Boundary adaptation scheme for spool valve pressure control
View Patent ↗A method for adapting a closing boundary for a proportional spool valve includes generating or detecting a pressure command signal. Next a pressure error is obtained by subtracting an actual wheel brake pressure from its pressure command. Then a modified pressure error is calculated a modified pressure error is calculated in such a way that steady state pressure errors resulting from mismatched feedforward term, control deadzone and other factors are subtracted from measured pressure error to leave the only error to be that due to boundary deviation. An estimator is then used to estimate an apply boundary deviation or a release boundary deviation of the spool valve, as appropriate. A boundary table is updated using the resultant boundary deviation estimate.
1. A method for adapting a closing boundary for a proportional spool valve comprising: (a) generating a pressure command signal; (b) obtaining a pressure error by subtracting an actual brake pressure from its pressure command; (c) calculating a modified pressure error; (d) using an estimator to estimate an apply boundary deviation or a release boundary deviation of the spool valve, as appropriate; and (e) updating a boundary table using the resultant boundary deviation estimate.
2. The method of claim 1 , wherein the boundary table is one of two tables including an apply boundary table and a release boundary table, the method further including after step (e), the step (f) repeating steps (a) through (e) while generating different pressure command signals for each step (a) so as to obtain boundary deviation estimates to update both the apply boundary table and the release boundary table for a plurality of data points on the apply and release portions, respectively, of an operating curve of the spool valve.
3. The method defined in claim 2 wherein the modified pressure error for an apply boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a negative boundary deviation estimate.
4. The method defined in claim 3 wherein the modified pressure error for a release boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a positive boundary deviation estimate.
5. The method defined in claim 2 wherein, in step (d), a gain scheduler is used for the estimator.
6. The method of claim 1 further including, following step (e), the following steps: (f) repeating steps (a) through (e) while generating different pressure command signals in step (a) to obtain boundary deviations associated with several points on a portion of the operating curve; (g) extrapolating boundary deviations associated with points on the rest of the operating curve based on the data obtained in steps (a) through (f); and (h) updating apply and release boundary tables using the boundary deviations obtained in steps (e) and (f) and extrapolated in step (g).
7. The method defined in claim 6 wherein the modified pressure error for an apply boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a negative boundary deviation estimate.
8. The method defined in claim 6 wherein the modified pressure error for a release boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a positive boundary deviation estimate.
9. The method defined in claim 6 wherein, in step (d), a gain scheduler is used for the estimator.
10. The method defined in claim 1 wherein a first estimator and a second estimator are used to estimate boundary deviations of apply and release functions, respectively, of a valve.
11. The method defined in claim 10 wherein the sign of the pressure command derivative over time is used to determine whether the first estimator or the second estimator is updated.
12. The method of claim 1 Further including following step (e) the following two steps: (f) mapping application specific generic data onto the boundary deviation determined in step (e); and (g) updating apply and release boundary tables using the mapped-on data.
13. The method defined in claim 1 wherein, in step (d), a gain scheduler is used for the estimator.
14. The method defined in claim 13 wherein the modified pressure error for an apply boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a negative boundary deviation estimate.
15. The method defined in claim 13 wherein the modified pressure error for a release boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a positive boundary deviation estimate.
16. A method for adapting a closing boundary for a proportional spool valve comprising: (a) detecting a pressure command signal; (b) obtaining a pressure command derivative over time; (c) using the sign of the pressure command derivative over time to determine which closing boundary should be updated; (d) obtaining a pressure error by subtracting an actual brake pressure from its pressure command; (e) calculating a modified pressure error; (f) implementing the modified pressure error in the estimation; (g) implementing an estimator to estimate a boundary deviation of the valve; and (h) updating a boundary table using the resultant boundary deviation estimate.
17. The method defined in claim 16 wherein a first estimator and a second estimator are used to estimate boundary deviations of apply and release functions, respectively, of a valve.
18. The method defined in claim 16 wherein a gain scheduler is used for the estimator and wherein the modified pressure error for an apply boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a negative boundary deviation estimate.
19. The method defined in claim 16 wherein a gain scheduler is used for the estimator and wherein the modified pressure error for a release boundary of the valve has alternating signs that result from an over-compensated deadband and wherein the modified pressure errors with alternating signs give a positive boundary deviation estimate.
20. The method defined in claim 17 wherein the sign of the pressure command derivative over time is used to determine whether the first or second estimator is updated.
21. The method defined in claim 16 wherein the estimators are updated only during gentle braking maneuvers and wherein gentle braking is braking where the absolute value of the pressure command derivative over time is less than a pre-specified value that further is less than the maximum allowable rate of change for wheel brake pressure for the valve.
22. The method defined in claim 21 wherein the pre-specified value is fixed across an entire operating pressure range for a braking system.
23. The method defined in claim 21 wherein the pre-specified value varies with pressure conditions.
24. The method defined in claim 23 wherein the pressure conditions comprise wheel brake pressure and pressure differential across the valve.
25. The method defined in claim 16 wherein a feedforward mismatch and a control deadzone are excluded from the pressure error.