Motor Speed Feedback Control for EV Resonance Vibration Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vibration suppression control systems for electric vehicles face challenges in accurately suppressing vibrations due to resonance, particularly due to high-order mathematical models, model mismatches with varying road conditions, and the effects of backlash and delay times in speed measurement and electric current control.
Innovation Solution
A vibration suppression control device that approximates the vehicle system model to a second-order transfer function, uses a disturbance torque observer to estimate disturbance torques, and implements predictive compensation for dead times, allowing for improved robustness and effective suppression of vibrations and backlash-induced hunting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a high-order vehicle system model is used for vibration suppression control, then the accuracy of vibration suppression is improved, but the complexity of the control system and difficulty of design increase
Solution Approach 1:
The high-order vehicle system model is segmented into a simplified second-order model that captures the essential torsional vibration characteristics. By focusing on the dominant resonance mode between motor and tire inertias connected through the elastic driveline, the complex multi-degree-of-freedom system is reduced to a manageable two-degree-of-freedom representation, making control design feasible while maintaining suppression effectiveness
Solution Approach 2:
The essential vibration suppression functionality is extracted from the complex high-order model by identifying and isolating the critical resonance characteristics. The second-order model extracts only the necessary dynamic parameters (motor inertia, tire inertia, elastic shaft stiffness) required for effective vibration suppression, discarding less critical system complexities
2Ease of manufacture
If a fixed vehicle system model is used, then the control design is simplified, but the robustness against model mismatches and varying road conditions deteriorates
Solution Approach 1:
The control system incorporates adaptive elements that allow the controller to dynamically adjust to varying road conditions and model mismatches. By using feedback from actual system behavior and adjusting control parameters accordingly, the system maintains robustness without requiring a perfectly accurate fixed model for every possible operating condition
Solution Approach 2:
The control approach uses parameter adaptation to handle model uncertainties. By adjusting control parameters based on observed system response and using a second-order model with key parameters (inertias and stiffness), the system achieves both design simplicity and robustness against varying conditions
3Device complexity
If speed measurement delay is not compensated, then the control system is simpler, but the vibration suppression performance deteriorates due to time mismatch
Solution Approach 1:
The system applies preliminary compensation for speed measurement delays by predicting the required compensation torque in advance. By calculating the compensation based on the delayed measurement and adjusting for the known delay characteristics, the control action is effectively advanced to compensate for the measurement lag, maintaining suppression performance without requiring complex real-time prediction algorithms
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided is a vibration suppression control device for a vehicle system, the device using motor speed information to suppress vibration due to resonance, and thereby improve vehicle riding comfort. The present invention is provided with a feedback control configuration in which: included is an approximation model in which a transmission characteristics model for torque input to a vehicle and a motor rotational speed is approximated to a transfer function that is the product of an integral term for a motor moment of inertia and a quadratic expression of filter characteristics; the approximation model is used to differentiate a speed detection component ωm of the motor rotational speed to determine a motor acceleration torque component TmA∗; the motor acceleration torque component TmA∗ is passed through a band-pass filter Fcomp(s) to obtain a compensation torque component TFcomp which is then subtracted from an input torque command Tref.