2-Drive Motor Control Unit for Electric Power Steering.
There is growing requirement of less fuel consumption and C[O.sub.2] emission since the fuel consumption is getting higher every year. The electric power steering (EPS) is increasing its number  because it reduces load to the engine compared to hydraulic power steering and it can keep working without engine in HV or EV. Moreover, EPS plays an important role in the driving assistance such as active safety and automated parking. It is expected to increase its number. Originally, the fail-safe concept of the early EPS system was designed to shut down the EPS assistance by the fail-safe functions in any case if there is a failure. Because the adaption of the early EPS systems started with the small vehicles, the driver needed less effort to make a turn even if the EPS system shut down. However, as the adaptation of the EPS expands to the heavier vehicles with the advance in the power electronics, it is more preferable to maintain assistance. It is more difficult for the driver to make a turn in the heavier vehicle such as pickup trucks and SUVs without assistance of the EPS. Reducing the very low possibility of shutting down the EPS even lower may be required in the heavier vehicle.
In order to reduce the possibility of shutting down the EPS, we've developed a new motor control unit (MCU). It is 2-Drive EPS-MCU with redundant motor drive units. Since this new MCU can maintain the EPS assistance by remaining motor drive unit even if one of the motor drive units doesn't work due to the failure in the system, our 2-Drive EPS-MCU can dramatically reduce the failure rate of the EPS. We have evaluated the conventional MCU and the new MCU according to the ISO 26262, one of the most well-known standards in the automotive industry.
CONFIGURATION OF 2-DRIVE MCU FOR EPS
1). Mechanical Structure of 2-Drive MCU
In Figure 1, the mechanical structure of 2-Drive EPS-MCU is shown. One feature of the mechanical structure is having the ECU integrated into the motor housing. ECU is mainly composed of the microcontroller, the power modules for 2-Drive inverter unit, the peripheral IC for the microcontroller abnormality detection and the heat sink for spreading heat which is generated by the joule heating of the power modules. By adopting the redundant design and by integrating the motor and the ECU, a more reliable and smaller product compared to our previous product has been achieved. The volume was reduced by 30% and the weight was reduced by 20%. Specific detailed technical viewpoints to reduce the product volume are the development of direct connecting method between motor wires and Power modules, called mechanical-electric connect. It allowed us to eliminate the bus bar module which connects the motor wires and the power modules for motor drive units. The conventional motor structure is shown in Figure 2.
Furthermore the 2-Drive motor drive concept could reduce the heat sink volume and the number of large size electrical parts by reduction of the maximum motor current to the half. Each power module manages the half of the required motor current. The sum of the output of the two power module configures the output of the motor.
As for the motor winding technology, there are two advantages. One advantage is noise reduction and the other advantage is better motor output density. This technology enabled our products to reduce the noise and vibration. We achieved the noise lowering of 7 dBA. Regarding the increase of the motor output power, we achieved 3.5% increase of the motor output by the arrangement of the windings.
2). System block of 2-Drive MCU
At first, EPS system consists of steering wheel, mechanical components which connect steering wheel and tire, torque sensor to detect driver's steering torque which is input data for EPS-MCU, and EPS-MCU which assists the driver's steering control. Figure 3 shows the system block diagram of 2-Drive MCU-EPS which we developed. As shown in Fig.3, our redundant concept is composed of the redundant torque sensors for detection of a driver's steering torque which is input signal for EPS-MCU, the redundant inverter units, the redundant winding wires and the redundant position sensors for detection of the motor rotation position. Each redundant function also has independent operation by each power supply. Therefore, if a malfunction happens in one of the drive line, the EPS steering assistance is able to be maintained by the remaining drive line. As for the microcontroller, in order to achieve ASIL-D based on ISO 26262, we adopted lock-step CPU architecture. The lock-step CPU architecture can detect a transient error and a permanent error, and this feature helps to enhance the EPS system reliability.
