloaded from IP address 171.229.232.139 on 29/11/2024 at 02:51

ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

Simulation and application of the electric power steering

system simulation platform based on Simulink

Senqi Gao, Rui Sun, Ying Fu*, Jianguo Xu

Research & Development Dept. BOSCH HUAYU Steering Systems Co., Ltd.
Shanghai, China
[email protected]
[email protected]
[email protected]
[email protected]

Abstract—The principle of the electric power steering system is introduced in this paper, and
the simulation platform based on Simulink/Simscape is established. The mechanical structure
of the steering gear, the timing scheduling strategy, the hardware control circuit, and the
integration of the software control algorithm are modeled and verified in detail. It is proved
that the simulation model is reliable and highly precise. The simulation platform is used to
analyze the actual problems, and propose an improvement plan. The results show that the
platform can improve the intelligence of development and quickly locate, effectively reproduce
and analyze actual problems, and reduce the trial-produce cost of the early production line,
which has a certain guiding significance for software and hardware development.

1. INTRODUCTION
The steering system is always an important part of automobiles and is also one of the products with the
highest functional safety requirements. It has high production process requirements, a high scrap rate,
and long and difficult sample modification. Improving EPS operation stability and safety has always
been the fundamental purpose of electric power steering. The current method is to process manual
samples for software, hardware verification, and manual driving and also build complex test benches to
verify and debug. Therefore, improving automated testing and rapid verification can save costs,
increase development speed, and reduce unnecessary manufacturing errors.
Electric Power Steering system (EPS) is a kind of Power Steering system with the auxiliary force
provided by a motor [1]. In the current research and development process of EPS, most of the
experiments will be carried out after software integration, including hardware-in-the-loop (HIL),
vehicle tuning, durability, and other tests [2]. Many problems will be found during the test. The current
mainstream problem-solving mode is that test engineers give feedback on problems, and the test data
will be sent to the engineers through a series of processes. Then they analyze, reproduce, improve, and
verify the problems. This mode requires a relatively long period and more human and equipment
resources. Therefore, a high-precision simulation platform is of great significance, which can greatly
reduce the time and resources required for problem reproduction, problem improvement, and problem
verification [3]. In this paper, a simulation model of EPS is established and analyzed in the
Simulink/Simscape simulation environment. The model mainly includes mechanical parts such as

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

steering gear, worm, gear and rack, power-assisted motor, circuit control hardware, and electronic
control unit ECU software control strategy. It can not only verify whether the mechanical parameters
are reasonable, but also optimize the software algorithm strategy.

2. EPSSYSTEM COMPOSITION AND TRANSMISSION PRINCIPLE

This paper uses the column-type electric power steering system as the system modeling and simulation
object. The principle of its mechanical structure transmission system is shown in Figure 1. This system
comprises a steering wheel, torque sensor, worm gear reduction mechanism, booster motor, steering
column and rack and pinion, and ECU electronic control unit. The basic working principle is to transmit
the vehicle speed signal sent by the vehicle and the torque signal measured by the torque sensor to the
electronic control unit, and then calculate the rotation direction and current size of the booster motor.
The steering column helps the driver overcome the resistance of the ground acting on the tires and the
inside of the system, thereby realizing power steering control.

TAS ECU Motor Turbo-Worm

Intermed
iate shaft

Rack and
pinion

Figure 1. EPS structure diagram

3. EPS SYSTEM MODEL

Simulink is a signal-flow-oriented simulation modeling environment, where signals flow directionally
from one to another. Simscape is module and block-oriented as a Simulink extension tool, and it can be
used to model multi-domain physical systems. As an object-oriented physical modeling method, the
corresponding mass, spring, and damping modules can be simply connected from the library if a mass-
spring-damper system needs to be modeled by Simscape. On the contrary, it is necessary to derive the
differential equation first and then model it in Simulink. The difference is shown in Figure 2.

Figure 2. The difference between the Simulink physical domain and the mathematical domain

The Simscape library provides basic components, including ten physical domains. There are
mechanical, electronic, and fluid types, involving resistors, springs, racks, and pinions, as well as more
complex components, such as electric drives, transmissions, heat exchangers, etc. According to the

2
ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

schematic, various model systems, such as motors, bridge rectifiers, and refrigeration systems, can be
built by assembling basic components.
In addition, Simscape also supports the creation of custom component libraries. Through Simscape,
users can quickly create physical system models for their products and build software-in-the-loop
systems. Firstly, for the column-type electronic power steering system studied in this paper, it is
necessary to build:
• PowerPack model: including motor and control circuit
• Steering gear model: including torque sensor, worm gear reduction mechanism, booster motor,
rack and pinion, Coulomb friction, and other mechanical structures
• ECU software control logic: including the corresponding control strategy based on model-based
development (MBD), basic functions of the application layer such as onboard assistance, active return,
etc., current control strategy, and task scheduler.

