Electricity: electrical systems and devices – Control circuits for electromagnetic devices – For relays or solenoids
Reexamination Certificate
2001-11-29
2003-12-16
Tso, Edward H. (Department: 2838)
Electricity: electrical systems and devices
Control circuits for electromagnetic devices
For relays or solenoids
Reexamination Certificate
active
06665165
ABSTRACT:
BACKGROUND AND SUMMARY OF THE INVENTION
This application claims the priority of German Application No. 100 59 285.6, filed Nov. 29, 2000, the disclosure of which is expressly incorporated by reference herein.
The invention relates to a control method for the electrical actuator of a variable valve train, the actuator being embodied in the form of a servo motor that drives a shaft using a-worm gear. A sensor detects a position value (measured position value) of the shaft, a control difference is generated in a feedback branch with the use of the measured position value, and the electrical actuator is activated as a function of the control difference.
In internal-combustion engines, one of the objectives is to alter the height of the valve stroke, particularly of the gas-intake valves, as a function of the operating point, for the charge-exchange process. If this is accomplished, the use of a throttle valve can be extensively eliminated, which would lead to reduced fuel consumption, because the disadvantageous effects of throttle valves include high consumption.
A known embodiment of a so-called completely variable valve train provides an eccentric shaft for adjusting the height of the valve stroke. The shaft is acted upon by a servo motor, and a worm gear is interposed between the two. As dictated by structural conditions, the worm gear possesses a so-called clearance, in which the eccentric shaft can rotate slightly without the actuator being activated. The effect of an external mechanical interference, as induced, for example, by the camshaft influence, may cause the shaft to move within the clearance of the worm gear. This in turn causes a position sensor that is disposed in the region of the main drive pinion of the worm gear pair to indicate a change in position, although the actuator has not been activated. In particular, the rapid change in the position signal due to the change in the contact of the flanks in the gear excites the control system, and, consequently, leads to an activation of the actuator. This results particularly in an increased current consumption by the system.
It is the object of the present invention to suppress interferences that occur in the control system when the gear clearance is passed through, without significantly limiting the control dynamics or control precision at the same time.
This object is accomplished by a control method for the electrical actuator of a variable valve train, the actuator being embodied in the form of a worm gear, in which a sensor detects a position value (measured position value) of an eccentric shaft, a control difference is generated in a feedback branch with the use of the measured position value, and the electrical actuator is activated as a function of the control difference. An observer value that is associated with the position value is formed from a pulse duty factor of a pulse-width modulator. The observer value is compared to a comparison value formed from the measured position value. The detected measured position value is not accepted if the deviation of the observer value from the comparison value exceeds a defined limit.
The electrical actuator is activated, preferably via pulse-width modulation, as a function of a control difference ascertained with a measured position value. To generate an associated pulse duty factor, a control difference is assessed in a controller. The controller itself is preferably embodied to operate digitally. The control difference comprises, for example, a desired value that is preset by the engine control unit, and an acknowledged position value for the eccentric shaft. Because the measured position value may be affected by interfering variables, it is necessary to perform a relevant plausibility check. For this purpose, an observer value that is associated with the position value is formed from the pulse duty factor of the pulse-width modulation. This observer value is compared to a comparison value obtained from the actual measured position value. If the difference between the observer value and the comparison value exceeds a defined limit or amount, the measured position value itself is not accepted.
The measured position value is thus checked in terms of its plausibility with available state information.
Because the pulse duty factor is approximately proportional to the current flowing through the electrical actuator when the actuator is at still stand, and is therefore proportional to the maximum torque that the actuator can deliver, the values associated with the second derivative of the position value are preferably compared to one another.
The torque generated by the actuator effects an acceleration
the system. A theoretical acceleration
determined as an observer value from the pulse duty factor is preferably compared to an actually determined acceleration
of the measured position value. This type of comparison of the angular accelerations allows the measured position values to be assessed, then filtered such that all ascertained sensor values that correspond to an angular acceleration located outside of the established limits are rejected as implausible.
If the sampling time of the controller T
A
is longer than that of the position sensor, it also appears useful to insert an actual-value filter. The actual-value filter filters out apparently implausible measured position values, prior to an averaging device, particularly for suppressing signal noise. If a differentiating transmission element is used in the feedback branch of the control method, the actual-value filter should be disposed upstream of this transmission element, of course, other transmission elements, such as proportional elements, can also be disposed, in series or in parallel, in the feedback branch. In many cases, however, filtering can be limited to the feedback branch with the differentiating transmission element, because the particularly rapid, temporary fluctuations that occur when the worm-gear clearance is passed through have a negative impact on this member. The dynamics of the control method, however, stipulate that the differentiating member be disposed in the feedback branch: high actuating dynamics can be attained with this arrangement.
The presetting of a desired value for forming the control difference and actuating the controller preferably originates in digital electronics of an engine that dictate the valve-stroke height by way of an adjustment of the eccentric shaft, depending on the operating point.
The actual position value is notably detected with a rotation sensor disposed on the eccentric shaft. All information can be detected and transmitted in time-discrete and digital form.
A nonlinear actual-value filter can preferably be used in the comparison of the observer and comparison values. Maximum and minimum values for the difference between the observation and comparison values can be defined in this nonlinear actual-value filter. A measured position value is accepted if the aforementioned difference lies in a range between the maximum and minimum values. As described above, the maximum value can be determined as the observer value from the pulse duty factor. A brake acceleration that occurs when the servomotor runs down without current can be selected as the minimum value.
If a pulse width ratio/pulse duty factor of zero is present, the positive and negative absolute values of this minimum value can be selected as the two limit values of the nonlinear actual-value filter.
If no accepted measured position values are present during a position-control cycle, the old state variables of the filter are used for the following cycle, especially if the nonlinear filter was disposed upstream for calculating them.
A so-called D
2
member is used to attain a second derivative of the measured position value. This member can effect a differentiation of detected values by forming finite differences. The acceleration is also possible through approximation with a cubic function (spline) using the last three valid position values.
REFERENCES:
patent: 5778840 (1998-07-01), Murata et al.
p
Tibbits Pia
Tso Edward H.
LandOfFree
Control method for the actuator of a variable valve train does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Control method for the actuator of a variable valve train, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Control method for the actuator of a variable valve train will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3165550