The present invention relates to a control system and method for a vehicle transmission.
Knowledge of various torque values within an automatic transmission is useful for many aspects of powertrain control. For example, knowing the input and output shaft torque of a transmission gear box is particularly useful for controlling the transmission to provide consistent and robust transmission shifts. Using these torque signals along with input and output shaft speeds facilitates accurate estimation of individual clutch torques during a shift. One such control system and method are described in United States patent application publication number US20100318269, filed on 23 Aug. 2010, which is hereby incorporated herein by reference.
For cost or other considerations it is not always possible to directly measure each torque within the transmission, and therefore, control system specialists rely on various transmission models to estimate torque values that are not directly measured. In one such model, a quasi-static nonlinear model of a torque converter within the transmission is used to provide torque estimates. Such models rely on multiple nonlinear functions, and are therefore not readily susceptible to adaptive modeling techniques where the values of the estimates are periodically improved. This is because the estimated value, when it is found to deviate from a verifiable value, may require adjustment of one or more of the nonlinear functions in the model, but it is not possible to know which function or functions requires adjustment. Therefore, a need exists for a system and method for a vehicle transmission that provides accurate and adaptable estimates of transmission torque.
At least some embodiments of the invention include a method for controlling a vehicle transmission using a torque estimate. At least one transmission component is automatically controlled during a shift event based on a first torque estimate defined in terms of one nonlinear function of a transmission parameter for a particular value of the transmission parameter. The first torque estimate is modified outside of a shift event based on a measured torque of the transmission at the particular value of the transmission parameter.
At least some embodiments of the invention include a method for controlling a vehicle transmission using a torque estimate. At least one transmission component is automatically controlled based on a first torque estimate for a particular value of a transmission parameter. The first torque estimate is based on one nonlinear function of the transmission parameter which combines a plurality of nonlinear functions of the transmission parameter.
At least some embodiments of the invention include a control system for a vehicle transmission. The control system includes a controller configured to: output a first torque estimate defined in terms of one nonlinear function of a transmission parameter for a particular value of the transmission parameter, receive a measured torque of the transmission at the particular value of the transmission parameter, and output a modified torque estimate for the particular value of the transmission parameter based on the measured torque.
As required, detailed embodiments of the present invention are disclosed herein; however, it is to be understood that the disclosed embodiments are merely exemplary of the invention that may be embodied in various and alternative forms. The figures are not necessarily to scale; some features may be exaggerated or minimized to show details of particular components. Therefore, specific structural and functional details disclosed herein are not to be interpreted as limiting, but merely as a representative basis for teaching one skilled in the art to variously employ the present invention.
Each of the boxes 22, 24, 26, 28 shown in
The torques are again balanced at 24, where the turbine torque (τt) from the torque converter 14 is balanced against an equal torque from the gearbox 16 (τs/Rg). The torque (τs/Rg) is the torque of the gearbox output shaft (τs) modified by the gear ratio (Rg) of the gearbox 16. The turbine speed (ωt) is shown as an input to the gearbox 16, and shown as an output is an output shaft speed (ωt/Rg), which is the turbine speed (ωt) modified by the gear ratio (Rg). Another torque balance occurs at 26 with the output shaft torque (τs) being balanced against itself. Unlike the dynamic equations used in 22, 24, 28, which are generally based on the equation F=ma, the equations in 26 are generally based on a spring-force relationship, F=kx. As shown at 26, the speed of the output shaft (ωt/Rg) is equal to the input speed of the FDR 18 (ωw/Rd). The input speed of the FDR 18 is the wheel speed (ωw), which is a speed on the output side of the FDR 18, modified by the ratio of the differential (Rd). Finally, the torques are balanced at 28 where a road load torque (τr) is balanced against the output torque (τsRd) of the FDR 18, which is the output shaft torque (τs) modified by the ratio of the differential (Rd).
In
As discussed above, it is often desirable for powertrain control specialists to know various torques within a powertrain, and in particular the input and output torques of the gearbox of the transmission. It was also noted that in many cases, direct measurement of both of these torques is, for various reasons, not possible. If there is a direct measurement of an output torque, for example, the torque (τs) of the gearbox 16, then the input torque (τt) can be easily calculated if the instant of interest is outside of a shift event for the transmission. This situation is illustrated by
Embodiments of the invention include a method for estimating a torque, such as an input torque to a gearbox, such as the gearbox 16 of the transmission 12. One way to do this is to provide a first torque estimate that is defined in terms of a single nonlinear function of a transmission parameter, for example, a speed ratio, at a particular value of that transmission parameter. Equation 1 shown below provides such an estimate for a torque, such as the turbine torque (τt) described above.
