Model predictive control with adaptive Kalman filter for premixed turbocharged natural gas engine.

Area of Science:

• Engineering
• Control Systems
• Automotive Engineering

Background:

• Controlling natural-gas engines is difficult due to complex dynamics and unknown load disturbances.
• Existing methods struggle with strong couplings and delays in multi-input multi-output (MIMO) systems.

Purpose of the Study:

• To develop a robust control framework for natural-gas engines facing unknown load disturbances.
• To enhance engine performance, stability, and efficiency through adaptive control strategies.

Main Methods:

• Integration of rate-based model predictive control (MPC) with a gain-scheduling scheme.
• Utilizing an adaptive Kalman filter with a novel adaptation mechanism for load torque estimation.
• Online torque estimation to compute equilibrium points and generate adaptive gain-scheduling parameters.

Main Results:

• The adaptive Kalman filter rapidly tracks load torque transients while minimizing steady-state noise.
• Gain-scheduled MPC significantly reduces speed and air-fuel ratio deviations.
• Experimental validation shows improved transient response and shorter settling times after load changes.

Conclusions:

• The proposed control framework offers improved disturbance rejection for natural-gas engines.
• The adaptive gain-scheduling approach enhances practical applicability for power-generation engines.
• This method provides a robust solution for dynamic engine control challenges.