检索结果-内蒙古大学图书馆

Visual feedback vibration control of flexible hinged plate system based on reinforcement learning algorithm

MECHANICAL SYSTEMS AND SIGNAL PROCESSING 2025年 224卷

作者： Qiu, Zhi-cheng Liu, Yi-hong South China Univ Technol Sch Mech & Automot Engn Guangzhou 510641 Peoples R China

In order to simulate the bending and torsional coupled vibration of solar panels under external disturbances, an experimental system of a fixed flexible hinged plate is established. The dynamic model is established by finite element method (FEM), and the model parameters are accurately identified by wavelet transform and simulated annealing (SA) optimization method. Two laser displacement sensors and binocular vision are combined to detect the bending and torsional vibrations and used for feedback control. A combined piezoelectric (PZT) control scheme is designed to suppress the bending and torsional vibrations of the flexible hinged plate. A vibration control method based on divergence augmented policy optimization (DAPO) reinforcement learning (RL) for the fixed flexible hinged plate is developed to train the modal controllers. Simulation and experiments are carried out to verify the effectiveness of the applied vibration control scheme and algorithm. The simulation and experimental results show that DAPO RL modal controllers have better vibration suppression effect than large gain proportional derivative (PD) controllers, especially for small amplitude residual vibration.

关键词： Flexible hinged plate Visual detection Vibration suppression Model identification Divergence-augmented policy optimization (DAPO) reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Intelligent Energy Management Strategy Based on an Improved reinforcement learning algorithm With Exploration Factor for a Plug-in PHEV

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2022年第7期23卷 8725-8735页

作者： Lin, Xinyou Zhou, Kuncheng Mo, Liping Li, Hailin Fuzhou Univ Coll Mech Engn & Automat Fuzhou 350108 Peoples R China West Virginia Univ Dept Mech & Aerosp Engn Morgantown WV 26506 USA

An intelligent energy management strategy (EMS) based on an improved reinforcement learning (RL) algorithm is developed to enhance the adaptability of the EMS and to further improve the fuel efficiency of a Plug-in Parallel Hybrid Electric Vehicle (PHEV). Both the numerical model and the energy management strategy of a plug-in PHEV are described. The improved RL with Q-learning algorithm is implemented to acquire the optimal control strategies for improving fuel economy. The Markov Chain is employed to calculate the Transition Probability Matrix of the required power. A Kullback-Leibler (KL) divergence rate is designed to activate the update of EMS, when a new corresponding driving cycle is expected. An Exploration Factor (FT) is proposed to overcome the disadvantages of the normal RL algorithm in convergence rate and reward cost evaluation. The diverse KL divergence rates are examined to seek optimal solutions. The normal-RI, strategy, rule-based strategy, and dynamic programming strategy are implemented as benchmark strategies to verify the effectiveness of the proposed strategy. The validation results indicate that the improved RL algorithm with FY makes it possible to promote the EMS capable of significantly improving the energy efficiency of a plug-in PHEV.

关键词： Hybrid electric vehicle energy management strategy reinforcement learning algorithm self-adaptive exploration factor

来源：评论

学校读者我要写书评

暂无评论

Optimization of river environmental management based on reinforcement learning algorithm: a case study of the Yellow River in China

引用

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH 2023年第3期30卷 8170-8187页

作者： Chen, Minghong Li, Ying Zhang, Xue Liao, Renkuan Wang, Changxi Bi, Xiangshuai China Agr Univ Coll Water Resources & Civil Engn Beijing 100083 Peoples R China China Agr Univ Coll Land Sci & Technol Beijing 100083 Peoples R China

Generating scientific management strategy contributes to the sustainable development of river ecological environment. In this study, a multi-objective coupled water and sediment regulation model aiming at minimizing sedimentation and inundation loss as well as maximizing ecological value in the lower Yellow River has been developed. A reinforcement Q-learning algorithm was used to obtain optimized strategies from the multi-objective of sediment reduction, flood control and ecological restoration under different hydrological years. The results showed that the simulated channel sedimentation is very close to the measured value, which proves the applicability of the developed model. Under dry, normal and wet hydrological year, the effects of various regulation strategies on silt reduction, flood control and ecological restoration were obviously different. The regulation scheme of discharge at 3700 m(3)/s was verified to be suitable for dry and wet year, and that of discharge at 2600 m(3)/s was more suitable for normal year. Increasing the spacing of the beach area was better in normal year and wet year. Our findings suggested optimized strategies to address environmental challenges of the lower Yellow River in different hydrological years. This paper provides a reliable reference for improving the management of the lower Yellow River.

