For the charging station construction of electric vehicle,location selecting is a key issue.There are two problems in location selection of the electric vehicle charging station.One is determining the location of char...For the charging station construction of electric vehicle,location selecting is a key issue.There are two problems in location selection of the electric vehicle charging station.One is determining the location of charging station;the other is evaluating the location of charging station.To determine the charging station location,an spatial clustering algorithm is proposed and programmed.The example simulation shows the effectiveness of the spatial clustering algorithm.To evaluate the charging station location,a multi-hierarchical fuzzy method is proposed.Based on the location factors of electric vehicle charging station,the hierarchical evaluation structure of electric vehicle charging station location is constructed,including three levels,4first-class factors and 14second-class factors.The fuzzy multi-hierarchical evaluation model and algorithm are built.The analysis results show that the multi-hierarchical fuzzy method can reasonably complete the electric vehicle charging station location evaluation.展开更多
A low-complexity distributed power allocation algorithm is proposed to reduce the interference and improve the transmitting rate of edge users. Different scenarios are considered and user experience of indoor communic...A low-complexity distributed power allocation algorithm is proposed to reduce the interference and improve the transmitting rate of edge users. Different scenarios are considered and user experience of indoor communication is promoted. The simulation results prove the effectiveness of our algorithm. The proposed power control scheme ensures that more users can achieve their required rate and the fairness of different users is improved. Besides, more than 5096 energy can be saved without loss in outage ability, and energy efficiency is also promoted. In addition, the proposed algorithm can be extended to scenarios that the required rates of pico stations can be changed periodically.展开更多
Tribological tests play an important role on the evaluation of long-term bio-tribological performances of prosthetic materials for commercial fabrication.Those tests focus on the motion simulation of a real joint in v...Tribological tests play an important role on the evaluation of long-term bio-tribological performances of prosthetic materials for commercial fabrication.Those tests focus on the motion simulation of a real joint in vitro with only normal loads and constant velocities,which are far from the real friction behavior of human joints characterized with variable loads and multiple directions.In order to accurately obtain the bio-tribological performances of artificial joint materials,a tribological tester with a miniature four-station tribological system is proposed with four distinctive features.Firstly,comparability and repeatability of a test are ensured by four equal stations of the tester.Secondly,cross-linked scratch between tribo-pairs of human joints can be simulated by using a gear-rack meshing mechanism to produce composite motions.With this mechanism,the friction tracks can be designed by varying reciprocating and rotating speeds.Thirdly,variable loading system is realized by using a ball-screw mechanism driven by a stepper motor,by which loads under different gaits during walking are simulated.Fourthly,dynamic friction force and normal load can be measured simultaneously.The verifications of the performances of the developed tester show that the variable frictional tracks can produce different wear debris compared with one-directional tracks,and the accuracy of loading and friction force is within ？5%.Thus the high consistency among different stations can be obtained.Practically,the proposed tester system could provide more comprehensive and accurate bio-tribological evaluations for prosthetic materials.展开更多
