检索结果-内蒙古大学图书馆

26th International Conference on Methods and Models in Automation and Robotics

作者： Musial, Jakub Stebel, Krzysztof Czeczot, Jacek Silesian Tech Univ Fac Automat Control Elect & Comp Sci Dept Automat Control & Robot PL-44100 Gliwice Poland

ISBN: (数字)9781665468589

ISBN: (纸本)9781665468596;9781665468589

This paper presents a new approach to the general-purpose self-improving controller based on q-learning control strategies. The previous approach was based on a three-dimensional q-matrix, significantly slowing down the learning process and limiting the ability of practical implementation in industrial practice. The proposed new algorithm solves this problem by reducing the size of the matrix and the number of tuning parameters of the method without sacrificing its accuracy and learning capabilities. The algorithm is validated by simulation on the two second-order dynamics models and the results show significant improvement compared to the previous version of the developed method.

关键词： process control q-learning algorithm reinforcement learning online learning intelligent control

来源：评论

学校读者我要写书评

暂无评论

A q-learning-based Multi-timescale Resilience Enhancement Approach for Power Grids with High Renewables 2

A Q-learning-based Multi-timescale Resilience Enhancement Ap...

引用

IEEE 2nd International Conference on Power Science and Technology (ICPST)

作者： Huang, Yanting Zhong, qing Wang, Akang Lin, Shunjiang Peng, Chaoyi Lei, Shunbo Chinese Univ Hong Kong Sch Sci & Engn Shenzhen Peoples R China Shenzhen Res Inst Big Data Shenzhen Peoples R China South China Univ Technol Sch Elect Power Engn Guangzhou Peoples R China China Southern Power Grid Power Dispatch & Control Ctr Guangzhou Peoples R China

ISBN: (纸本)9798350349047;9798350349030

Power grid resilience is critically undermined by extreme weather events, especially when coupled with uncertain renewable energy generation, which can lead to a severe power outage. When facing such outages, it is essential for system operators to restore damaged lines in a sequence that maximizes the overall power supply. Identifying the optimal restoration sequence can be cast as a multi-stage problem where at each stage a single line is first restored in an hourly basis and then numerous power flow problems within the timescale of minutes are optimized under the resulting network topology. Evaluating the economic effect of restoring a specific line entails the optimization of power flow problems with time-varying renewables, lacking a closed-form expression. To address this, we employ the q-learning algorithm, utilizing a well-designed q-function to approximate the effect of restoring each line and embedding such a function into a training process. Simulations on the IEEE 33-Bus and IEEE 39-Bus Systems confirm the efficacy of the proposed approach, showcasing significant enhancements in resilience and recovery capabilities.

关键词： Power Grid Resilience q-learning algorithm Power Flow Optimization Renewable Energy Generation Multi-timescale

来源：评论

学校读者我要写书评

暂无评论

q-learning Dynamic Path Planning for an HCV Avoiding Unknown Threatened Area

Q-Learning Dynamic Path Planning for an HCV Avoiding Unknown...

引用

Chinese Automation Congress (CAC)

作者： Lv, Yali Hao, Dongchen Gao, Yang Li, Yuankai Univ Elect Sci & Technol China Sch Aeronaut & Astronaut Chengdu 611731 Peoples R China Univ Elect Sci & Technol China Prov Key Lab Aircraft Swarm Intelligent Sensing & Chengdu 611731 Peoples R China

ISBN: (纸本)9781728176871

For a hypersonic cruise vehicle (HCV) avoiding unknown threats from ground or space, dynamic path planning is an essential issue to fly safely in near space. q-learning is the most typical one of reinforcement learning methods that has been effectively used for robotic path planning. However, few works is reported for an HCV with hypersonic flying characteristics considered. In this paper, a q-learning based dynamic path planning method is presented. In the method, the vehicle is assumed as a mass point with time-invariant altitude and velocity for simplicity and the flight zone of the vehicle is modeled as an average grid map, where an epsilon-greedy strategy is designed to establish the q-learning algorithm. A simulation case has illustrated the effectiveness of the given method and shows that the method is suitable for the cruise phase path planning of an HCV.

关键词： q-learning algorithm reinforcement learning epsilon-greedy strategy hypersonic cruise vehicle path planning

来源：评论

学校读者我要写书评

暂无评论

Evaluation of q-learning for Search and Inspect Missions Using Underwater Vehicles

Evaluation of Q-Learning for Search and Inspect Missions Usi...

