检索结果-内蒙古大学图书馆

Chinese Automation Congress (CAC)

作者： Yuan, Jun Chen, Zitao Wang, Xiaodong Zeng, Xianxian Zhang, Yun Guangdong Univ Technol Sch Automat Guangzhou Peoples R China

ISBN: (纸本)9781728176871

This paper proposes a new home energy management system (EMS) based on the adaptive dynamic programming (ADP) algorithm for economic dispatch problem of microgrid with distributed generations. The electricity price, maximum power output of renewable generation, and load are considered as uncontrollable state variables in this paper. Then two neural networks are used to implement the ADP to minimize a given utility function over a certain time horizon. In this work, the fluctuating amounts of available renewable energy are trained and evaluated in this ADP approach. EMS reduces the daily household energy costs, maximizing the PV self-consumption and increase customer benefits through economically managing the operation of the battery. Then demonstrated via computer simulation that the EMS can reduce the overall energy cost.

关键词： adaptive dynamic programming neural networks smart grid energy management system

来源：评论

学校读者我要写书评

暂无评论

Expandable depth and width adaptive dynamic programming for economic smart generation control of smart grids

引用

ENERGY 2021年 232卷 120964-120964页

作者： Yin, Linfei Luo, Shikui Ma, Chenxiao Guangxi Univ Coll Elect Engn Nanning 530004 Guangxi Peoples R China

The uncertainty of renewable energy sources (RESs) puts pressure on the economic dispatch of smart grids. However, many traditional multiple time-scales dispatching approaches lead to problems such as inverse regulation and cost increase. To optimize the economic dispatch of smart grids, firstly, a three state energy prosumers (TSEPs) model is proposed to facilitate flexible dispatching of RESs. Secondly, an economic smart generation control (ESGC) framework is designed to replace the traditional multiple time-scales framework. The ESGC allows TSEPs to enter and exit at any time and improves the economy of economic dispatch. Furthermore, an expandable depth and width adaptive dynamic programming (EDP) algorithm is proposed. The EDP can match the entering and exiting characteristics of TSEPs, and has high algorithm accuracy and calculation speed with reducing the network redundancy of the EDP. Finally, compared with compared algorithms, the EDP reduces 69.559% and 61.2% in |Df| and improves 1.783% and 0.668% in profit under a two-area smart grid and a modified 1888-bus French based smart grid, respectively. The simulation results verify the feasibility and reliability of the EDP for the ESGC of smart grids. (c) 2021 Elsevier Ltd. All rights reserved.

关键词： Expandable depth and width adaptive dynamic programming Economic smart generation control Three-state energy prosumer deep neural networks adaptive dynamic programming

来源：评论

学校读者我要写书评

暂无评论

adaptive Neural-Network Optimal Tracking Control for Permanent-Magnet Synchronous Motor Drive System via adaptive dynamic programming

Adaptive Neural-Network Optimal Tracking Control for Permane...

引用

IEEE-Industry-Applications-Society Annual Meeting

作者： El-Sousy, Fayez F. M. Amin, Mahmoud M. Aziz, Ghada A. Abdel Al-Durra, Ahmed Prince Sattam Bin Abdulaziz Univ Elect Engn Dept Al Kharj Saudi Arabia Manhattan Coll Elect & Comp Engn Dept Riverdale NY 10471 USA Elect Res Inst Power Elect & Energy Convers Dept Cairo Egypt Khalifa Univ Adv Power & Energy Ctr EECS Dept Abu Dhabi U Arab Emirates

ISBN: (纸本)9781728171920

This paper presents an adaptive neural-network optimal tracking control (ANOTC) scheme for permanent-magnet synchronous motor (PMSM) servo drive with uncertain dynamics via adaptive dynamic programming (ADP). The proposed ANOTC scheme consists of an adaptive steady-state controller, an adaptive optimal feedback controller and a robust controller. The adaptive steady-state controller is designed for attaining the targeted tracking response during the steady-state. The adaptive optimal feedback controller is designed for stabilizing the dynamics of tracking error at the transient in an optimal manner. Accordingly, critic and actor neural-networks are employed for facilitating the online solution of the Hamilton-Jacobi-Bellman (HJB) equation for approximating the adaptive optimal control laws via ADP method. Further, the robust controller is developed for compensating the approximation errors of neural-network (NN). Based on Lyapunov approach, the closed-loop stability of the PMSM servo drive system is proved to demonstrate that the proposed ANOTC scheme can ensure the system state tracking the targeted trajectory effectively. The proposed ANOTC scheme validation is performed via experimental analysis. From the experimental validation results, the PMSM servo drive dynamic behavior using the proposed ANOTC scheme can attain the optimal control response regardless the compounded disturbances and parameter uncertainties.

关键词： Actor-critic NN adaptive control adaptive dynamic programming optimal tracking control PMSM robust control Hamilton-Jacobi-Bellman

来源：评论

学校读者我要写书评

暂无评论

adaptive dynamic programming-based Decentralized Sliding Mode Optimal Control for Modular and Reconfigurable Robots

Adaptive Dynamic Programming-based Decentralized Sliding Mod...

