سال انتشار:
2020
عنوان انگلیسی مقاله:
Reinforcement learning approach for optimal control of multiple electric locomotives in a heavy-haul freight train: A Double-Switch-Q-network architecture
ترجمه فارسی عنوان مقاله:
رویکرد یادگیری تقویتی برای کنترل بهینه چندین لوکوموتیو الکتریکی در قطار باری سنگین حمل و نقل : معماری شبکه دو سوئیچ- Q
منبع:
Sciencedirect - Elsevier - Knowledge-Based Systems, 190 (2020) 105173. doi:10.1016/j.knosys.2019.105173
نویسنده:
Huiyue Tang a,c, Yuan Wangb,c,∗, Xiang Liu c, Xiaoyun Feng a
چکیده انگلیسی:
Electric locomotives provide high tractive power for fast acceleration of heavy-haul freight trains,
and significantly reduce the energy consumption with regenerative braking. This paper proposes
a reinforcement learning (RL) approach for the optimal control of multiple electric locomotives
in a heavy-haul freight train, without using the prior knowledge of train dynamics and the predesigned
velocity profile. The optimization takes the velocity, energy consumption and coupler force as
objectives, considering the constraints on locomotive notches and their change rates, speed restrictions,
traction and regenerative braking. Besides, since the problem in this paper has continuous state
space and large action space, and the adjacent actions’ influences on states share similarities, we
propose a Double-Switch Q-network (DSQ-network) architecture to achieve fast approximation of
the action-value function, which enhances the parameter sharing of states and actions, and denoises
the action-value function. In the numerical experiments, we test DSQ-network in 28 cases using the
data of China Railways HXD3B electric locomotive. The results indicate that compared with tablelookup
Q-learning, DSQ-network converges much faster and uses less storage space in the optimal
control of electric locomotives. Besides, we analyze 1)the influences of ramps and speed restrictions
on the optimal policy, and 2)the inter-dependent and inter-conditioned relationships between multiple
optimization objectives. Finally, the factors that influence the convergence rate and solution accuracy
of DSQ-network are discussed based on the visualization of the high-dimensional value functions.
Keywords: Reinforcement learning | Double-Switch Q-network | Optimal control | Electric locomotive | Heavy-haul freight train
قیمت: رایگان
توضیحات اضافی:
تعداد نظرات : 0