Peng Yue

Peng Yue

Auto-generated profile, based on publicly available information

Peng Yue

State Key Laboratory of Synthetical Automation for Process Industries, Northeastern University, Shenyang, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 August 2023 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
PYPeng Yue
Peng Yue
YJYaochu Jin
Yaochu Jin
XDXuewu Dai
Xuewu Dai
ZFZhenhua Feng
Zhenhua Feng
DCDongliang Cui
Dongliang Cui
Train Timetable Rescheduling (TTR) is a crucial task in the daily operation of high-speed railways to maintain punctuality and efficiency in the presence of unexpected disturbances. However, it is challenging to promptly create a rescheduled timetable in real time. In this study, we propose a reinforcement-learning-based method for real-time rescheduling of high-speed trains. The key innovation of the proposed method is to learn a well-generalized dispatching policy from a large amount of samples, which can be applied to the TTR task directly. At first, the problem is transformed into a multi-stage decision process, and the decision agent is designed to predict dispatching rules. To enhance the training efficiency, we generate a small yet good-quality action set to reduce invalid explorations. Besides, we propose an action sampling strategy for action selection, which implements forward planning with consideration of evaluation uncertainty, thus improving search efficiency. Extensive experimental results demonstrate the effectiveness and competitiveness of the proposed method. It has been proven that the local policies trained by the proposed method can be applied to numerous problem instances directly, rendering it unnecessary to use human-designed rules.

13 February 2024 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
PYPeng Yue
Peng Yue
YJYaochu Jin
Yaochu Jin
XDXuewu Dai
Xuewu Dai
ZFZhenhua Feng
Zhenhua Feng
DCDongliang Cui
Dongliang Cui
Train timetable rescheduling (TTR) aims to promptly restore the original operation of trains after unexpected disturbances or disruptions. Currently, this work is still done manually by train dispatchers, which is challenging to maintain performance under various problem instances. To mitigate this issue, this study proposes a reinforcement learning-based approach to TTR, which makes the following contributions compared to existing work. First, we design a simple directed graph to represent the TTR problem, enabling the automatic extraction of informative states through graph neural networks. Second, we reformulate the construction process of TTR’s solution, not only decoupling the decision model from the problem size but also ensuring the generated scheme’s feasibility. Third, we design a learning curriculum for our model to handle the scenarios with different levels of delay. Finally, a simple local search method is proposed to assist the learned decision model, which can significantly improve solution quality with little additional computation cost, further enhancing the practical value of our method. Extensive experimental results demonstrate the effectiveness of our method. The learned decision model can achieve better performance for various problems with varying degrees of train delay and different scales when compared to handcrafted rules and state-of-the-art solvers.

18 December 2023 in IEEE Transactions on Automation Science and Engineering
Verified
Journal Article
Closed access
RLRuiguang Liu
Ruiguang Liu
DCDongliang Cui
Dongliang Cui
XDXuewu Dai
Xuewu Dai
PYPeng Yue
Peng Yue
ZYZhiming Yuan
Zhiming Yuan
In High-Speed Railway (HSR) networks with hub stations connecting multiple HSR lines, Train Timetable Rescheduling (TTR) under disruptions (such as speed restrictions caused by high wind) has been a challenging problem, which requires collaborative consideration of the traffic and impacts on all lines. Compared to the first principle model of complex railway networks, data-driven modeling provides a better solution to describe how the performance of one HSR line is affected by a train rescheduling decision made for another lines, but it faces the challenges of incompleteness, imbalance and lack of comprehensiveness of history data as disruptions in railways (e.g. delays, accidents) are relatively rare compared to normal operations. This paper proposes a multi-line rescheduling framework consisting of an interactive railway operation simulation and experiment (iROSE) system, a surrogate model and a heuristic algorithm to enable network-wise optimal rescheduling of multiple lines. To compensate for the limits of incomplete history data, a relatively low-cost but accurate enough surrogate model is developed from simulation data of the realistic but computation-intensive iROSE simulator. To reduce the demand for data and the time on running the costly simulator, a multi-surrogate search method is developed. A data expansion-based knowledge transfer method and joint distribution adaptation and tradaboost are also adopted to further improve the accuracy of the surrogate model. Our extensive experiments show that the proposed method can obtain higher precision fine search models with few simulations and solve the problem of TTR under wind-caused speed restrictions in complex railway networks with multiple lines. Note to Practitioners —This paper was motivated by the Train Timetable Rescheduling problem of complex high-speed railway networks of multiple lines connected via hub stations, in which the delays caused by high-wind speed restrictions on one line may easily affect trains on other lines in the network. Thus the impacts of a local-line TTR decision on other parts of the HSR network should be evaluated appropriately in the sense of precision and real-time, to assist the local dispatcher in making a network-wise decision. However, the incomplete and imbalanced historical data may not accurately capture how the system behaves during disruptions. In order to address these challenges, this paper proposes a data-driven rescheduling optimization framework to allow network-wise optimal decision-making. The proposed framework consists of an on-demand iROSE system, a surrogate model representing the operation performance of the whole railway network, and a heuristic method responsible for the traffic rescheduling of partial HSR lines. The realistic iROSE simulator is able to compensate the imbalanced actual history operation data by giving a precise evaluation of the network’s performance. Then a multi-surrogate search method and a knowledge transfer method are developed to avoid the time-consuming caused by expensive simulation. The developed surrogate model is able to capture the insights of delay propagation in a multi-line HSR network and enable the dispatchers to have a quick and comprehensive evaluation of how a TTR rescheduling decision made for one line affects other lines in the network. As a result, a network-wise better decision on train rescheduling can be made.

