检索结果-内蒙古大学图书馆

Run Time Assured Reinforcement Learning for Safe Satellite Docking

JOURNAL OF AEROSPACE INFORMATION SYSTEMS 2023年第1期20卷 25-36页

作者： Dunlap, Kyle Mote, Mark Delsing, Kaiden Hobbs, Kerianne L. Parallax Adv Res Beavercreek OH 45431 USA Pytheia Corp Atlanta GA 30308 USA Cedarville Univ Cedarville OH 45314 USA US Air Force Res Lab Dayton OH 45431 USA

Reinforcement learning promises high performance in complex tasks as well as low online storage and computation cost. However, the trial-and-error learning approach of reinforcement learning could explore unsafe behavior in the search for an optimal solution. Run time assurance (RTA) approaches can be applied to monitor behavior and ensure safety constraint satisfaction during reinforcement learning. This paper investigates the effect of RTA on reinforcement learning training performance in terms of training efficiency, safety constraint satisfaction, control efficiency, task efficiency, and training duration. For the purposes of demonstration, a custom reinforcement learning environment is created where the objective is to develop a policy that moves a satellite into docking position with another satellite in a two-dimensional relative-motion reference frame. Six different policies are trained. The first features no RTA, the second features no RTA but a higher penalty for safety violations, and four others use different RTA techniques to enforce a dynamic velocity constraint during training. The trained policies are analyzed with standardized test points. It is shown that the policies trained without RTA do not learn to adhere to the constraint, whereas all policies trained with RTA do learn to adhere to the constraint. Although more complex RTA frameworks can be better for operational use, it is found that a simple RTA framework provides the best overall results for reinforcement learning training.

关键词： Reinforcement Learning Satellites Neural Networks Spacecraft Docking optimization algorithm Closed Loop

来源：评论

学校读者我要写书评

暂无评论

Optimal Robust Two-Body Trajectory Design with Corrective Maneuvers

引用

JOURNAL OF SPACECRAFT AND ROCKETS 2023年第6期60卷 1966-1982页

作者： Kelly, Scott Geller, David Utah State Univ Mech & Aerosp Engn Dept Logan UT 84322 USA

The focus of this research is the optimization of impulsive transfer trajectories in the presence of stochastic effects. A trajectory design method is introduced that accounts for a normally distributed initial state dispersion and minimizes the sum of nominal impulsive & UDelta;V plus 3s trajectory correction maneuver (TCM) magnitude. Four main problems are presented. First, a deterministic optimal trajectory is developed;in the coplanar cases, the optimal solution is Hohmann transfer. For the second problem, an initial state dispersion and a target position dispersion constraint are introduced. It is shown to be possible to modify the nominal two-impulse trajectory to satisfy the dispersion constraint. In the third problem, a TCM is performed at the optimal point along the deterministic optimal trajectory, resulting in a more efficient method to influence the target position dispersion. Problem 4 is the development of a robust trajectory, where the nominal impulsive maneuvers and 3s TCM dV are simultaneously optimized. The result is a different nominal trajectory and TCM that is less expensive than the total cost of problem 3. The optimal TCM is rapidly computed along each nominal trajectory using the numerically propagated state transition matrix history as a step inside the optimization algorithm.

关键词： Spacecraft Design Orbital Maneuvers optimization algorithm Optimal Control Problem Geostationary Transfer Orbit Sun Synchronous Orbit Stochastic Optimal Control Robust optimization Gradient-Based Design optimization State-Transition Equation

来源：评论

学校读者我要写书评

暂无评论

Facial emotion recognition on video using deep attention based bidirectional LSTM with equilibrium optimizer

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2023年第19期82卷 28681-28711页

作者： Vedantham, Ramachandran Reddy, Edara Sreenivasa Vasireddy Venkatadri Inst Technol CSE Dept Nambur Andhra Pradesh India Acharya Nagarjuna Univ ANU Coll Engn & Technol Nambur Andhra Pradesh India