TT1=ωe2TCF(ωt/ωe) Eq. 1
In equation 1, the torque (TT1) is the input torque to the gearbox 16 (it is also the turbine torque (τt) as shown in
One such equation using multiple nonlinear functions for a torque estimate is illustrated below as equation 2.
In this equation, the torque estimate (TT2) is defined by two separate nonlinear functions of the same transmission parameter (ωt/ωe) as used in equation 1. The first function, K(ωt/ωe), is illustrated in graph 32 in
Similarly, the second function used in equation 2, TR(ωt/ωe), is illustrated in graph 34 in
As noted above, using a torque estimate that is defined in terms of a single nonlinear function, such as shown in the equation 1, provides advantages over the torque estimate shown in equation 2. For example, if the torque of the input shaft of the gearbox 16 is estimated according to equation 2, and the estimation takes place outside of a shift event of the transmission 12, the estimated torque (TT2) can be compared directly to a measured torque of the gearbox output shaft adjusted by the gear ratio (τs/Rg). If it is determined that the estimated torque (TT2) is not as accurate as desired, it could be a candidate for an adaptive learning process by which the estimate is adjusted based on a measured value. The problem is that with two nonlinear functions in the definition of the torque (TT2), it is not possible to determine which of the two need to be adjusted so as to adapt the torque estimate equation to a more accurate model.
In contrast, using embodiments of the present invention, such as a method using the torque estimate (TT1) shown in equation 1, allows the torque estimate to be adapted to measured torque values obtained outside of transmission shift events where there is a known relationship between the measured output shaft torque and the input shaft torque. Thus, in at least some embodiments of the present invention, a first torque estimate, such as the torque (TT1), is defined in terms of a single nonlinear function of a transmission parameter—see equation 1 with the single, nonlinear function being TCF(ωt/ωe), and the transmission parameter being the ratio of speeds (ωt/ωe) across the torque converter 14. The first torque estimate may be calculated by the controller 20, and thus “provided” to itself; conversely, it may be calculated elsewhere in a control system and a signal sent to the controller 20, which may use this signal along with other information related to the powertrain 10 to control various elements of the powertrain, such as the transmission 12. The first torque estimate may be used, for example, to automatically control at least one transmission component, such as a clutch or clutches, during a shift event. To provide an actual value, the first torque estimate (τT1) is provided to the controller 20 at a particular value of the transmission parameter, which in this case is a value of the speed ratio.
Next, the torque of the output shaft (τs) is measured at the particular value of the transmission parameter—i.e., at the same value of the speed ratio. As described in detail above, the output shaft torque (τs) is easily related to the input shaft torque (τt) when the measurement and estimate are taken outside of a shift event. Thus, if the torque estimate (τT1) is subject to adaptive learning and needs to be adjusted, it is then modified based on the measured torque. Going forward, when the torque is again estimated using equation 1, a modified torque estimate is provided by or to the controller 20 where the value can again be compared to a measured torque if available. Conversely, if the modified torque estimate is provided by or to the controller 20 during a shift event, no comparison is possible, but the modified and presumably more accurate torque estimate is now used by the controller 20 to control the powertrain 10—e.g., to control the torque of the OFG and ONC clutches in the transmission 12.
