Search

Deep reinforcement learning (DRL), which excels at solving a wide variety of Atari and board games, is an area of machine learning that combines the deep learning approach and reinforcement learning (RL). However, to the authors' best knowledge, there seem to be few studies that apply the latest DRL algorithms on real-world powertrain control problems. If there are any, the requirement of classical model-free DRL algorithms typically for a large number of random exploration in order to realize good control performance makes it almost impossible to implement directly on a real plant. Unlike most of the other DRL studies, whose control strategies can only be trained in a simulation environment—especially when a control strategy has to be learned from scratch—in this study, a hybrid end-to-end control strategy combining one of the latest DRL approaches—i.e., a dueling deep Q-network and traditional Proportion Integration Differentiation (PID) controller—is built, assuming no fidelity simulation model exists. Taking the boost control of a diesel engine with a variable geometry turbocharger (VGT) and cooled (exhaust gas recirculation) EGR as an example, under the common driving cycle, the integral absolute error (IAE) values with the proposed algorithm are improved by 20.66% and 9.7% respectively for the control performance and generality index, compared with a fine-tuned PID benchmark. In addition, the proposed method can also improve system adaptiveness by adding another redundant control module. This makes it attractive to real plant control problems whose simulation models do not exist, and whose environment may change over time. [ABSTRACT FROM AUTHOR]

Deep reinforcement learning (DRL) is an area of machine learning that combines a deep learning approach and reinforcement learning (RL). However, there seem to be few studies that analyze the latest DRL algorithms on real-world powertrain control problems. Meanwhile, the boost control of a variable geometry turbocharger (VGT)-equipped diesel engine is difficult mainly due to its strong coupling with an exhaust gas recirculation (EGR) system and large lag, resulting from time delay and hysteresis between the input and output dynamics of the engine's gas exchange system. In this context, one of the latest model-free DRL algorithms, the deep deterministic policy gradient (DDPG) algorithm, was built in this paper to develop and finally form a strategy to track the target boost pressure under transient driving cycles. Using a fine-tuned proportion integration differentiation (PID) controller as a benchmark, the results show that the control performance based on the proposed DDPG algorithm can achieve a good transient control performance from scratch by autonomously learning the interaction with the environment, without relying on model supervision or complete environment models. In addition, the proposed strategy is able to adapt to the changing environment and hardware aging over time by adaptively tuning the algorithm in a self-learning manner on-line, making it attractive to real plant control problems whose system consistency may not be strictly guaranteed and whose environment may change over time. [ABSTRACT FROM AUTHOR]