3.1. Mechanical Model

In this paper, the column-type electric power steering mechanism model is established by using the
Simscape toolbox. This method does not require force analysis or the establishment of complex
mathematical differential equations. It only requires modular modeling of the actual structure of the
product.

Figure 3. EPS mechanical model

As shown in Figure 3, all connections between modules are physical domain connections, which
means that all connection points can transmit corresponding physical information like a real system,
such as torque transmission and displacement transmission. The steering system model established by
Simscape will restore as many physical properties as possible to real products. Depending on the
mechanical structure of the product, model variants such as double pinion assist, rack assist, and
steering by wire can be quickly carried out.

3.2. Motor And Hardware Model

The PowerPack part of the electric power steering system is shown in Figure 4. The model of the motor
and hardware part can be simplified as a permanent magnet synchronous motor, a battery, and a three-
phase bridge.

3
ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

Figure 4. Motor and hardware model

3.2.1. Motor
The motor model uses the permanent magnet synchronous motor model PMSM in the Simscape library.
The vector control of PMSM is based on the dq0 coordinate system of the motor, as shown in Figure
5[4]. The permanent magnet generates the rotor magnetic field. The permanent magnets generate the
rotor magnetic field so that the magnetic flux varies sinusoidally with the rotor angle. For the axis as
specified in Figure 6, when the rotor mechanical angle is zero, the fluxes of the a-phase and the
permanent magnets are aligned. According to the coordinate transformation theory, the voltage, flux
linkage, electromagnetic torque, and power equations of this model can be obtained [5].

Figure 5. PMSM structure of permanent magnet synchronous motor model

The motor model parameters used in this paper are set based on the parameters calibrated by a real
motor on this platform. The motor output end R will be connected to the worm gear of the mechanical
model to output power assist. The data collected by the R, such as the electrical angle, and the motor
speed, will be sent back to the ECU for the field-oriented control algorithm.

3.2.2. Three-phase Bridge

The three-phase inverter circuit consists of 6 MOSFETs. As shown in Figure 6, the positive and
negative electrodes of the circuit are connected to the battery. The gate of each MOSFET is connected
to the PWM wave generated by the FOC, and the upper and lower groups are grouped. The switching
sequence of MOSFETs is controlled by PWM, thereby generating three-phase sinusoidal voltages with
a phase difference of 120 degrees to drive the motor.

Figure 6. Three-phase bridge model

4
ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

3.2.3. Battery
The battery model simulates a standard 12V vehicle battery to provide the power needed to operate the
electric power steering system. Each cell is built according to the example described in the Simscape
case library, and the battery model consists of several cells connected in series. The model simulates the
voltage conditions of the battery under different SOC (State of Charge) conditions, where the SOC is
the ratio of the remaining capacity to the total battery charge [6]. Through the battery capacity and
current state, the current SOC change is calculated, which can be expressed as:

Q −  idt (1)
SOC = 100%
Q
In the formula: Q is the total power of the battery, which is set to 50 AH in this model.

3.3. ECU Software Control Model

3.3.1. OS Timing Scheduling Model

The ECU software design is based on the AUTOSAR software architecture, which includes functional
modules and a timing scheduler that schedules these functional modules. The timing schedule is
designed according to the Tasking schedule, which is an ideal timing sequence without considering
interruption. It can reproduce the real software execution sequence better, as shown in Figure 7.

Figure 7. The working principle of the software

3.3.2. Control Strategy Model

The software control logic design mainly includes the application layer, complex driver layer, and
system service layer, as shown in Figure 8.
The application layer mainly includes the following four parts:
• The signal processing module is used to receive and transmit the steering wheel torque and
vehicle speed signals from the internal sensor signals;
• Steering function module, including basic power assist curve, inertia compensation to ensure
sensitive power switching at high and low speed, friction compensation power to reduce mechanical
loss, active damping to provide damping torque;
• Passive steering function module, including active return force, when the rack reaches the end,
the soft limit function is used to reduce the power assist to protect the mechanical structure from
collision;
• In the torque limit protection module, when the temperature of the hardware, such as the motor,
is too high, the output of the requested torque will be limited to protect the hardware system.
The complex drive layer is mainly the motor control module, which receives the requested torque
signal from the application layer, uses the PID algorithm to accurately track and control the target
current through the torque and current control module, and calculates the real-time duty cycle output to
control the three-phase bridge switch.

5
ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

The system service layer is mainly the task scheduler of the application layer and the motor control
module, which sets the priority of the runnable tasks in each module.
Application layer

OS Task
Active
Assist Curve
Damping
TAS Torque

Sensor signal
Inertia Friction
Compensation Compensation
Motor Speed Motor Control
Steering function module over-temperature
+ protection current control PWM Motor

Vehicle Speed Torque Limit module -

Vehicle signal Active software
Signal processing Return stroke limit
module
Passive steering function module

Figure 8. Control strategy logic

4. SIMULATION MODEL VERIFICATION

In the MATLAB/Simulink environment, the model of each module group is firstly analyzed for the
individual matching degree by using the built-in toolbox in Simulink, and then the entire platform is
analyzed under the premise of ensuring a single accuracy. To verify quickly, the experimental bench is
used to collect data, which is used as the input signal of the platform and the baseline signal of the
benchmark. The fault tolerance is set to ±10%. This paper mainly introduces some of the results. For
the EPS system, the torque output signal and the rack position signal are one of the most important
signals.