引用

Oceans Conference

作者： Frost, Gordon Lane, David M. Heriot Watt Univ Sch Engn & Phys Sci Ocean Syst Lab Edinburgh EH14 4AS Midlothian Scotland

ISBN: (纸本)9781479949182

An application for offline Reinforcement learning in the underwater domain is proposed. We present and evaluate the integration of the q-learning algorithm into an Autonomous Underwater Vehicle (AUV) for learning the action-value function in simulation. Three separate experiments are presented. The first compares two search policies: the c - least visited, and random action, with respect to convergence time. The second experiment presents the effect of the learning discount factor, gamma, on the convergence time of the c - least visited search policy. The final experiment is to validate the use of a policy learnt offline on a real AUV. This learning phase occurs offline within the continuous simulation environment which had been discretized into a grid-world learning problem. Presented results show the system's convergence to a global optimal solution whilst following both sub-optimal policies during simulation. Future work is introduced, after discussion of our results, to enable the system to be used in a real world application. The results presented, therefore, form the basis for future comparative analysis of the necessary improvements such as function approximation of the state space.

关键词： autonomous underwater vehicles convergence learning (artificial intelligence) military computing ε-least visited search policy AUV q-learning algorithm q-learning evaluation action-value function autonomous underwater vehicle convergence time function approximation inspect missions reinforcement learning search missions Computer architecture Convergence Robot sensing systems Software Sonar Vehicles

来源：评论

学校读者我要写书评

暂无评论

Fuzzy q - learning for Uniform Price Wholesale Power Markets

Fuzzy Q - Learning for Uniform Price Wholesale Power Markets

引用

3rd International Conference on Communication Systems and Network Technologies (CSNT)

作者： Thanhquy Bach Yao, Jiangang Yang, Shengjie Hunan Univ Fac Elect Engn Changsha 410082 Hunan Peoples R China

ISBN: (纸本)9780769549583;9781467356039

The generators gain more profits through bidding strategies become an important in the open competitive market. In the incomplete information market, the generators are looking for a best bidding strategy according to the changing market structure and their own conditions are a significant. The reinforcement q - learning combined with fuzzy control are considered to solve this problem. The day - ahead uniform price wholesale power market model is presented. The model is suitable for multi area power system and which simulated throughout IEEE 30 bus power system with six generators in three areas. The resulting simulations demonstrated the Fuzzy q - learning (Fq - leaning) has convergent more rapidly than the conventional q - learning and the supplier agents can choose the best bidding strategy according to the changing power market structure.

关键词： q-learning algorithm reinforcement learning optimal strategy bidding wholesale power market locational marginal price

来源：评论

学校读者我要写书评

暂无评论

An Approach for Maximizing Container Loading and Minimizing the Waste of Space Using q-learning

An Approach for Maximizing Container Loading and Minimizing ...

引用

International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE)

作者： Tijjani, Sani Bucak, Ihsan Omur Mevlana Rumi Univ Dept Comp Engn Selcuklu Konya Turkey

ISBN: (纸本)9781467356138;9781467356121

This paper suggests a methodology for tackling container loading problem by using a q-learning technique. One of the important issues for the efficient operation of container loading problem is to find software that can determine the order in which the containers are to be arranged in order to speed up the process. However, the efficient way to arrange containers involves new and specialized logistic process, a number of logistics plan and automated system to handle a great number of containers. This study proposes a method for maximizing the load of container and consequently minimizing the waste of space.

关键词： Container loading problem (CLP) q-learning algorithm quay crane (qC) Yard Crane (YC)

来源：评论

学校读者我要写书评

暂无评论

Optimization and simulation of distribution system in a supply chain based on q-learning

Optimization and simulation of distribution system in a supp...

引用

3rd International Conference on Service Systems and Service Management

作者： Li Suicheng Lin Jun Yin Hongying Xian Univ Technol Sch Business Adm Xian Peoples R China

ISBN: (纸本)1424404509

A distribution system consisting of multiple manufactures and one retailer was studied in this paper. In this system, the demand of the manufactures' products is sensitive to service level, and their products have substitution. The manufactures motivate the retailer to maximize their profits by providing bonus. The retailer provides corresponding service level to maximize its profits by obtaining bonus from manufactures. A simulation was done to this system under both non-cooperation and cooperation situations based on q-learning algorithm. The simulated results show the manufactures' profits are damaged in non-cooperation situation. The manufactures would be benefited as they adopt the cooperation action, but the retailer's profit is damaged.

