System and method for controlling a clutch fill event
View Patent ↗A method optimizes a fill event of an apply chamber of a fluid-actuated clutch, and includes determining input values describing the fill event, and then estimating a fill time using the input values. The method includes filling the apply chamber using the estimated fill time (EFT) or within an allowable range of the EFT. The input values can include a command line pressure, command fill stroke pressure, and an estimated viscosity of the fluid, although other values can be used. The input values are processed through a neural network having an input layer, an optional hidden layer, and an output layer. An assembly includes a fluid-actuated clutch having an apply chamber and a controller operable for estimating the fill time required for filling the apply chamber, and for controlling the fill of the apply chamber within the EFT.
1. A method of controlling a fill event of an apply chamber of a fluid-actuated clutch, the method comprising:
determining a plurality of input values describing the fill event, including a calibrated command line pressure, a calibrated command fill stroke pressure, and a viscosity of a fluid used for filling the apply chamber, wherein the plurality of input values collectively defines a clutch fill signature (CFS);
processing the plurality of input values using a neural network having an input layer, a hidden layer, and an output layer each with one or more nodes to thereby recognize a pattern of the CFS;
estimating a fill time required for filling the apply chamber with the fluid using the pattern that is recognized by the neural network; and
filling the apply chamber with the fluid using the estimated fill time (EFT) to thereby optimize the fill event.
2. The method of claim 1 , wherein filling the apply chamber with the fluid using the EFT includes filling the apply chamber within a predetermined range of the EFT.
3. The method of claim 2 , wherein the predetermined range is approximately 95 to 105 percent of the EFT.
4. The method of claim 1 , wherein determining a plurality of input values includes estimating the viscosity of the fluid using a temperature of the fluid.
5. The method of claim 4 , wherein estimating the viscosity of the fluid using a temperature of the fluid is selected from the group consisting of: calculating the viscosity using the temperature of the fluid and selecting the viscosity from a lookup table which is indexed by the temperature of the fluid.
6. The method of claim 1 , wherein the input layer of the neural network includes a plurality of input neurons, and wherein processing the plurality of input values using a neural network includes: processing each of the plurality of input values through a corresponding one of the plurality of input neurons.
7. The method of claim 6 , wherein the hidden layer of the neural network includes a plurality of hidden neurons that is less than the plurality of input neurons, the method further comprising:
processing an output value from the input layer through each of the plurality of hidden neurons.
8. A method of controlling a fill event of a fluid-actuated clutch, the method comprising:
determining a calibrated command line pressure and a calibrated command fill stroke pressure within the fluid-actuated clutch;
estimating a viscosity of a fluid used for actuating the fluid clutch;
processing the calibrated command line pressure, the command fill stroke pressure, and the estimated viscosity through a neural network, wherein the calibrated command line pressure, calibrated command fill stroke pressure, and viscosity collectively define a clutch fill signature (CFS);
determining an estimated fill time (EFT) required for filling an apply chamber of the fluid-actuated clutch by recognizing a pattern of the CFS via the neural network; and
controlling the filling of the apply chamber using the EFT.
9. The method of claim 8 , further comprising:
determining if EFT corresponds to an observed fill time; and
modifying the neural network when the EFT that is predicted by the neural network does not agree with the observed fill time.
10. The method of claim 8 , wherein estimating a viscosity of a fluid used for actuating the fluid clutch includes measuring a temperature of the fluid and then estimating the viscosity using the measured temperature.
11. A clutch fill system comprising:
a fluid-actuated clutch having a piston disposed within an apply chamber; and
a controller configured for determining a plurality of input values describing a fill event of the fluid-actuated clutch, wherein the plurality of input values collectively defines a clutch fill signature (CFS) and includes a calibrated command line pressure of the fluid-actuated clutch, a command fill stroke pressure of the fluid-actuated clutch, and an estimated viscosity of a fluid used to actuate the clutch;
wherein the controller includes a neural network having an input layer with a plurality of input nodes, a hidden layer having a plurality of hidden nodes that is less than the plurality of input nodes, and an output layer having a single output node, and wherein the controller is configured for using the neural network for estimating a fill time required for filling the apply chamber by recognizing a pattern of the CFS, and is operable for controlling a filling of the apply chamber within the estimated fill time.
12. The system of claim 11 , wherein the hidden layer is a single hidden layer having two hidden nodes.
13. The system of claim 12 , wherein the two hidden nodes each utilize a non-linear transfer function.
14. The method of claim 1 , wherein each of the input layer, the hidden layer, and the output layer includes at least one neuron employing a sigmoidal transfer function.