Nowadays, the Multi-Shuttle and Multi-Station Transportation System (MMTS)<span><span><span style="font-family:;" "=""> is one of the most interesting research topics in many...Nowadays, the Multi-Shuttle and Multi-Station Transportation System (MMTS)<span><span><span style="font-family:;" "=""> is one of the most interesting research topics in many fields of industries. It is an effective solution to reduce unexpected accidents that occur during transportation as well as increase productivity in manufacturing. The aim of this paper is to introduce the controller design for the MMTS which is built in our BK-Recme BioMech Lab at Ho Chi Minh City University of Technology (VNU-HCM), Viet Nam. Based on the design of this system, the control algorithms will be conducted to check the operation of the whole system. To evaluate the feasibility and effectiveness of this model, we design a series of random instances for different quantities of nodes as well as the different quantities of shuttles. Our system includes 4 stations and 6 shuttles which are assembled in the serial chain system. However, the number of stations and number of shuttles can be expanded to any desired ones which are based on the requirement of the industries. In this paper, we mainly focus on the controller design of this system to make it operate in an effective way that the goods will be transported and delivered to the target station as fast as possible. In order to solve the large</span></span></span><span><span><span style="font-family:;" "="">-</span></span></span><span><span><span style="font-family:;" "="">scale instances and realistic transport prob<span>lems, we propose three algorithms for three progresses as shuttles calling</span>, path reading and shuttles communicating. The shuttles calling is to decide which <span>shuttle should be called to the star</span></span></span></span><span><span><span style="font-family:;" "="">t</span></span></span><span><span><span style="font-family:;" "="">-node. Path reading to determine the shortest</span></span></span><span><span><span style="font-family:;" "=""> <span>way to go from start-node to end-node. Finally, shuttles communicating,</span> which allow one shuttle to interact with the next shuttles so we have a loop of orders (shuttle 1 to shuttle 2;shuttle 2 to shuttle 3;etc</span></span></span><span><span><span style="font-family:;" "="">.</span></span></span><span><span><span style="font-family:;" "="">;shuttle n-1 to shuttle n). This proposes solution can help us to solve the huge numbers of shuttles <span>and stations in the system. The specific result of this study is applying</span> Dijkstra’s algorithm to propose an algorithm that allows handling a transportation system without caring about the number of shuttles as well as the number of stations for the closed-loop path. Several test problems are carried out in order to check the feasibility and the effectiveness of our purposed control algorithm.</span></span></span>展开更多
大容量电池储能电站功率分配的决策变量多,且策略需考虑多个优化目标及能自动适应场景的不确定性。为此,提出了一种基于多智能体深度强化学习(multi-agent deep reinforcement learning,MADRL)的电池储能电站功率分配决策方法。首先,基...大容量电池储能电站功率分配的决策变量多,且策略需考虑多个优化目标及能自动适应场景的不确定性。为此,提出了一种基于多智能体深度强化学习(multi-agent deep reinforcement learning,MADRL)的电池储能电站功率分配决策方法。首先,基于大容量电池储能电站结构及其功率分配特性构建基于MADRL的功率分配决策框架,每个储能单元设置一个功率分配智能体,多个智能体构成合作关系;然后,设计考虑储能电站有功功率损耗、荷电状态(state of charge,SOC)一致性和健康状态损失最小优化目标的功率分配智能体模型,采用深度确定性策略梯度(deep deterministic policy gradient,DDPG)算法去中心化训练各智能体网络参数,算法收敛后得到储能子系统充放电功率值。最后,算例验证了所提方法的有效性,能在有效提高储能子系统SOC均衡性的同时降低有功功率损耗、健康状态损失和充放电切换次数。展开更多
国际月球科研站ILRS (International Lunar Research Station)的建立是中国顺应当前形势的重要太空战略部署。为研究基于多源数据的形貌分析在ILRS建设过程中的应用,本研究将“ILRS计划”分为选址和科考阶段,分析ILRS选址阶段需考虑的...国际月球科研站ILRS (International Lunar Research Station)的建立是中国顺应当前形势的重要太空战略部署。为研究基于多源数据的形貌分析在ILRS建设过程中的应用,本研究将“ILRS计划”分为选址和科考阶段,分析ILRS选址阶段需考虑的约束条件,简述正射影像、月表地形、微波辐射计、多/高光谱成像、测月雷达、伽马射线等多源数据在选址阶段的应用,利用嫦娥二号DOM、DEM、LRO Diviner辐射计、LOLA激光高度计以及地质制图等多源数据对月球南极地形特征、温度条件、光照条件、对地能见度和地质特征进行分析,提出在ILRS选址阶段应考虑的诸多因素和适合着陆选址的区域特点,并以Amundsen区域为例,选择了3个候选着陆点。最后,以Shackleton、Shoemaker、de Gerlache和Amundsen撞击坑为代表,基于多源数据分析结果为ILRS科考阶段巡视路线的规划、水冰探测的分析、观测基站的设立提供参考。综上,本研究基于多源数据的形貌分析结果对ILRS的选址和完成科学考察任务具有重要参考价值。展开更多
基金supported by the National Natural Science Foundation of China(No.51575047)
文摘For the charging station construction of electric vehicle,location selecting is a key issue.There are two problems in location selection of the electric vehicle charging station.One is determining the location of charging station;the other is evaluating the location of charging station.To determine the charging station location,an spatial clustering algorithm is proposed and programmed.The example simulation shows the effectiveness of the spatial clustering algorithm.To evaluate the charging station location,a multi-hierarchical fuzzy method is proposed.Based on the location factors of electric vehicle charging station,the hierarchical evaluation structure of electric vehicle charging station location is constructed,including three levels,4first-class factors and 14second-class factors.The fuzzy multi-hierarchical evaluation model and algorithm are built.The analysis results show that the multi-hierarchical fuzzy method can reasonably complete the electric vehicle charging station location evaluation.