关键词： supply chain management distribution system q-learning algorithm

来源：评论

学校读者我要写书评

暂无评论

A Flexible q-relearning Method to Accelerate learning Under the Change of Environments by Reusing a Portion of Useful Policies

A Flexible <i>Q</i>-relearning Method to Accelerate Learning...

引用

Annual Conference of the Society-of-Instrument-and-Control-Engineers (SICE)

作者： Saito, Masanori Masuda, Kazuaki Kurihara, Kenzo Kanagawa Univ Fac Engn Kanagawa Japan

ISBN: (纸本)9784907764401

We propose a relearning method for q-learning algorithm, called "q-relearning," to accelerate learning under the change of environments by reusing a portion of useful policies. To deal with the problem that q-learning algorithm can't adapt to the change of environments efficiently, we developed an original q-relearning method in the past study. However, there was little flexibility in choosing the portion to be reused. In this paper, we revise the q-relearning method to permit any choice of the portion by adding the possibility to update the fixed q-values under particular conditions. We show the effectiveness of the improved q-relearning method by numerical experiments.

关键词： machine learning reinforcement learning q-learning algorithm relearning

来源：评论

学校读者我要写书评

暂无评论

Self-improving q-learning based controller for a class of dynamical processes

引用

ARCHIVES OF CONTROL SCIENCES 2021年第3期31卷 527-551页

作者： Musial, Jakub Stebel, Krzysztof Czeczot, Jacek Silesian Tech Univ Fac Automat Control Elect & Comp Sci Dept Automat Control & Robot Ul Akad 16 PL-44100 Gliwice Poland

This paper presents how q-learning algorithm can be applied as a general-purpose self-improving controller for use in industrial automation as a substitute for conventional PI controller implemented without proper tuning. Traditional q-learning approach is redefined to better fit the applications in practical control loops, including new definition of the goal state by the closed loop reference trajectory and discretization of state space and accessible actions (manipulating variables). Properties of q-learning algorithm are investigated in terms of practical applicability with a special emphasis on initializing of q-matrix based only on preliminary PI tunings to ensure bumpless switching between existing controller and replacing q-learning algorithm. A general approach for design of q-matrix and learning policy is suggested and the concept is systematically validated by simulation in the application to control two examples of processes exhibiting first order dynamics and oscillatory second order dynamics. Results show that on-line learning using interaction with controlled process is possible and it ensures significant improvement in control performance compared to arbitrarily tuned PI controller.

关键词： process control q-learning algorithm reinforcement learning intelligent control on-line learning

来源：评论

学校读者我要写书评

暂无评论

A q-learning artificial bee colony for distributed assembly flow shop scheduling with factory eligibility, transportation capacity and setup time

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2023年第PartA期123卷

作者： Wang, Jing Tang, Hongtao Lei, Deming Wuhan Univ Technol Sch Automat Wuhan Peoples R China Wuhan Univ Technol Sch Mech & Elect Engn Hubei Digital Mfg Key Lab Wuhan Peoples R China

Distributed assembly flow shop scheduling problem (DAFSP) has been considered;however, DAFSP with factory eligibility,transportation capacity and setup time is seldom studied even though these three constraints often exist simultaneously in real-life multi-factory assembly production processes. In this study, DAFSP with Pm-+ 1 layout, factory eligibility, transportation capacity and setup time is proposed and a q-learning artificial bee colony (q-LABC) is presented to minimize both makespan and total tardiness. A heuristic is designed to produce initial solution. In employed bee phase, factory set based heuristic is developed, population is divided into two parts by non-dominated sorting and two global search operators are applied in two parts respectively. A q-learning algorithm has a fixed q-table when a given condition is met, and is used to dynamically decide the way of selecting food source and the number of the selected food source for onlooker bee. Extensive experiments on 90 instances are conducted to show the effectiveness of new strategies and compare q-LABC with two comparative algorithms. The computational results demonstrate that the new strategies of q-LABC are effective and q-LABC performs better than its comparative algorithms on at least 83.33% and 94.44% instances, respectively.

关键词： Distributed scheduling Assembly Factory eligibility Artificial bee colony q-learning algorithm

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：