Since "2-Drive EPS-MCU" system architecture looks a bit complex and increases the number of the electrical parts, it looks like that the cost of redundant concept "2-Drive EPS-MCU" is increased, but it's not true. Because the optimization of 2-Drive motor current control allows us to reduce a number of large size electrical parts, such as an aluminum electrolytic capacitor. Also, the 2-Drive concept allows us to select smaller chip size power semiconductor devices. Since the current for the each power module is reduced, the heat generation on the power semiconductor devices by the joule heating becomes 25%. The reduction of the heat also contributes to the reduction of the size of the heat sink. Details of redundant 2-drive motor technology and 2-drive current control technologies are explained in the later chapters.
3). Electrical Configuration of 2-Drive MCU
Figure 4 shows an electrical circuit diagram of the single motor drive as the conventional concept. There is one electrical circuit and one winding. Figure 5 shows an electrical circuit diagram of the dual motor drive as the 2-Drive EPS-MCU concept which we developed. There are two electrical circuits and two windings.
As for the specific behavior of the conventional concept, single motor drive motor control (conventional MCU for EPS) during failure occurrence, if a failure happens in any part of the electric power steering, the EPS power assistance is shut off by fail-safe functions. Since the vehicle will revert to manual steering, steering control can be maintained, but greater driver's effort would be required especially at low vehicle speeds. On the other hand, our new redundant concept, the dual motor drive, can reduce the driver's effort because the EPS assistance is maintained by the remaining drive line even if one of the drive line is stopped by a malfunction occurrence. Figure 6 shows specific operation during fault occurrence in one drive line such as Inverter unit 2. In this case, Inverter Unit 2 is shut off by fail-safe functions, but since Inverter Unit 1 is able to keep providing motor current, a driver can steer the steering wheel with greater effort, but still less than if failure occurred in a single motor drive system.
The maximum torque of the 1-drive mode, or failure state in developed system, is half of the maximum torque in 2-drive mode, or normal state. In EPS, usually large torque is required only in stopped vehicles. Therefore we think that a half of the torque is sufficient to maintain EPS assist in the moving vehicle. Comparison of the maximum torque of the normal state in the conventional and the developed MCU, the failure state in the developed MCU, and the failure state in the conventional MCU is shown in Figure 7.
MOTOR TECHNOLOGY IN 2-DRIVE MOTOR CONTROL UNIT
The motor technology of the dual winding motor is introduced in this chapter. As shown in Figure 8, two windings are arranged in the common stator. The two windings are electrically independent. The detailed layout of the windings is shown in Figure 9. Two windings are arranged in the stator with [pi]/6 radians phase difference in the electrical angle.
In Figure 10, the phase current of the two windings and the output of the each winding with 6th harmonic distortion in the magnetic circuit are shown. As the windings are arranged with the [pi]/6 radians phase difference, the phase current of the two windings also has the [pi]/6 radians phase difference. With the phase difference, 6th harmonic distortion in the magnetic circuit of the windings can be cancelled in the total output of the windings.
The pole and slot combination is also redesigned in the 2-Drive MCU. 10-60 combination was chosen to reduce the noise and the vibration. Comparison of the deformation mode in the conventional motor and the developed motor is shown in Figure 11. The deformation mode of the new slot combination is higher compared to the conventional slot combination. With the new technologies, the vibration was reduced by 90% and the noise was reduced by 7dBA in MCU as previously mentioned in the former chapter.
CURRENT CONTROL TECHNOLOGY IN 2-DRIVE MCU
The control scheme is introduced in this chapter. To control the current of the dual winding motor, the sum of the current of the two windings and the difference of the current of the two windings are controlled . Equations (1) and (2) are the voltage equations of the windings with the mutual inductance. With the magnetic coupling, the windings interact with one another. The term of the mutual inductance represents the magnetic coupling of the dual winding motor.