With regard to the controller 20, a method might be summarized as follows. The controller 20 either calculates or receives from another controller the first torque estimate and “outputs” this value either to itself or as part of a signal to control the powertrain 10. Then, it receives the measured torque, for example, from a torque sensor on the output shaft of the gearbox 16. The controller 20 then determines a “measured” torque for the input shaft of the gearbox 16 based on the measured output shaft torque and the gear ratio of the gearbox 16. If, after comparing the estimated torque to the measured torque, the controller 20 determines that an adaptive learning process should be applied, it modifies the torque estimate, for example, by modifying the function TCF(ωt/ωe) to account for the difference between the first torque estimate and the measured torque. Finally, the controller 20 “outputs” a modified torque estimate for the particular speed ratio when the torque estimate for that speed ratio is required. As discussed above, this adaptive learning is only relevant outside of transmission shift events, because the measured torque of the output shaft of the gearbox 16 cannot be related to the torque of the input shaft when a shift event is occurring—see
In order to define a torque estimate, such as the torque estimate (TT1) shown in the equation 1, in terms of only one nonlinear function, embodiments of the invention use certain other nonlinear functions from a different torque estimate and combine them to define the single nonlinear function, such as the function TCF(ωt/ωe) described above. For example, the two nonlinear functions described above from equation 2, can be combined to create a function TCF(ωt/ωe). When the speed term (ωt2) is removed from equation 1, what remains is a combination of the two nonlinear functions that can be combined to create the single nonlinear function TCF(ωt/ωe). Embodiments of the invention may combine two nonlinear functions such as these to create a single nonlinear function, as shown below in equation 4.
In summary, a preliminary torque estimate (TT2) from equation 2 is defined in terms of a plurality of certain nonlinear functions of a transmission parameter: K(ωt/ωe) and TR(ωt/ωe). The two nonlinear functions are combined into one nonlinear function, TCF(ωt/ωe), and this function is used to define a first torque estimate shown in equation 1 above, the value for which can be provided by or to the controller 20 for use in controlling the powertrain 10.
In order to use the one nonlinear function TCF(ωt/ωe) in the torque estimate (TT1) shown in equation 1, it may first be desirable to generate a plurality of values for the function at certain chosen values of the speed ratio (ωt/ωe). Generating these values will allow them to be input into a controller, such as the controller 20 where they could be used as part of a lookup table, or for graphical interpretation they may be plotted in a plot similar to the graphs 32, 34 shown respectively in
Another way to generate values of the function TCF(ωt/ωe) so that it can be used to estimate torque as described above is to use equation 1 along with torque values measured outside of transmission shift events for known values of the speed ratio (ωt/ωe) and engine speed (ωe). In this case, the measured torque values are substituted into (TT1), which is now a known rather than an unknown, and with the speed ratio (ωt/ωe) and engine speed (ωe) values known, the only unknown is the value of the function TCF(ωt/ωe) itself, and thus it can be easily solved for by manipulating equation 1. Once the values of the function TCF(ωt/ωe) are known over a range of the speed ratio (ωt/ωe) and engine speed (ωe), equation 1 can be used to provide the torque estimate (TT1) for the predetermined value of the speed ratio (ωt/ωe). As above, if the predetermined value of the speed ratio (ωt/ωe) corresponds to one of the values for the speed ratio (ωt/ωe) used to calculate the value of the function TCF(ωt/ωe), this value of the function TCF(ωt/ωe) can be used directly in equation 1. If, however, the predetermined value of the speed ratio (ωt/ωe) is different from one of the values previously used, interpolation can be used to find the value of the function TCF(ωt/ωe), which is then used in equation 1 to provide the first torque estimate.
As described in detail above, this torque estimate can be compared to measured torque values and an adaptive learning process can be implemented to further refine the function TCF(ωt/ωe) as needed. Shown in
While exemplary embodiments are described above, it is not intended that these embodiments describe all possible forms of the invention. Rather, the words used in the specification are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the invention. Additionally, the features of various implementing embodiments may be combined to form further embodiments of the invention.
Number | Name | Date | Kind |
---|---|---|---|
6052640 | Wu | Apr 2000 | A |
6427109 | Doering et al. | Jul 2002 | B1 |
6594573 | Rossmann et al. | Jul 2003 | B1 |
7340336 | Yasui et al. | Mar 2008 | B2 |
7478572 | Maten et al. | Jan 2009 | B2 |
7580786 | Yanakiev | Aug 2009 | B2 |
7603219 | Joshi et al. | Oct 2009 | B2 |
7752021 | Holtz et al. | Jul 2010 | B2 |
20060245929 | McDonald et al. | Nov 2006 | A1 |
20070294017 | Joshi et al. | Dec 2007 | A1 |
20090181824 | Baeuerle | Jul 2009 | A1 |
20100063696 | Runde et al. | Mar 2010 | A1 |
20100318269 | Yanakiev et al. | Dec 2010 | A1 |
Number | Date | Country |
---|---|---|
2013029140 | Feb 2013 | JP |
Number | Date | Country | |
---|---|---|---|
20130345022 A1 | Dec 2013 | US |