引用

IEEE Symposium Series on Computational Intelligence (SSCI)

作者： An, Tianjiao Men, Xiaodong Zhang, Xinyao Li, Yuanchun Dong, Bo Changchun Univ Technol Dept Control Sci & Engn Changchun Peoples R China Northeastern Univ Res Ctr Mech Kinet & Reliabil Qinhuangdao Hebei Peoples R China

ISBN: (纸本)9781728124858

A model-free decentralized sliding mode control (SMC) is proposed via adaptive dynamic programming (ADP) algorithm to solve the problem of optimal tracking control of modular and reconfigurable robots (MRRs) in this paper. The dynamic formulation of MRR is expressed by a synthesis of joint subsystems with interconnected dynamic couplings (IDCs). Based on SMC technique, the optimal control of robotic system is transformed into an optimal compensation problem of unknown dynamics of each subsystem and a neural network (NN) identifier is set up to approximate IDC dynamics. Based on ADP and policy iteration (PI) method, the Hamilton-Jacobi-Bellman (HJB) equation can be addressed by using the critic NN and the optimal control policy can be obtained. The closed-loop robotic system is proved to be asymptotic stable by using the Lyapunov theory. Finally, simulation results are provided to demonstrate the effectiveness of the method.

关键词： Optimal control Decentralized control adaptive dynamic programming Modular and reconfigurable robot

来源：评论

学校读者我要写书评

暂无评论

Learning-Based Control of Multiple Connected Vehicles in the Mixed Traffic by adaptive dynamic programming ⁎

引用

IFAC-PapersOnLine 2021年第14期54卷 370-375页

作者： Tong Liu Leilei Cui Bo Pang Zhong-Ping Jiang New York University Brooklyn NY 11201 USA

The emergence of connected and autonomous vehicles (CAVs) has increased opportunities to mitigate the traffic congestion, improve safety and reduce accidents. In this paper, we consider the mixed traffic case with multiple heterogeneous human-driven vehicles and multiple CAVs on freeways. Under mild conditions, the stabilizability of the overall system is proved. With the tracking errors of relative distance and velocity as the states, we design an input-to-state stabilizing controller that solves a linear quadratic regulator problem by means of reinforcement learning and adaptive dynamic programming techniques. The priori knowledge of the vehicle network model is not needed. For a string of connected human-driven and automated vehicles, we give the sufficient conditions to guarantee the general string stability. The proposed learning-based control methodology is validated by means of simulation results.

关键词： Connected Autonomous Vehicles Stabilizability String Stability adaptive dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Observer Based Event-triggered Fault Compensation Control for Nonlinear Systems via adaptive dynamic programming 10

Observer Based Event-triggered Fault Compensation Control fo...

引用

10th International Conference on Information Science and Technology (ICIST)

作者： Luo, Fangchao Zhao, Bo Liu, Derong Guangdong Univ Technol Sch Automat Guangzhou 510006 Peoples R China Beijing Normal Univ Sch Syst Sci Beijing 100875 Peoples R China

ISBN: (纸本)9781728155586

This paper develops an observer based event-triggered fault compensation control method for a class of nonlinear continuous-time systems with actuator failures via adaptive dynamic programming (ADP). The proposed control method incorporates an event-triggered optimal control policy and an adaptive observer-based compensator. Due to the excellent estimation performance of neural networks, the actuator failure is precisely estimated by the online learning observer. By combining the event-triggered mechanism with ADP, the eventtriggered optimal control policy is obtained by adopting the critic neural network to solve Hamilton-Jacobi-Bellman equation. The triggering condition is provided along with the stability analysis of the close-loop system. Finally, the simulation of single link robot arm system is presented to confirm the efficiency of the proposed fault compensation control method.

关键词： adaptive dynamic programming Event-triggered mechanism adaptive fault compensation Optimal control Neural network

来源：评论

学校读者我要写书评

暂无评论

Optimization control of UAVs based on self-learning adaptive dynamic programming 35

Optimization control of UAVs based on self-learning adaptive...

引用

35th Youth Academic Annual Conference of Chinese-Association-of-Automation (YAC)

作者： Ye, Shuai Zhou, Ying-Jiang Jiang, Guo-Ping Lin, Qiong Nanjing Univ Posts & Telecommun Dept Automat & Artificial Intelligence Nanjing 210023 Peoples R China

ISBN: (纸本)9781728176840

In UAVs, optimal control has attracted more and more attention. In this paper, a self-learning adaptive dynamic programming (ADP) architecture based reinforcement learning (RL) is proposed to obtain optimal control for UAVs. 1 Compared with the traditional ADP architecture including two networks, one is used to make policy, and the other is used to evaluate policy, We propose to add a third network to replace external reward signals, that is, the agent can acquire reward signals by itself and do not need to interact with the environment. The proposed self-learning ADP method can improve the control performance by online learning while ensuring the state of the system stable at the equilibrium point. Finally, the proposed control algorithm is applied to quadrotor UAVs, and the experimental results show that the effectiveness of the algorithm.

