基于深度强化学习的发动机涡轮鲁棒控制方法设计与应用测试  

Design and Application Test of Robust Control Method for Engine Turbine Based on Deep Reinforcement Learning

作　　者：刘腾 LIU Teng(Shanxi Railway Vocational and Technical College,Taiyuan 030013,Shanxi,China)

机构地区：[1]山西铁道职业技术学院,山西太原030013

出　　处：《凿岩机械气动工具》2025年第3期159-161,共3页Rock Drilling Machinery & Pneumatic Tools

摘　　要：文章提出一种基于深度强化学习的发动机涡轮鲁棒控制方法,旨在提高发动机涡轮在复杂工况下的运行稳定性和性能。建立涡轮系统数学模型,引入了深度强化学习方法,通过设计状态空间、动作空间和奖励函数,采用深度Q网络算法进行控制策略优化。测试结果表明,文章所提方法在超调量和稳态误差方面均表现出显著优势,能够有效抑制系统过度响应,减小稳态误差,提高系统的稳定性和一致性。This paper proposes a robust control method for engine turbine based on deep reinforcement learning,aiming to improve the operation stability and performance of engine turbine under complex working conditions.The mathematical model of the turbine system is built,the deep reinforcement learning method is introduced,and a deep Q-network algorithm is used to optimize the control strategy by designing the state space,action space and reward function.The test results show that the proposed method has significant advantages in overshoot and steady-state error,which can effectively suppress the over-response of the system,reduce the steady-state error,and improve the stability and consistency of the system.

关 键 词：深度强化学习 发动机 涡轮 鲁棒控制 

分 类 号：V434.211[航空宇航科学与技术—航空宇航推进理论与工程]