Facial emotion recognition (FER) from videos is now considered a significant role in HCI (Human-Computer Interaction). The dynamic variations shown by various facial movements need to be realized quickly without degrading the recognition performance. Therefore, the procedure of classifying the facial emotions from videos is now demanded as a challenging and interesting issue. This work proposes a practical methodology for identifying emotions through facial expressions from videos. At first, the Lucas-Kanade (LK) based optical flow scheme is used for motion detection from the input videos. After finding the set of LK frames, the pre-processing scheme is applied. In this process, the Viola-Jones algorithm is utilized for face detection, and then the gray scale conversion is involved. Moreover, the FAST corner detection approach is used to detect the facial landmark points over the gray scale frame. The Neighborhood Difference Features (NDF) are extracted in feature extraction (FE). The optimal set of features is selected from the mined features using the Modified Plant Genetics-Inspired Evolutionary optimization (MPGEO) algorithm in the feature selection (FS). Finally, the chosen features are fed into the Deep Attention-based Bidirectional LSTM with Equilibrium Optimizer (DABLEO) classifier for the emotion classification. The proposed scheme is performed in the Python software using four standard datasets like FAMED, CK+, AFEW, and MMI, and it delivers the classification accuracy of 96.5%, 99.2%, 90%, and 92% individually. As related to other schemes, the proposed scheme is better for all performances of emotion recognition.

关键词： Emotion recognition optimization algorithm Deep network Optical flow Feature extraction And selection

来源：评论

学校读者我要写书评

暂无评论

An optimal internal model proportional integral controller to improve pressure tracking profile of artificial ventilator

引用

MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING 2023年第7期61卷 1807-1820页

作者： Acharya, Debasis Das, Dushmanta Kumar Natl Inst Technol Nagaland Chumukedima India

Mechanical ventilator (MV) is a breathing support medical equipment used during medical surgery or in an intensive care unit (ICU) for critical patients to reduce the breathing work of the patient under ventilation. In this work, a piston-driven mechanical ventilator is studied. Such types of ventilators are generally used as anaesthesia ventilators with volume-controlled modes. A modern ventilator can operate in different modes, such as volume control, pressure control, or both. Pressure-controlled mode may be better suited for some cases, such as patients undergoing laparoscopic surgery, as compared to volume-controlled mode because it may increase compliance during pneumoperitoneum, enhance oxygenation, and lower the stress response postoperatively. The pressure profile of a PCV must exactly track the pressure profile of a patient under ventilation to avoid ventilator-induced diaphragmatic dysfunction. Therefore, an optimal internal model control base proportional integral (IMC-PI) controller is proposed. The optimum parameter of the filter component of the IMC-PI controller is found by an enhanced class topper optimization (ECTO) algorithm. The performance of the proposed controller is analyzed in different scenarios and compared with existing results.

关键词： Artificial ventilation system (AVS) optimization algorithm Pressure control ventilator (PCV) Positive end-expiratory pressure (PEEP) Internal model control (IMC)

来源：评论

学校读者我要写书评

暂无评论

Sliding Mode Control of FMII Systems: Handling Rice Fading Issues Under 2-D Frame

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2023年第11期53卷 6909-6920页

作者： Lv, Xinyu Niu, Yugang Park, Ju H. East China Univ Sci & Technol Minist Educ Key Lab Smart Mfg Energy Chem Proc Shanghai 200237 Peoples R China Yeungnam Univ Minist Educ Dept Elect Engn Gyongsan 38541 South Korea

The Lth order Rice model can effectively describe imperfect transmission, including amplitude attenuation, time delays, and stochastic disturbance, and is a popular channel fading model in one-dimensional (1-D) systems. However, due to the state evolution feature of two-dimensional (2-D) systems, the previous state signals of 2-D systems are not as directly determinable as those of 1-D systems such that the construction of the conventional and obvious Lth order Rice model become difficulty for 2-D systems. Motivated by these, this work will investigate the sliding mode control (SMC) problem of the 2-D Fornasini-Marchesini (FMII) system under Rice fading channels. First, the previous p-step state signals are defined according to the evolution feature of the FMII model. Afterwards, a new applicable Lth-order Rice fading model for 2-D FMII systems is constructed, which not only has 2-D structural characteristics but also has similar properties to 1-D cases. A 2-D-type sliding function involving the predefined previous L-step state signals and channel attenuation coefficients is designed, which exhibits a similar structural feature to the FMII model. A feasible SMC law is designed via the available fading states to achieve the ultimately uniformly boundedness in the mean square of the closed-loop system and the reachability of the sliding domain around the specified sliding surface. Furthermore, an optimization-solving algorithm is introduced to minimize the convergent bound. Finally, an example is applied to demonstrated the proposed results.