4 March 2025 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
YHYuxiang Hu
Yuxiang Hu
XDXuewu Dai
Xuewu Dai
PYPeng Yue
Peng Yue
JDJinliang Ding
Jinliang Ding
TCTianyou Chai
Tianyou Chai
Early faults, with their small amplitudes and slow variations, are easily masked by noise or process trends, making detection difficult. Also, observer-based methods struggle with timely and accurate early fault detection. To address these challenges, a novel observer-driven temporal graph convolutional network (OD-TGCN) is proposed for early fault detection of control systems subject to semi-stationary periodic input disturbances and measurement noise. We employ graph representation to describe the mechanistic relationships and capture temporal features between variables in the control system. First, graph nodes are constructed based on the control inputs, output signals of the control systems and the fault detection residual of the disturbance decoupling generalized proportional-integral observer (DD-GPIO). Then, a method for constructing the adjacency matrix of the graph based on the system matrix is provided. Finally, the output of the temporal graph convolutional network (TGCN) is passed to a Multi-Layer Perceptron (MLP) for graph-based fault detection. The proposed method is applied to a two-wheeled self-balancing robot. Comparative results show that OD-TGCN significantly outperforms DD-GPIO and typical TGCN in early fault detection accuracy. Additionally, OD-TGCN exhibits notable robustness across datasets with different disturbances/noise.

12 September 2025 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
Verified
Journal Article
Closed access
PYPeng Yue
Peng Yue
YJYaochu Jin
Yaochu Jin
XDXuewu Dai
Xuewu Dai
DCDongliang Cui
Dongliang Cui
Train timetable rescheduling (TTR) aims to promptly restore the original train operations in the event of unexpected disturbances. In recent years, learning-based methods, as a novel technology, have been widely studied, hoping to overcome the high computational cost of traditional methods. However, to our best knowledge, timetable robustness, as another critical factor in TTR, is often ignored in existing work. To fill this gap, this study proposes a learning-based method for rapidly restoring train delays with robustness guarantees. Compared with the existing work, our approach offers several benefits. First, we design a network architecture for learning-based methods when solving the robust TTR problem. The dual-headed network can deal with the reordering and re-timing operation simultaneously in the TTR problem. Second, we reformulate the robust TTR problem into a constrained Markov decision process (CMDP), and a Lagrange-based proximal policy optimization (LPPO) algorithm is employed to train the designed network. Through testing on real-scale problems, the learned model can not only provide the rescheduled timetable within seconds but also strike a fine balance between operation efficiency and the scheme’s robustness.

15 July 2023
Verified
Conference Paper
Open access
PYPeng Yue
Peng Yue
SLShiqing Liu
Shiqing Liu
YJYaochu Jin
Yaochu Jin
Vehicle routing problem with time windows (VRPTW) is a typical class of constrained path planning problems in the field of combinatorial optimization. VRPTW considers a delivery task for a given set of customers with time windows, and the target is to find optimal routes for a group of vehicles that can minimize the total transportation cost. The traditional heuristics suffer from several limitations when solving VRPTW, such as poor scalability, sensitivity to hyperparameters and difficulty in handling complex constraints. Recent advance in machine learning makes it possible to enhance heuristic approaches via learned knowledge. In this paper, we propose a graph Q-learning assisted ant colony optimization algorithm named GQL-ACO to solve VRPTW. Compared to vanilla ant colony optimization (ACO), our proposed method first employs the learned heuristic values by using graph Q learning, instead of handcrafted ones, to define the hyperparameters of ACO. Second, we design a collaborative search strategy by combining ACO and Q-learning effectively, which can adaptively adjust the hyperparameters of ACO based on the search experiences.