Electrified Powertrain Creep Control Using Feedforward Torque Observer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional solutions for creep control in electrified powertrains with torque converters or launch clutches require high calibration efforts and do not effectively account for system dynamics, leading to suboptimal performance and closed-loop windup.
Innovation Solution
A feedforward control system utilizing a model reference observer, such as a Kalman filter, to determine optimal input torque for the torque transfer device based on operating parameters, actuator achieved torque, and torque converter characteristics, reducing calibration effort and improving drivability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional heuristic techniques are used to compute optimal load torque, then the control system can be implemented, but high calibration effort is required
Solution Approach 1:
The patent replaces conventional heuristic calibration methods with a model reference observer-based computational approach. The observer uses system dynamics models and measured signals to compute optimal load torque without requiring extensive manual calibration, thereby reducing calibration effort while maintaining control sophistication.
Solution Approach 2:
The model reference observer automatically adapts to system variations by continuously estimating states and parameters based on measured signals and system models. This self-adjusting capability eliminates the need for manual recalibration under different operating conditions, reducing overall calibration effort.
2Reliability
If modeling from wheel to actuator is used to compute optimal load torque, then the control system can be implemented, but it does not account for variations that cause closed-loop windup
Solution Approach 1:
The model reference observer incorporates feedback from measured system signals (motor speed, torque, transmission input speed) to continuously update state estimates and adapt to variations. This feedback mechanism allows the system to account for real-time variations without causing closed-loop windup, improving both reliability and adaptability.
Solution Approach 2:
The patent employs a dynamic model reference observer that adapts to changing system conditions by continuously estimating states based on current measurements and system dynamics. This dynamic adaptation capability allows the system to account for variations in operating conditions while maintaining stability, preventing closed-loop windup.
3Ease of operation
If the feedforward control system accurately regulates input speed and torque, then creep drivability is improved, but the system complexity increases
Solution Approach 1:
The feedforward control component computes the optimal load torque in advance based on the desired transmission input speed and system models. This preliminary action provides the ideal torque command before disturbances occur, improving creep drivability by proactively regulating speed and torque rather than merely reacting to deviations.
Solution Approach 2:
The model reference observer acts as an intermediary between the feedforward torque command and the actual actuator control. It processes the desired speed and torque inputs along with measured signals to generate an accurate estimate of the optimal load torque, mediating between theoretical commands and practical implementation to improve drivability while managing complexity.
Data Source
AI summary
A feedforward control method for an electrified powertrain including a torque transfer device arranged between an electric motor and a transmission includes monitoring a set of operating parameters of the electrified powertrain, determining a desired input speed for the torque transfer device based on the set of operating parameters, determining a desired input torque for the torque transfer device based on a characteristics model or map of the torque transfer device and the desired input speed, performing an observer-based determination of a final feedforward torque for the torque transfer device based on the desired input speed, the desired input torque, a filtered actuator achieved torque for the electric motor, and minimum and maximum torque limits for the transmission, and controlling the electric motor based on the final feedforward torque for the torque transfer device.