论文部分内容阅读
目的:混合动力挖掘机的能量管理策略直接影响着系统的燃油经济性。本文旨在通过研究混合动力挖掘机能量管理系统,得到最优能量管理策略,并开发实时能量管理控制器,降低系统的燃油消耗。创新点:1.通过强化学习算法,设计时间无关的实时能量管理控制器;2.通过极大值原理求得最优能量管理问题的解析解,并用来辅助实时能量管理控制器设计。方法:1.建立负载的马尔科夫模型,运用强化学习算法,得到实时能量管理控制器;2.运用极大值原理,求得最优能量管理问题的解析解,并将其作为初始能量管理策略;3.通过仿真模拟和实验研究,验证所设计的实时能量控制器的性能。结论:1.基于强化学习的能量管理控制器是一个可以在线应用的与时间无关的实时能量管理控制器;2.基于强化学习的能量管理控制器优于广泛使用的恒温控制器和等效消耗最小化策略控制器;3.基于强化学习的能量管理控制器由于其闭环特性可适用于不同类型的作业工况。
Objective: The energy management strategy of hybrid excavators has a direct impact on the fuel economy of the system. The purpose of this paper is to study the energy management system of hybrid excavator to get the optimal energy management strategy and to develop a real-time energy management controller to reduce the system’s fuel consumption. Innovations: 1. Real-time energy management controller with non-time-related design through reinforcement learning algorithm; 2. Analytic solution to the optimal energy management problem by maximum principle and used to assist real-time energy management controller design. Method: 1. Establish the Markov model of load, and use the reinforcement learning algorithm to get the real-time energy management controller. 2. Apply the maximum principle to get the analytical solution of the optimal energy management problem, and use it as the initial energy management Strategy; 3. Through the simulation and experimental study, verify the performance of the designed real-time energy controller. Conclusion: 1.The energy management controller based on reinforcement learning is a time-independent real-time energy management controller that can be applied online.2. The energy management controller based on reinforcement learning is superior to the widely used thermostatic controller and its equivalent consumption Minimization Strategy Controller; 3. Energy Management Controller Based on Reinforcement Learning Because of its closed-loop characteristics, it can be applied to different types of working conditions.