关键词： River siltation River scouring River dredging reinforcement learning algorithm Water ecological value

来源：评论

学校读者我要写书评

暂无评论

Observer-Based Adaptive Optimized Control for Uncertain Cyclic Switched Nonlinear Systems: reinforcement learning algorithm Approach

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS 2024年第5期71卷 2203-2216页

作者： Yan, Chengyuan Xia, Jianwei Park, Ju H. Sun, Wei Xie, Xiangpeng Liaocheng Univ Sch Math Sci Liaocheng 252000 Shandong Peoples R China Yeungnam Univ Dept Elect Engn Gyongsan 38541 South Korea Nanjing Univ Posts & Telecommun Inst Adv Technol Nanjing 210023 Peoples R China

This article addresses adaptive optimized tracking control problem for strict-feedback cyclic switched nonlinear output constrained systems with average cyclic dwell time (ACDT). Different from most of the existing results on optimized control of switched nonlinear systems, a new mode-dependent reinforcement learning (RL) algorithm of identifier-critic-actor architecture is designed. Technically, with the help of neural networks (NNs) switched state observer, the virtual and actual optimal controllers are developed by solving the Hamilton-Jacobi-Bellman (HJB) equation. Meanwhile, to reduce the impact of systems switching on the overall optimization control, the information of the switching signal is considered into the optimal performance index functions. In an attempt to settle the output constraints, a nonlinear output-dependent time-varying function is used to ensure that the system output never transgresses the prescribed regions. More importantly, by combining the improved ACDT method and Lyapunov stability theorem, a novel adaptive optimized control scheme is put forward which ensures that the boundedness of all signals in the closed-loop system. Finally, the effectiveness of the proposed optimized control algorithm is verified by numerical as well as practical simulations.

关键词： Neural networks switched nonlinear systems adaptive optimized control reinforcement learning algorithm switched state observer

来源：评论

学校读者我要写书评

暂无评论

Adaptive Fault-Tolerant Tracking Control for MIMO Discrete-Time Systems via reinforcement learning algorithm With Less learning Parameters

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2017年第1期14卷 299-313页

作者： Liu, Lei Wang, Zhanshan Zhang, Huaguang Northeastern Univ Coll Informat Sci & Engn Shenyang 110819 Liaoning Peoples R China State Key Lab Synthet Automat Proc Ind Shenyang 110819 Liaoning Peoples R China

This paper is concerned with a reinforcement learning-based adaptive tracking control technique to tolerate faults for a class of unknown multiple-input multiple-output nonlinear discrete-time systems with less learning parameters. Not only abrupt faults are considered, but also incipient faults are taken into account. Based on the approximation ability of neural networks, action network and critic network are proposed to approximate the optimal signal and to generate the novel cost function, respectively. The remarkable feature of the proposed method is that it can reduce the cost in the procedure of tolerating fault and can decrease the number of learning parameters and thus reduce the computational burden. Stability analysis is given to ensure the uniform boundedness of adaptive control signals and tracking errors. Finally, three simulations are used to show the effectiveness of the present strategy. Note to Practitioners-As the practical engineering systems are becoming more and more complex, it is more likely to suffer from faults which can lead to unpredictable behaviors and serious damages. Therefore, adaptive fault tolerant control technique plays an important role in modern engineering systems. As a matter of fact, it is always desired to minimize the maintenance cost even when some faults occur in the systems, which can reduce the energy consumption in the industrial production process. One way is to integrate the adaptive reinforcement learning algorithm into the fault tolerant controller. Furthermore, a major restriction of the learning algorithm is that a large number of parameters should be tuned online, which directly increase the computational burden. In order to tackle this difficulty, by adjusting the estimated values of the network weight vectors instead of their weights, the number of adaptive learning parameters is decreased, and the computational burden is reduced dramatically.

关键词： Adaptive critic design fault tolerant control multiple-input multiple-output discrete-time systems neural networks reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Adaptive Fault-Tolerant Tracking Control for Discrete-Time Multiagent Systems via reinforcement learning algorithm

引用

IEEE TRANSACTIONS ON CYBERNETICS 2021年第3期51卷 1163-1174页

作者： Li, Hongyi Wu, Ying Chen, Mou Guangdong Univ Technol Sch Automat Guangzhou 510006 Peoples R China Guangdong Univ Technol Guangdong Prov Key Lab Intelligent Decis & Cooper Guangzhou 510006 Peoples R China Bohai Univ Coll Engn Jinzhou 121013 Peoples R China Nanjing Univ Aeronaut & Astronaut Coll Automat Engn Nanjing 211106 Peoples R China

This article investigates the adaptive fault-tolerant tracking control problem for a class of discrete-time multiagent systems via a reinforcement learning algorithm. The action neural networks (NNs) are used to approximate unknown and desired control input signals, and the critic NNs are employed to estimate the cost function in the design procedure. Furthermore, the direct adaptive optimal controllers are designed by combining the backstepping technique with the reinforcement learning algorithm. Comparing the existing reinforcement learning algorithm, the computational burden can be effectively reduced by using the method of less learning parameters. The adaptive auxiliary signals are established to compensate for the influence of the dead zones and actuator faults on the control performance. Based on the Lyapunov stability theory, it is proved that all signals of the closed-loop system are semiglobally uniformly ultimately bounded. Finally, some simulation results are presented to illustrate the effectiveness of the proposed approach.

