PCE-based uncertain PWA modeling and control for DMHP mode transition system

Area of Science:

• Automotive Engineering
• Control Systems Engineering
• Hybrid Electric Vehicle Technology

Background:

• Manufacturing variations, friction, and wear introduce random uncertainties in powertrain parameters.
• These uncertainties, combined with nonlinear dynamics, degrade the mode transition process (MTP) quality in hybrid powertrains.
• Dual-motor hybrid powertrains (DMHPs) are particularly susceptible to these MTP performance issues.

Purpose of the Study:

• Investigate the impact of parameter uncertainties on MTP performance in a DMHP.
• Develop a coordinated control strategy to mitigate these adverse effects.
• Ensure robust and smooth MTP under uncertain operating conditions.

Main Methods:

• Performed local and global sensitivity analyses to identify critical powertrain parameters affecting MTP.
• Utilized Polynomial Chaos Expansion (PCE) to transform the uncertain piecewise-affine (PWA) model of MTP into a deterministic system.
• Designed a PWA-PCE H2 coordinated control strategy for robust MTP.

Main Results:

• Sensitivity analyses pinpointed key parameters significantly influencing MTP performance.
• The PCE method successfully converted the uncertain MTP model into a deterministic framework.
• The proposed PWA-PCE H2 control strategy demonstrated effectiveness in handling parameter uncertainties.
• Simulation and hardware-in-the-loop (HiL) tests validated the strategy's ability to maintain high-quality MTP.

Conclusions:

• Parameter uncertainties pose a significant challenge to smooth MTP in DMHPs.
• The developed coordinated control strategy effectively addresses these uncertainties.
• The proposed approach ensures robust and high-quality MTP performance even with system variations.