Learning-Based Decision Making for Combustion Phasing Control in Multi-Fuel CI Engines with Latent Fuel Reactivity Estimation

2026-06-16

AI Summary

This study addresses the challenge of precise combustion phasing (CA50) control in multi-fuel compression-ignition engines, where fuel cetane number (CN) varies over time and is inherently unmeasurable. The problem is formulated as a partially observable sequential decision-making task, and a recurrent deep deterministic policy gradient (DDPG) framework incorporating gated recurrent units (GRUs) is proposed. This approach jointly optimizes online estimation of fuel reactivity from combustion history and control policy, circumventing the train-deploy mismatch inherent in conventional “estimate-then-control” paradigms. By integrating a Gaussian process surrogate model, GRU-based latent representations, and actuator signals, the method enables end-to-end reactivity-aware closed-loop control. Evaluated on unseen CN trajectories, the controller achieves a CA50 tracking mean absolute error below 0.25° crank angle, while producing smooth and physically consistent outputs for injection timing and glow plug power.

Abstract

Multi-fuel compression-ignition engines offer fuel flexibility but introduce uncertain, time-varying fuel reactivity, represented by cetane number (CN), which complicates cycle-to-cycle combustion-phasing control. This work formulates CA50 regulation under latent CN variation as a partially observable sequential decision problem and systematically evaluates controllers with increasing temporal and representational capacity, including LinUCB, history-augmented contextual bandits, observation-only DDPG, recurrent DDPG, and a proposed GRU-guided RL framework. A Gaussian-process surrogate trained on experimental multi-fuel engine data provides a controlled and reproducible evaluation environment. Results show that myopic and fixed-history bandit methods degrade under CN variation, observation-only RL suffers from latent-state aliasing, and generic recurrence is insufficient when CN evolves rapidly. The proposed framework learns a compact GRU-based representation of fuel reactivity from combustion history and conditions both actor and critic on this estimated signal rather than oracle CN. By training the policy on the same imperfect fuel-reactivity information available at deployment, the controller avoids train-deploy inconsistency in conventional online estimate-then-control pipelines. Across unseen CN trajectories, the policy achieves stable CA50 regulation with mean absolute tracking error below 0.25° CA at the training setpoint, while producing smooth, physically consistent SOI and glow-plug-power actuation. These results show that combustion control under latent, continuously evolving fuel dynamics requires more than standalone estimation or generic recurrence. By aligning fuel-reactivity inference with control policy learning, the proposed framework enables reactivity-aware decision-making using the same estimated state available during deployment.

Problem

• combustion phasing control
• fuel reactivity
• cetane number
• multi-fuel CI engines
• latent state

Innovation

• GRU-guided RL
• latent fuel reactivity estimation
• combustion phasing control
• partially observable sequential decision
• multi-fuel CI engines