Abstract: Provided is a processor-implemented method and a processor in a vehicle for estimating the value of a quantity for which a physical sensor is not available for measurement. The method includes: receiving a plurality of measured signals representing values of measurable variables; computing, in real-time, time derivatives of the measured signals; and applying a trained feedforward neural network, in real-time, to estimate values for a plurality of unmeasurable variables, the unmeasurable variables being variables that are unmeasurable in real-time, the feedforward neural network having been trained using test data containing time derivatives of values for the measurable variables and values for the unmeasurable variables; wherein the vehicle uses the estimated values for the unmeasurable variables for vehicle operation.

Abstract: A system or method for determining virtual data of a system, relative to a measurement point having a sensor located nearby, is determined by a controller. The system calculates modeled data at the measurement point, filters the modeled data to determine filtered data, and calculates a differential between the modeled data and the filtered data to determine a compensation term. The system also determines raw-sensed data from the sensor at the measurement point, and combines that raw-sensed data with the compensation data to calculate the virtual data at the measurement point. In some configurations, the modeled data is determined from a physics-based model. Furthermore, filtering the modeled data may include using a low-pass filter, and a time constant for the low-pass filter may be calculated based on operating conditions of the system.

Abstract: Provided is a processor-implemented method and a processor in a vehicle for estimating the value of a quantity for which a physical sensor is not available for measurement. The method includes: receiving a plurality of measured signals representing values of measurable variables; computing, in real-time, time derivatives of the measured signals; and applying a trained feedforward neural network, in real-time, to estimate values for a plurality of unmeasurable variables, the unmeasurable variables being variables that are unmeasurable in real-time, the feedforward neural network having been trained using test data containing time derivatives of values for the measurable variables and values for the unmeasurable variables; wherein the vehicle uses the estimated values for the unmeasurable variables for vehicle operation.

Abstract: A system or method for determining virtual data of a system, relative to a measurement point having a sensor located nearby, is determined by a controller. The system calculates modeled data at the measurement point, filters the modeled data to determine filtered data, and calculates a differential between the modeled data and the filtered data to determine a compensation term. The system also determines raw-sensed data from the sensor at the measurement point, and combines that raw-sensed data with the compensation data to calculate the virtual data at the measurement point. In some configurations, the modeled data is determined from a physics-based model. Furthermore, filtering the modeled data may include using a low-pass filter, and a time constant for the low-pass filter may be calculated based on operating conditions of the system.

Abstract: An exhaust gas after-treatment system for an internal combustion engine that includes a selective catalyst reduction on filter (SCRF) exhaust gas after-treatment device in communication with exhaust gases from the internal combustion engine and having a treated exhaust gas output. An oxides of nitrogen (NOx) sensor is coupled to the treated exhaust gases and has a NOx sensor output signal that is NOx and ammonia (NH3) cross-sensitive. A closed loop observer (CLO) is operatively coupled to receive the NOx sensor output signal and provides a NOx concentration signal to an electronic control unit that is operatively associated with the exhaust gas after-treatment system and the internal combustion engine.

Abstract: An engine assembly includes an engine and a plurality of actuators. The plurality of actuators includes a first turbine serially connected to a second turbine, the first turbine being a relatively high pressure turbine and the second turbine being a relatively low pressure turbine. A controller is configured to transmit respective command signals to the plurality of actuators. The controller is programmed to obtain respective transfer rates for the plurality of actuators based at least partially on an inversion model. The controller is programmed to control an output of the engine by commanding the plurality of actuators to respective operating parameters via the respective command signals. Prior to obtaining the respective transfer rates, the controller is programmed to determine a respective plurality of desired values and respective correction factors for the plurality of actuators.

Abstract: A method and apparatus is disclosed to control the operation of an air charging system of an internal combustion engine. A plurality of output parameters of the air charging system are monitored. An error is calculated between the monitored output parameters and a target value thereof. The calculated errors are applied to a linear controller that yields a virtual input used to calculate a plurality of input parameters for the air charging system. The input parameters is used to determine the position of a corresponding actuator of the air charging system for operating the actuators according to the determined position thereof. The inputs parameters are calculated with a non-linear mathematical model of the air charging system configured such that the virtual inputs are in a linear relation with only one of the output parameters and vice versa.

Abstract: An engine assembly includes an engine and a plurality of actuators. The plurality of actuators includes a first turbine serially connected to a second turbine, the first turbine being a relatively high pressure turbine and the second turbine being a relatively low pressure turbine. A controller is configured to transmit respective command signals to the plurality of actuators. The controller is programmed to obtain respective transfer rates for the plurality of actuators based at least partially on an inversion model. The controller is programmed to control an output of the engine by commanding the plurality of actuators to respective operating parameters via the respective command signals. Prior to obtaining the respective transfer rates, the controller is programmed to determine a respective plurality of desired values and respective correction factors for the plurality of actuators.

Abstract: An automotive vehicle includes an internal combustion engine and an exhaust system. The exhaust treatment system includes a dosing system that injects NH3 into an exhaust gas stream generated by the engine. An SCR device stores an amount of the NH3 and converts NOx into diatomic nitrogen (N2) and water (H2O) based on the stored amount of the NH3. The vehicle further includes an SCR status estimator device and a controller. The SCR status estimator device determines an NH3 coverage ratio (R), which indicates a stored amount of NH3 with respect to a maximum NH3 storage capacity of the SCR device. The controller determines a target NOx reduction efficiency (?NOx) of the SCR device, and an NH3 coverage ratio set point (Rsp) based on the ?NOx. The controller also generates an NH3 control signal (u) that controls the dosing system based on a comparison between the R and the Rsp.

Abstract: An automotive vehicle includes an internal combustion engine and an exhaust system. The exhaust treatment system includes a dosing system that injects NH3 into an exhaust gas stream generated by the engine. An SCR device stores an amount of the NH3 and converts NOx into diatomic nitrogen (N2) and water (H2O) based on the stored amount of the NH3. The vehicle further includes an SCR status estimator device and a controller. The SCR status estimator device determines an NH3 coverage ratio (R), which indicates a stored amount of NH3 with respect to a maximum NH3 storage capacity of the SCR device. The controller determines a target NOx reduction efficiency (?NOx) of the SCR device, and an NH3 coverage ratio set point (Rsp) based on the ?NOx. The controller also generates an NH3 control signal (u) that controls the dosing system based on a comparison between the R and the Rsp.

Abstract: A method and apparatus is disclosed to control the operation of an air charging system of an internal combustion engine. A plurality of output parameters of the air charging system are monitored. An error is calculated between the monitored output parameters and a target value thereof. The calculated errors are applied to a linear controller that yields a virtual input used to calculate a plurality of input parameters for the air charging system. The input parameters is used to determine the position of a corresponding actuator of the air charging system for operating the actuators according to the determined position thereof. The inputs parameters are calculated with a non-linear mathematical model of the air charging system configured such that the virtual inputs are in a linear relation with only one of the output parameters and vice versa.