[mathematical expression not reproducible] (1)
[mathematical expression not reproducible] (2)
[V.sub.d1], [V.sub.d2] : d - axis voltage of inverter1 and inverter2
[V.sub.q1], [V.sub.q2] : q - axis voltage of inverter1 and inverter2
[I.sub.d1], [I.sub.d2] : d - axis current of inverter1 and inverter2
[I.sub.q1], [I.sub.q2] : q - axis current of inverter1 and inverter2
[omega] : angular velocity of the motor
R : resistance of the winding
L : inductance of the winding
M : mutual inductance between winding 1 and winding 2
[[empty set].sub.0] : magnet flux
From the equations (1) and (2), equations (3) and (4) are derived. Equations (3) and (4) are the plant models when controlling the sum and the difference of the current of the two windings. Equation (5) represents the relationship between the currents of the windings and the sum or the difference of the current of the two windings.
[mathematical expression not reproducible] (3)
[mathematical expression not reproducible] (4)
[mathematical expression not reproducible] (5)
The block diagram of the plant is shown in Figure 12. The controller is designed based on the plant model.
The control block of the dual winding motor is shown in Figure 13. The controller consists of two sets of PI controllers and four transformations. The first transformation is transformation from the sum and the difference of the current of the two windings to current of the each winding. The second transformation is transformation from the d-q axis to three phases. The third and the fourth transformations are the inverse transformations of the first and the second transformation. With the transformations, the plant model of the dual winding motor can be assumed as the plant model shown in Figure 12. The plant model is less complicated compared to the plant model based on the equations (1) and (2). The PI controllers are designed to control the 4 different first-order lag systems.
INVERTER DRIVE TECHNOLOGY IN 2-DRIVE MOTOR CONTROL UNIT
The PWM strategy which contributes to the reduction of the size of the capacitor is introduced in this chapter. The capacitor placed at DC bus line of the inverter is necessary for the stable DC bus voltage and the smooth current control of the inverter. The capacitor is one of the large components in the inverter. Its size is determined by the required capacitance and the degrading of the capacitance in the application. The degrading of the capacitance is determined by the ambient temperature and the root mean square of the capacitor current. If the capacitor current increases, the capacitance degrades more. It is necessary to increase the capacitance and the size of the capacitor increases. However, if the capacitor current decreases, the capacitance degrades less. The size of the capacitor is reduced since less capacitance is necessary.
To reduce the size of the capacitor, the timing of the charging and the discharging of the capacitor by the inverter 1 and the inverter 2 is changed. In the PWM control of the inverter, there is state called the zero voltage vector and the effective voltage vector. In the state called zero voltage vector, all three high side switches are on or off. At the zero voltage vector, capacitor is charged. In the state called effective voltage vector, one or two high side switches are on or off. At the effective voltage vector, the capacitor is discharged. The capacitor current in the zero voltage vector and the effective voltage vector is shown in Figure 14.To change the timing of the charging and the discharging by the two inverters, the neutral voltage of the inverter 1 is shifted to higher voltage and the neutral voltage of the inverter 2 is shifted to lower voltage. The modulation wave of the inverter 1 and the inverter 2 is shown in Figure 15.
The capacitor current at the interval 1 in Figure15 is shown in Figure16. With the neutral voltage shifting, the timing of the charging and the discharging by the two inverters are changed. Since the timing is changed, the peak of the capacitor current is reduced and the root mean square of the capacitor current is reduced. The capacitor current at the interval 2 in Figure15 is shown in Figure17. At the interval 2, the reduction of the capacitor current is less compared to interval 1.
APPROACH TO REDUCTION OF TOTAL FAILURE RATE
There are several methods to estimate the failure rate of electrical components. In automotive industry, IEC TR 62380 is one of the most common standards for failure rate estimation. Generally, increase in the number of parts leads to the increase in the failure rates. There is a case when the total failure rate increases in the redundant system. However we could install the 2-Drive function without increasing the total failure rate. We could reduce the total failure rate by 20%.
First of all, we identified the higher failure rate function. In case of 2-Drive EPS MCU which we developed, Figure 18 shows the distribution of the total failure rate by each function.
As shown in Figure 18, the highest failure rate of the EPS-MCU functions is the motor drive unit. The motor drive unit is composed of the two power modules, which have 11 MOSFETs, and the pre-drivers, which have functions of driving MOSFETs, and the passive components such as resistor and capacitor. Among these components in the motor drive unit, the highest failure rate component is the power module. It accounts for about 75% of Motor Drive Unit's failure rate. Therefore reduction of the failure rate of the power module is effective in reducing the total failure rate.