Figure 9. Comparison of motor output torque

As can be seen from Figure 9, the overall trend of the motor output torque signal is consistent, and
the accuracy can reach more than 90%. There will be glitches in the acquisition signal, and the actual
cycle of the sampling hardware is not completely fixed step size. While the simulation is a fixed step
size and non-continuous change, this error at the peak will appear.

Figure 10. Comparison of rack positions

6
ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

It can be seen from Figure 10 that the overall trend of the rack position is consistent, but there is a
small amount of deviation in accuracy. It is less than 10%. The springs and racks of the rear bench will
be worn to a certain extent, and the simulation model is based on the ideal spring stiffness and mass.
To sum up, the accuracy of this simulation platform is as high as 90% or more, and it can be used in
fields such as verification of control algorithms, analysis of practical problems, test evaluation before
software release, etc. It can assist in software development and release.

5. SIMULATION ANALYSIS OF EPS SYSTEM MODEL ACTUAL CASES

Based on the verification of the simulation model in Chapter 3, this simulation platform can be used for
qualitative and quantitative analysis of torque direction. An actual case is selected as an example in this
paper during the test of the real vehicle and the bench. When the steering wheel is turned rapidly, the
test engineer subjectively feels that the assistance is not enough. The torque data flow in the model can
be observed by simulating the torque at the steering wheel end on the simulation platform, given a
torque input of 5s increasing from 0 to 4.2. As shown in Figure 11, it can be found that the torque-speed
(TN) limit curve of the torque limit protection module in the application layer greatly affects the
requested torque output in the second half.

Figure 11. Requested torque comparison chart

Figure 12. Comparison of TN limit curves

Figure 13. Comparison of motor output torque

7
ISRIMT-2022 IOP Publishing
Journal of Physics: Conference Series 2402 (2022) 012007 doi:10.1088/1742-6596/2402/1/012007

By comparing with the TN limit curve in the motor control, as shown in Figure 12, it can be found
that the limit value in the application layer is smaller than that in the motor control module. Therefore,
if the limit in the application layer is enlarged or the two limit curves are set to be consistent, the actual
request torque output of the application layer can be completely transmitted to the motor control
module. As shown in Figure 13, the output torque of the motor after changing the limit is significantly
larger than the result before the change in the second half, which can improve the problem of
inadequate steering wheel assist feeling by drivers.
If the problem occurs in the mass production stage, the production line will be stopped; if it occurs
in the testing stage, it will take longer to solve the problem, in the absence of a simulation platform. The
time required before and after the investigation is shown in Table 1. It can be seen that efficiency has
increased by more than 90%.

Table 1 Troubleshooting Efficiency Comparison

Efficiency
Check method Number of staff Required resource
Time
required information
OE, computer and
Field test 3 other equipment, 20h
vehicles, test sites
OE, computer, test
Bench test 1 bench, and other 10h
equipment
Simulation test 1 computer 3h

6. CONCLUSIONS
Based on Simulink/Simscape tool, the mechanical structure, power-assisted motor, hardware control
circuit, and control logic strategy of the column-type electric power steering system are modeled in
detail. This platform can help analyze and solve the problems in the actual vehicle test, and a software
test environment that fits the actual product can be built, which can conduct targeted simulation and
optimization of the ECU control logic in combination with the actual problem. Then the software
development and mechanical hardware verification can be iterated quickly. At the same time, this
scheme can be further extended to the development of different types of electric power steering systems,
providing practical simulation ideas for the development of power steering by wire.

REFERENCES
[1] Pan G, Min D, Wang Q. Electric power steering system control strategy and simulation
technology research[J]. Automobile Applied Technology, 2017(24):33-40.
[2] Liao L, Shi H, Zhang J, et al. Design of Performance Test System for Vehicle Electric Power
Steering System [J]. Journal of Chongqing University of Technology (Natural Science),
2018,32(06):41-46.
[3] Khalkhali A, Shojaeefard M H, Dahmardeh M, et al. Optimal design and applicability of electric
power steering system for automotive platform[J]. Journal of Central South University,
2019,26(04):839-851.
[4] Yao C, Haibin L U. Permanent Parameter Simulation of PMSM for EV[J]. Micromotors,
2019,52(1):21-25.
[5] Ning B, Cheng S, Lu S, et al. A direct torque control strategy of PMSM based on duty ratio
technology[J]. Chinese High Technology Letters, 2019,29(11):1095-1103.
[6] Li B, Bei S. Estimation algorithm research for lithium battery SOC in electric vehicles based on
adaptive unscented Kalman filter[J]. Neural Computing and Applications, 2019(31):8171-
8183.