关键词： adaptive dynamic programming Self-learning adaptive dynamic programming Reinforcement learning UAVs Online learning

来源：评论

学校读者我要写书评

暂无评论

Event-triggered Multi-agent Optimal Regulation Using adaptive dynamic programming

Event-triggered Multi-agent Optimal Regulation Using Adaptiv...

引用

International Joint Conference on Neural Networks (IJCNN) held as part of the IEEE World Congress on Computational Intelligence (IEEE WCCI)

作者： Zhong, Xiangnan He, Haibo Florida Atlantic Univ Dept Comp & Elect Engn & Comp Sci Boca Raton FL 33431 USA Univ Rhode Isl Dept Elect Comp & Biomed Engn Kingston RI 02881 USA

ISBN: (纸本)9781728169262

This paper develops an event-triggered multi-agent control method based on adaptive dynamic programming (ADP) techniques. Different from the traditional ADP-based multi agent control with fixed sampling period, our method designs an adaptive controller only based on the efficiently reduced samples. The sampling instants are decided by an adaptive triggering condition to guarantee the stability of the event-triggered learning process. The theoretical analysis of the proposed method is also provided in this paper. It is proved that the designed event triggered ADP controller can make all the agents synchronize to the leader's dynamics with reduced sampled data, and also reach Nash equilibrium at the same time. Therefore, the proposed method can save the computational resources in the learning process. Finally, the simulation results verify the theoretical analysis and also demonstrate the performance of the developed method.

关键词： Event-triggered control adaptive dynamic programming multi-agent systems online learning

来源：评论

学校读者我要写书评

暂无评论

Knowledge-shareable adaptive deep dynamic programming for hierarchical generation control of distributed high-percentage renewable energy systems

引用

RENEWABLE ENERGY 2024年 228卷

作者： Zhao, Lulin Yin, Linfei Guangxi Sci & Technol Normal Univ Sch Mech & Elect Engn Laibin 546199 Guangxi Peoples R China Guangxi Univ Guangxi Key Lab Power Syst Optimizat & Energy Tec Nanning 530004 Guangxi Peoples R China

The increased proportion of distributed renewable energy sources connected to systems leads to frequency regulation difficulty, such as uncoordinated algorithms, inconsistent optimization and control objectives, and long training time. This work proposes a knowledge-shareable adaptive deep dynamic programming (KSADDP) to improve the active control accuracy and generation economy of the proposed hierarchical generation control framework, which contains two-layer control structures. In the first layer, droop control is employed to optimize the outputs of controllable generators, follow the frequency changes, reduce frequency deviation, and improve the operation economy of battery energy storage systems. In the second layer, a KSADDP is proposed to mitigate local optimum and accelerate the hybrid cooperative structure optimization by sharing the parameters of neural networks. These two layers synergistically optimize control performance and reduce generation costs. The simulation results of a three-high-percentage renewable energy system show that: with two layers of cooperation and sharing training information, the KSADDP outperforms the comparison algorithms in seven evaluation indices and reduces the generation cost by 4.85%-7.41 % compared to the comparison algorithms. The proposed KSADDP cooperates the time scales between multiple layers, cooperates the objectives between multiple layers, increases the parameter-sharing process, and reduces training time.

关键词： Automatic generation control Deep neural networks adaptive dynamic programming Knowledge -shareable method Distributed high -percentage renewable energy systems

来源：评论

学校读者我要写书评

暂无评论

adaptive dynamic programming-Based Sliding Mode Optimal Position-Force Control for Reconfigurable Manipulators with Uncertain Disturbance

Adaptive Dynamic Programming-Based Sliding Mode Optimal Posi...

引用

第32届中国控制与决策会议

作者： Xinye Zhu Bing Ma Bo Dong Keping Liu Yuanchun Li Department of Control Science and Engineering Changchun University of Technology

In this paper, a sliding mode optimal position-force control method is proposed for constrained reconfigurable manipulators with disturbance based on adaptive dynamic programming(ADP) method and policy iteration(PI) algorithm. The sliding mode control is developed to compensate the model uncertainties, and the improved performance index function includes the sliding mode function and disturbance function. Then, the position-force control problem of the reconfigurable manipulator system under disturbance is transformed into an optimal control issue. The solution of Hamiltonian-Jacobi-Bellman(HJB) equation can be solved by using ADP and PI methods, and then the approximated sliding mode optimal control policy can be derived by constructing the critic neural network(NN). The closed-loop robotic system is proved to be asymptotic stability by using Lyapunov theory. Finally, simulations are provided to demonstrate the effectiveness of the proposed method.

关键词： Reconfigurable manipulators adaptive dynamic programming Sliding mode optimal position-force control Policy iteration algorithm Critic neural network

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：