According to the equation (6) from IEC TR 62380 8.5 Power Transistor, the failure rate of the power devices consists of the die failure rate, the package failure rate and the overstress failure rate.
[lambda] = [[lambda].sub.die] + [[lambda].sub.package] + [[lambda].sub.overstress] (6)
From the equation, it is known that the die failure rate is very small (less than 1% of power module failure rate) and if overstress factor is same in the same application, the EPS, the package failure rate decides the failure rate of power module. Therefore we focused on reducing the package failure rate. Generally, smaller packages can reduce its failure rate because heat shock stress becomes smaller.
As I previously mentioned, our 2-Drive EPS MCU has two power modules for redundancy and there is 50% of motor current flowing in the each power module. It means that the sum of the output of two modules is 100%. Reduction of maximum motor current allows us to select smaller chips of power device. As a result, we are able to reduce the power module size with the smaller chips size power device.
The well-known standards IEC TR 62380 for failure rate evaluation of variety electrical components shows failure rate equation for the power transistor shown in equation (7). The failure rate of this power module is able to be evaluated as well.
[mathematical expression not reproducible] (7)
Figure 19 shows evaluated total failure rate ratio of the conventional motor drive unit (MOSFETs) and the new design (Power Module). As for total of conventional, it was estimated by multiplying 11 to the MOSFET failure rate estimated based on the equation (7). The failure rate ratio in Figure 19 is defined so that the total failure rate of the conventional motor drive unit is 1.0.
The MCU that is more safe with redundant 2-drive design and is better in installation with integration of motor and ECU is newly developed. Even if a failure happens in one of the drive unit, developed MCU can maintain the EPS power assistance by the remaining drive unit. This paper has presented an approach to reduce a total failure rate of a product by reducing the size of the power device chip. Moreover, 2-drive technologies were adopted in the MCU. Motor windings were placed with [pi]/6 radians phase difference to reduce torque ripple and increase torque, 3 phase neutral voltage was shifted to changing timings of discharging from the capacitor to lower the RMS of the capacitor current, and the sum of the currents in two windings were directly controlled for high response torque control.
[1.] Kuebler, E. "Electric Power Steering System: Market Requirement and Application Range with Special Focus on Column-EPS" Proceedings of FISITA 2012 World Automotive Congress, Lecture Notes in Electrical Engineering 198
[2.] Hayashi, J. "Road Map of the Motor for an Electric Power Steering System" Chassis Tech plus 4th International Munich Chassis Symposium, 2013, p317-326
[3.] Suzuki Takashi, Kabune Hideki, Ito Norihisa, Ito Akira: "2-Drive Motor Control Unit for Electric Power Steering", Proceedings of FISITA 2014 World Automotive Congress, F2014-IVC-117, 2014
[4.] IEC TR 62380 First Edition 2004-08 p.44
Application Engineer - Safety Systems Engineering Division
DENSO International America, Inc.
24777 Denso Drive, Southfield, MI 48033 USA firstname.lastname@example.org
EPS - Electric Power Steering
MCU - Motor Control Unit
PWM - Pulse Width Modulation
MOSFET - Metal Oxide Semiconductor Field Effect Transistor
Mikihiro Hiramine, Yoshitaka Hayashi, and Takashi Suzuki
|Printer friendly Cite/link Email Feedback|
|Author:||Hiramine, Mikihiro; Hayashi, Yoshitaka; Suzuki, Takashi|
|Publication:||SAE International Journal of Passenger Cars - Electronic and Electrical Systems|
|Article Type:||Technical report|
|Date:||Aug 1, 2017|
|Previous Article:||Shared Autonomous Vehicles as a Sustainable Solution to the Last Mile Problem: A Case Study of Ann Arbor-Detroit Area.|
|Next Article:||Use of Electric Driveline for a Cooperative Braking Strategy.|