关键词： reinforcement learning Artificial neural networks Actuators Fault tolerance Fault tolerant systems Estimation Multi-agent systems Discrete-time multiagent systems (MASs) fault-tolerant control neural networks (NNs) reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

RESEARCH ON ELEVATOR GROUP SCHEDULING STRATEGY AND SIMULATION BASED ON reinforcement learning algorithm

引用

MECHATRONIC SYSTEMS AND CONTROL 2024年第4期52卷 251-259页

作者： Tian, Rui Gao, Weimin Guangdong Construction Polytech Sch Mech & Elect Engn Guangzhou 510440 Guangdong Peoples R China Hunan Univ Med Sch Med Humanities & Informat Management Huaihua 418000 Hunan Peoples R China

The multi-objective elevator group optimisation problem has attracted widespread attention due to its high practical significance. The elevator group control system (EGCS) is a typical multi- objective system designed to increase passenger service and reducing costs, such as energy consumption (EC). Based on the characteristics of elevator group scheduling problem, this paper treats EGCS as multiple reinforcement learning agents that cooperate with each other. Each agent controls the operation of an elevator according to the average reward reinforcement learning algorithm (RLA), aiming to minimise the average waiting time (AWT) of passengers. Additionally, a neural network is employed to store and update the state behaviour value functions while dynamically classifying the state of unknown environments. The simulation results show that the reinforcement learning based-scheduling algorithm has better scheduling effects than the static partition scheduling algorithm (SPSA) and cultural algorithm (CA).

关键词： Elevator group control systems adaptive multi-objective optimisa- tion reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

New Trust Management Scheme Based on Blockchain and KNN reinforcement learning algorithm

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS 2024年第4期15卷 1158-1176页

作者： Aljohani, Ahdab Hulayyil Al-Shammri, Abdulaziz Imam Muhammed ibn Saud Islamic Univ IMISIU Comp Sci Informat Secur Dept Riyadh Saudi Arabia

There has been a continual rise in the quantity of smart and autonomous automobiles in recent decades. the effectiveness of communication among vehicles in Vehicular Ad-hoc Networks (VANET) is critical for ensuring the safety of drivers' lives. the primary objective of VANET is to share critical information regarding life-threatening events, such as traffic jams and accident alerts in a timely and accurate manner. Nevertheless, typical VANETs encounter several security issues involving threats to confidentiality, integrity, and availability. This paper proposes a new decentralized and tamper-resistant scheme for privacy preservation. We designed a new trust management system that utilizes blockchain technology. We strive to establish trust between vehicles and infrastructure and preserve privacy by guaranteeing the authenticity and integrity of the information exchanged in VANETS. Our proposal adopts the principles of reinforcement learning to dynamically evaluate and allocate trust scores to vehicles and infrastructure based on their behavior. The scheme's performance has been evaluated based on key metrics. The results show that our new system provides an effective behavior management technique while preserving vehicle privacy.

关键词： Vehicular Ad hoc Networks (VANETs) Blockchain trust management reinforcement learning algorithm privacy preservation network security

来源：评论

学校读者我要写书评

暂无评论

reinforcement learning algorithm for One-warehouse Multi-retailer Inventory Problem

Reinforcement Learning Algorithm for One-warehouse Multi-ret...

引用

2015 International Conference on Automation,Mechanical and Electrical Engineering(AMEE 2015)

作者： C.Y.Li X.T.Wang T.W.Zhang School of Computer Science and Technology Harbin University of Science and Technology School of Computer Science and Technology Harbin Institute of Technology

We considered a supply chain inventory scheduling problem in which a central warehouse serves n-retailers under. Mathematical model was developed to obtain the optimal revenue for the proposed policy and the objective function is to minimize revenue per unit time. reinforcement learning algorithm was used to solve the mathematical model, including state determination criteria, reward function, and other parameters. After determining the appropriate parameters of the algorithm, the effectiveness of the algorithm was verified by numerical example. The performance of the algorithm was tested for randomly generated problems. Results show the robust performances of the proposed algorithm.

关键词： Supply chain management one-warehouse multi-retailer problem reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

A Memory-based reinforcement learning algorithm for Partially Observable Markovian Decision Processes

A Memory-based Reinforcement Learning Algorithm for Partiall...

引用

International Joint Conference on Neural Networks

作者： Zheng, Lei Cho, Siu-Yeung Quek, Chai Nanyang Technol Univ Sch Comp Engn Singapore Singapore

ISBN: (纸本)9781424418206

This paper presents a modified version of U-Tree [1], a memory-based reinforcement learning (RL) algorithm that uses selective perception and short-term memory to handle partially observable Markovian decision processes (POMDP). Conventional RL algorithms rely on a set of pre-defined states to model the environment, even though it can learn the state transitions from experience. U-Tree is not only able to do that, it can also build the state model by itself based on raw sensor inputs. This paper enhances U-Tree's model generation process. The paper also shows that because of the simplified and yet effective state model generated by U-Tree, it is feasible and preferable to adopt the classical Dynamic Programming (DP) algorithm for average reward MDP to solve some difficult POMDP problems. The new U-Tree is tested using a car-driving task with 31,224 world states, with the agent having very limited sensory information and little knowledge about the dynamics of the environment.

关键词： reinforcement learning algorithm Partially Obersvable Markovian Decision Processs Dynamic Programming Average Reward

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：