Novel Clamping Force Control For Electric Parking Brake Systems
Novel Clamping Force Control For Electric Parking Brake Systems
Novel Clamping Force Control For Electric Parking Brake Systems
SaB1.4
AbstractIn this paper we present a novel claming force control method for electric parking brake systems without using force sensors. The control method consists of three steps for a simple control structure. First, we show how to approximately detect the initial contact point between brake pads and a brake disk with the angular velocity of motor. Second, the clamping force is estimated with the function of angular displacement from the maximum angular velocity. Third, since the motor continues to rotate about several tens milliseconds even after power-off we propose a novel on-off control method to decrease the inertia effect of DC motor. The proposed control method is validated by experiments. It enables low cost manufacturing of electric parking brake systems due to its simple control structure.
I. INTRODUCTION
electric parking brake (EPB) system that generates clamping force for parking by controlling a DC motor has been introduced to improve safety and convenience for drivers. At the press of a button on the dashboard a driver can easily apply or release the parking brake. The EPB systems provide some advantages. They allow for a larger interior space than conventional parking brake systems. And drivers with physical difficulties can activate the parking brake easily just by pushing the button. The EPB systems also support dynamic braking and anti-lock braking system (ABS) as well as the conventional parking brake function [1], [2]. Because of these advantages, the production of vehicles equipped with EPB systems is increasing. Force sensors have been used to measure clamping force in EPB or electric mechanical brake (EMB) systems, which are brake-by-wire systems. However, it is difficult to install the force sensor due to the limited mounting space and the cost of the sensor [3]. Therefore, a clamping force estimation method
ECENTLY,
Manuscript received February 15, 2009. This work was supported in part by the Minster of Knowledge and Economy, Republic of Korea under Grant 10014728. M. Jang, Y. O. Lee, W. Lee, C. W. Lee are with Dept. of Electrical Engineering, Hanyang University, Seoul 133-791, KOREA (e-mail: [email protected], e-mail: [email protected], e-mail: [email protected], and [email protected]) C. C. Chung is with Div. of Electrical and Biomedical Engineering, Hanyang University, Seoul 133-791, KOREA (+82-2-2220-1724; fax:+82-2-2291-5307; e-mail: [email protected]) Y. Son is with Central R&D Center, MANDO Corporation, Kyonggi-Do 446-901, KOREA (e-mail: [email protected])
978-89-956056-9-1/09/2009 ACA
1588
SaB1.4
In Fig. 2, we see that the final clamping forces are different even for the same angular displacement due to different clearances. To resolve this problem, we studied the relationship between the angular velocity of motor and the contact time.
5 4.5
II. CLAMPING FORCE ESTIMATION A. Structure and specifications of EPB system In this paper, we study a stretch type EPB system depicted in Fig. 1. It includes a controller, parking cables, a DC motor, a gear box, an assembly of screw, a nut, a controller, a motor driver, and a current sensor. The parking cables of EPB system are connected to the brake pads. There are two operating modes: applying and releasing force mode. In the applying force mode, the clamping force is increased by pulling the parking cables using the DC motor until it reaches the target force [13]. In the releasing force mode, the clamping force is decreased by releasing the parking cables reversely. In order to measure the angular displacement of motor, the current sensor is used by using current ripples as [5], [6]. In Table I, specifications of EPB systems are listed [7].
TABLE I SPECIFICATIONS OF EPB SYSTEM Target force Settling time Permitted error bound 80 100 [kgf] Less than 1 [s] 10[%] of target force
magnitude[not to scale]
3.5 3
2.5 2
1.5 1
0.5 0 The initial contact point 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
-0.5 0
time [s] Fig. 3. The clamping force, the angular velocity and the current in applying force mode
Fig. 3 shows the motor current, the angular velocity, and the clamping force in applying force mode. The results show that the three signals have correlations. The angular velocity decreases and the motor current increases when the clamping force increases. Thus we will derive the relationship between the angular velocity and the clamping force from now on. The dynamics of the DC motor may be represented by (1) - (3) [8].
Tm = K m im (t )
B. Detecting the contact point The clearance between the brake pads and the brake disk is different whenever the brake is released. To overcome this problem, the conventional method requires clearance management [4]. Three experimental results for different clearances are plotted in Fig. 2.
120
where R is the resistance, L is the inductance, and Kb is the back emf constant of the motor. The torque developed by the motor is Tm and the motor torque constant is Km. The motor current is im(t), the applied input voltage is va(t) and the back emf is vemf(t). The torque relation in EPB system is given by Tm = Ta + Ti + T f (4)
100
80
force [kgf]
where Ta is the applying torque of EPB system, Ti is the inertia torque, and Tf is the friction torque. Then Ta is proportional to the clamping force, f(eff), which is defined by Ta = t f ( eff ) (5)
60
40
20
where eff (t ) := (t ) (tmax ). t is a constant determined by the screw gain and gear ratio. Details on the modeling are referred to [9]. The clamping force, f(eff), is a function of the angular displacement from the contact point. This function will be derived in Section . C. And Ti is proportional to the
0 0
20
40
60
80
100
120
angular displacement [rev] Fig. 2. Clamping force versus angular displacement for different clearances.
1589
SaB1.4