关键词： optimization algorithm previous state signals rice fading model sliding mode control (SMC) two-dimensional (2-D) Fornasini-Marchesini (FMII) system

来源：评论

学校读者我要写书评

暂无评论

An Effective Spectral Processing Method for Low-Cost Fiber Loop Ring-Down Systems

引用

ADVANCED THEORY AND SIMULATIONS 2023年第5期6卷

作者： Cha, Guozhi Wang, Ruoyu Ma, Jiangtao Shao, Guifang Liu, Tundong Xiamen Univ Sch Aerosp Engn Xiamen 361102 Peoples R China Dali Univ Sch Engn Dali 671003 Peoples R China Xiamen Univ Xiamen Key Lab Big Data Intelligent Anal & Decis Xiamen 361102 Peoples R China

Fiber loop ring-down spectroscopy (FLRDS) is a highly sensitive spectroscopic technology. Here, a spectral processing method is proposed for low-cost fiber loop ring-down systems. In the proposed method, the amplitude modulation theory is used to establish a new ring-down spectral model, and the particle swarm optimization algorithm is introduced into FLRDS. The experimental results demonstrate that the proposed method can effectively process the ring-down spectral signals acquired from the low-cost fiber loop ring-down system. Moreover, this method demonstrates higher accuracy and stronger adaptability than the current widely used method. Owing to its excellent performance, the proposed method has considerable potential to be applied in low-cost fiber loop ring-down systems.

关键词： amplitude modulation fiber optics optimization algorithm spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Optimal Torque Sharing Function Control for Switched Reluctance Motors Based on Active Disturbance Rejection Controller

引用

IEEE-ASME TRANSACTIONS ON MECHATRONICS 2023年第5期28卷 2600-2608页

作者： Feng, Liyun Sun, Xiaodong Yang, Zebin Diao, Kaikai Jiangsu Univ Automot Engn Res Inst Zhenjiang 212013 Peoples R China Jiangsu Univ Sch Elect & Informat Engn Zhenjiang 212013 Peoples R China

In this article, an optimal torque sharing function (TSF) control method incorporated with improved linear active disturbance rejection control (LADRC) and modified coyote optimization algorithm (MCOA) is introduced and applied in switched reluctance motors (SRMs). First, the piecewise TSF is employed to reduce the torque ripple, which exhibits little effects on the rate of current change and the current peak value. The improved linear extended state observer with the variable gain in speed controller using LADRC is designed to guarantee better antidisturbance ability and torque ripple reduction simultaneously. Then, the MCOA is introduced to automatically tune the parameters including the turn-on angle, conduction angle, and coefficients for better comprehensive performances. Finally, comparisons among the conventional TSF with LADRC, the piecewise TSF with LADRC, and the proposed TSF with improved LADRC are conducted both in the simulations and experiments. The results indicate that the proposed method behaves best in aspects of torque ripple suppression and anti-interference ability among the three methods.

关键词： Active disturbance rejection control (ADRC) optimization algorithm switched reluctance motor (SRM) torque sharing function (TSF)

来源：评论

学校读者我要写书评

暂无评论

Web information mining and semantic analysis in heterogeneous unstructured text data using enhanced latent Dirichlet allocation

引用

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE 2023年第1期35卷 e7410-e7410页

作者： Venugopal, Madamanchi Sharma, Virendra K. Sharma, Kalpana Bhagwant Univ Dept CSE Ajmer 305004 Rajasthan India Bhagwant Univ Dept EEE Ajmer Rajasthan India

Information mining and semantic analysis have gained significant attention over recent years to obtain appropriate information from unstructured data. Several approaches have been introduced for web information mining. However, the expected accuracy is not reached by these approaches. Therefore, hybrid fuzzy clustering and enhanced latent Dirichlet allocation (ELDA) are proposed for the accuracy increment in this work. The information clustering process is performed using the hybrid fuzzy clustering algorithm called fuzzy-C-medoids optimized using improved whale algorithm. The clustering procedure entails grouping data points into more similar clusters than data points from other clusters. Finally, the context of the text is recognized by analyzing the semantic information with ELDA, which offers a suitable index for accurate and fast data extraction. PYTHON tool is used to develop the proposed web mining model, and the simulation analysis of the proposed model is carried out using the BibTex dataset and compared with baseline models. The performance of the proposed method is evaluated using purity, normalized mutual information, accuracy, and precision metrics. Also, it is compared with different existing algorithms. Thus, the analysis of the results proved that the proposed approach achieves better outcomes than the existing approaches.

