Abstract: The method comprises the steps of inputting a first set of data into both a physical system and into an SVR model. The method includes collecting a second set of system data from the physical system and stacking a plurality of SVR models to form an output prediction without feeding back the model output. Another aspect of the invention includes a method of hybrid modeling having delayed output measurement. This method includes the steps of inputting a first set of data into both a physical system and into an SVR model and collecting a second set of system data from the physical system. A modeling error is injected into both the second set of system data from the physical system and a third set of system data from the SVR model wherein the modeling error thereby leads to substantially improved model output.

Abstract: Examples of the present invention include air charge estimation models using linear programming support vector regression (LP-SVR). Air model parameters may be updated during operation of a vehicle, allowing the air model performance to be improved in the presence of part-to-part variation and part aging.

Abstract: The method comprises the steps of inputting a first set of data into both a physical system and into an SVR model. The method includes collecting a second set of system data from the physical system and stacking a plurality of SVR models to form an output prediction without feeding back the model output. Another aspect of the invention includes a method of hybrid modeling having delayed output measurement. This method includes the steps of inputting a first set of data into both a physical system and into an SVR model and collecting a second set of system data from the physical system. A modeling error is injected into both the second set of system data from the physical system and a third set of system data from the SVR model wherein the modeling error thereby leads to substantially improved model output.

Abstract: Example methods of modeling a nonlinear dynamical system such as a vehicle engine include providing a model using linear programming support vector regression (LP-SVR) having an asymmetric wavelet kernel, such as derived from a raised-cosine wavelet function. The model may be trained to determine parallel model parameters while in a series-parallel configuration, and operated in the parallel configuration allowing improved and more flexible model performance. An improved engine control unit may use an LP-SVR with an asymmetric wavelet kernel.

Abstract: Example methods of modeling a nonlinear dynamical system such as a vehicle engine include providing a model using linear programming support vector regression (LP-SVR) having an asymmetric wavelet kernel, such as derived from a raised-cosine wavelet function. The model may be trained to determine parallel model parameters while in a series-parallel configuration, and operated in the parallel configuration allowing improved and more flexible model performance. An improved engine control unit may use an LP-SVR with an asymmetric wavelet kernel.

Abstract: Examples of the present invention include air charge estimation models using linear programming support vector regression (LP-SVR). Air model parameters may be updated during operation of a vehicle, allowing the air model performance to be improved in the presence of part-to-part variation and part aging.

Abstract: A support vector machine with wavelet kernel was developed for accurate modeling of nonlinear systems. A method of providing an optimized model of a nonlinear system includes using a support vector machine (SVM) having a wavelet kernel, where support vectors include a family of multidimensional wavelets. Training the SVM allows optimization of the number of support vectors, the weights of the support vectors, and the translation factors of the support vectors. Use of a novel linear programming approach reduces computational demands required for training, allowing optimized support vectors to give an optimized model of the nonlinear system. Further, on-line retraining is possible, so that the model can be adapted to changing conditions.

Abstract: A support vector machine with wavelet kernel was developed for accurate modeling of nonlinear systems. A method of providing an optimized model of a nonlinear system includes using a support vector machine (SVM) having a wavelet kernel, where support vectors include a family of multidimensional wavelets. Training the SVM allows optimization of the number of support vectors, the weights of the support vectors, and the translation factors of the support vectors. Use of a novel linear programming approach reduces computational demands required for training, allowing optimized support vectors to give an optimized model of the nonlinear system. Further, on-line retraining is possible, so that the model can be adapted to changing conditions.

Abstract: A power-on clutch-to-clutch downshift control is initiated by partially releasing an off-going torque trasmitting device to permit an increase in the transmission input speed. At the same time, an on-coming torque transmitting device is filled in preparation for engagement. When the slip across the on-coming device is near zero, the on-coming device pressure is progressively increased while the off-going clutch pressure is controlled to maintain the near-zero on-coming slip speed condition. When the on-coming device develops significant capacity, a tie-up develops between the on-coming and off-going devices, causing a drop in output shaft acceleration. The drop is detected and used to release the off-going device, the shift being complete when the on-coming device is fully engaged. Engine torque controls are employed to ensure a smooth output torque transition at the end of the shift.

Abstract: An event-sequenced control for power-on clutch-to-clutch upshifting of a multiple speed ratio automatic transmission. The average output shaft acceleration during the fill period of the on-coming clutch is determined and used to compute a target acceleration value for the completion of the torque phase. The initial achievement of on-coming clutch torque capacity is detected when the input acceleration momentarily becomes negative or when the output acceleration drops by a first specified percentage of the difference between the initial and target values, whichever comes first. As such point, the reduction of the off-going clutch pressure is intensified but its slippage is limited to a near-zero value. An impending completion of the torque phase is detected when the output acceleration drops by a second specified percentage of the difference between the initial and target values.

Abstract: A driver circuit for the solenoid actuator coils of a direct pressure electronic transmission control. The circuit is remotely configurable for satisfying the requirements of both shifting and nonshifting modes of operation. In shifting modes of operation, a common snubber element is activated to permit fast commutation of the coil current while protecting the switching devices. In nonshifting modes of operation, dedicated free-wheeling diodes are activated to circulate the coil energy at turn-off of the switching devices to facilitate reduction energy consumption and heating.

Abstract: A clutch-to-clutch power-on downshift in a motor vehicle automatic transmission is initiated by reducing the torque capacity of the off-going clutch, permitting the transmission input speed to increase toward its post shift value. When the slippage across the off-going clutch exceeds a reference, the reduction of its torque capacity is suspended, permitting continued increase of the transmission input speed. When the slippage across the on-coming clutch falls below a reference value, the on-coming and off-going clutches are concurrently applied and released to complete the shift. The transmission input speed substantially achieves its post shift value at the time of such apply and release, resulting in reduced shift-related output torque disturbance.

Abstract: A load control system for controlling the field current of an alternating current generator and the idle speed control system for the engine that drives the generator. A voltage regulator senses the output voltage of the generator and controls field current by causing a semiconductor switch connected in series with the field winding to switch on and off in accordance with the magnitude of the sensed voltage. The voltage regulator is controlled to operate in a conventional manner when engine speed is higher than engine idle speed. When engine speed is in an idle speed range the duty cycle or on time of the semiconductor is controlled such that consecutive occurring on times are gradually increased to gradually increase field current when the output voltage of the generator is below a desired regulated value.