Have you ever experienced shift shock in a car with an automatic transmission? Shift quality design is one of the significant challenges in automotive development for a stress-free driving experience. Virtual prototyping has proven to have a great advantage over physical prototyping in respect to cost, speed and efficiency. Modeling the automatic transmission system requires a capability of hydro-mechanical multi-domain modeling. An example of this can be found in Hydraulics Library 4.5.
To study and design automatic transmission systems, mechanical and hydraulic domains should be taken into account. Such multi-domain modeling can quickly be achieved in Modelica, in particular when based on a good choice of libraries.
We demonstrate this with a newly added example in Modelon’s Hydraulics Library. Here the automatic transmission hardware is separated into two subsystems: the mechanical gear set and the hydraulic valve body.
The mechanical gear set is modeled with components like clutches and planetary gears that are available in the Modelica Standard Library (MSL). The hydraulics part is modeled with the Hydraulics Library which includes component models like spool valves, cylinders and hydraulic pistons.
The hydraulic subsystem is connected through force signals with the mechanical subsystem, which enables the control of MSL based clutch models. Thus, the hydraulic valve body enables actuation of the mechanical clutches to change the transmission gear speed ratio based on a controller signal.
The pressure in each valve in the valve body is measured for feedback control purpose. This model is summarized in Figure 1.
An equivalent power train system is set up for shift shock simulation, as shown in Figure 2.
A typical shift shock torque profile is shown in Figure 3. It has a torque hole (undershoot) and an overshoot, which are caused by clutch slippage and inertial dynamics respectively. One of the main challenges for the system designer is to reduce the under- and overshoot, and other oscillatory behavior.
In addition to the physical plant model, two controller set ups are presented as examples in the Hydraulics Library.
An open loop controller delivers simple voltage inputs to control the hydraulic system and change the gear speed ratio from gear 1 to 4 (Figure 4).
The simulation result shows a certain amount of shift shock - see Figure 5.
The alternative closed loop control modulates each valve pressure by using pressure measurement feedback (Figure 6). An advantage of this is that once you get the optimal pressure profile the controller tries to maintain the performance even if the hydraulic specification is changed.
The closed loop example gives much better shift quality, as shown in Figure 7.
Contact us to learn more!
Hiroo Iida is a simulation engineer for chassis performance development at Toyota Motor Corporation. He spent several years to develop trade-off design methodology with virtual optimization workflow. Hiroo is currently specializing in modeling and simulation methodology with Modelica and FMI.
Anand Pitchaikani has been working in model based engineering since 2000. With Modelon, he works on simulation consulting and product development with a focus on fluid systems (Hydraulics and Pneumatics).