关键词： clustering optimization algorithm preprocessing semantic analysis text mining web mining

来源：评论

学校读者我要写书评

暂无评论

Application of an intelligent method for hydrogen-based energy hub in multiple energy markets

引用

INTERNATIONAL JOURNAL OF HYDROGEN ENERGY 2023年第93期48卷 36485-36499页

作者： Li, Ziyuan He, Tao Farjam, Hashem Xiamen Inst Technol Business Sch Xiamen 361000 Fujian Peoples R China Wenzhou Vocat & Tech Coll Sch Intelligent Mfg Wenzhou 325000 Zhejiang Peoples R China Sun Life Co EED Baku Azerbaijan

Use of different energy carriers together, known as an energy hub, has been a hot topic of research. The hub manager manages the energy in the hub and supplies the energy demanded by the consumers. Herein, a multi-carrier energy storage system comprising a thermal energy storage system, ice storage system, and hydrogen storage system is considered for a multi-carrier energy system to exploit their economic benefits. The proposed combined cooling, heating and electricity microgrid not only participates in the electricity, gas and heat market to meet the needs of electricity, heating and cooling, but also can use the new hydrogen power technology used in the hydrogen storage system to increase the efficiency of the whole system participate in the market. In addition, a multienergy demand response model is used as a new concept of demand response on electrical and thermal loads, which provides more options for multi-energy end users in energy management policies. To cover the uncertainty of the existence of different energy sources in the used model and the price of electricity, a new method has been used. The LHS sampling is used to develop diverse scenarios, and the backward scenario reduction method is adopted to decrease the number of scenarios. To solve the objective function and optimization, the metaheuristic whale algorithm improved by a local search algorithm is utilized. The main goal is to reduce the operation costs of the multi-energy carrier system. The obtained results show that the hydrogen storage utilization and multi-energy demand response can reduce the system operation cost by 6% for the studied experimental system. Furthermore, as the uncertainty budget increased, the total cost of operations increased by 10%. This is because the level of planning conservatism increases with each increase in the uncertainty budget. (c) 2023 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

关键词： Energy hub optimization algorithm Demand response Multi -carrier energy

来源：评论

学校读者我要写书评

暂无评论

Large eddy simulation on effects of trench geometry on film cooling effectiveness through fan-shaped holes

引用

INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER 2023年第1期216卷

作者： Zamiri, Ali Abdeh, Hamed Barigozzi, Giovanna Chung, Jin Taek Korea Univ Sch Mech Engn Seoul South Korea Univ Bergamo Dept Engn & Appl Sci Dalmine Italy

Modern gas turbines require the use of external film-cooling techniques to effectively cool the turbine blades, which are heated by hot combustion gases. Large eddy simulation (LES) is conducted to numerically evaluate the influences of trench configurations on film-cooling effectiveness and flow-field behavior of a shaped hole. The cooling hole was a laidback fan-shaped hole with a 17.5-degree forward expansion angle located on a flat plate surface with a 30-degree injection angle at a constant blowing ratio of 2.0 and a constant density ratio of 1.5. The computed film-cooling effectiveness was validated with that obtained in the experimental data for the reference case at two blowing ratios. Two geometrical parameters of the trench height and width were selected as design variables, and overall area-averaged film-cooling effectiveness was considered as an objective function. The Latin Hypercube sampling (LHS) method was applied to generate the designed cases and two different optimization algorithms, the response surface methodology (RSM) and the Kriging method, were employed to maximize the objective function. The computational LES results revealed that trenched cases with a larger trench height and a relatively smaller trench width improved cooling performance. The optimal trench configurations based on the RSM and Kriging method, demonstrated improvements in cooling performance by 27% and 25%, respectively, compared to the reference case without a trench slot. The transient analysis also showed that the flow unsteadiness and velocity disturbances near the trench exit at the mixing region were weakened for the cooling holes with the optimal trench cases.

关键词： Film-cooling effectiveness Fan-shaped cooling hole optimization algorithm Large eddy simulation

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：