基金Supported by National S&T Major Program of China(2013ZX03003002-003)
文摘A low-complexity distributed power allocation algorithm is proposed to reduce the interference and improve the transmitting rate of edge users. Different scenarios are considered and user experience of indoor communication is promoted. The simulation results prove the effectiveness of our algorithm. The proposed power control scheme ensures that more users can achieve their required rate and the fairness of different users is improved. Besides, more than 5096 energy can be saved without loss in outage ability, and energy efficiency is also promoted. In addition, the proposed algorithm can be extended to scenarios that the required rates of pico stations can be changed periodically.
基金Supported by Funding of State Key Laboratory of Mechanical Transmissions,Chongqing University,China(Grant No.SKLMT-KFKT-201504)National Natural Science Foundation of China(Grant No.51275381)Science and Technology Planning Project of Shaanxi Province,China(Grant No.2012GY2-37)
文摘Tribological tests play an important role on the evaluation of long-term bio-tribological performances of prosthetic materials for commercial fabrication.Those tests focus on the motion simulation of a real joint in vitro with only normal loads and constant velocities,which are far from the real friction behavior of human joints characterized with variable loads and multiple directions.In order to accurately obtain the bio-tribological performances of artificial joint materials,a tribological tester with a miniature four-station tribological system is proposed with four distinctive features.Firstly,comparability and repeatability of a test are ensured by four equal stations of the tester.Secondly,cross-linked scratch between tribo-pairs of human joints can be simulated by using a gear-rack meshing mechanism to produce composite motions.With this mechanism,the friction tracks can be designed by varying reciprocating and rotating speeds.Thirdly,variable loading system is realized by using a ball-screw mechanism driven by a stepper motor,by which loads under different gaits during walking are simulated.Fourthly,dynamic friction force and normal load can be measured simultaneously.The verifications of the performances of the developed tester show that the variable frictional tracks can produce different wear debris compared with one-directional tracks,and the accuracy of loading and friction force is within ？5%.Thus the high consistency among different stations can be obtained.Practically,the proposed tester system could provide more comprehensive and accurate bio-tribological evaluations for prosthetic materials.
文摘Nowadays, the Multi-Shuttle and Multi-Station Transportation System (MMTS)<span><span><span style="font-family:;" "=""> is one of the most interesting research topics in many fields of industries. It is an effective solution to reduce unexpected accidents that occur during transportation as well as increase productivity in manufacturing. The aim of this paper is to introduce the controller design for the MMTS which is built in our BK-Recme BioMech Lab at Ho Chi Minh City University of Technology (VNU-HCM), Viet Nam. Based on the design of this system, the control algorithms will be conducted to check the operation of the whole system. To evaluate the feasibility and effectiveness of this model, we design a series of random instances for different quantities of nodes as well as the different quantities of shuttles. Our system includes 4 stations and 6 shuttles which are assembled in the serial chain system. However, the number of stations and number of shuttles can be expanded to any desired ones which are based on the requirement of the industries. In this paper, we mainly focus on the controller design of this system to make it operate in an effective way that the goods will be transported and delivered to the target station as fast as possible. In order to solve the large</span></span></span><span><span><span style="font-family:;" "="">-</span></span></span><span><span><span style="font-family:;" "="">scale instances and realistic transport prob<span>lems, we propose three algorithms for three progresses as shuttles calling</span>, path reading and shuttles communicating. The shuttles calling is to decide which <span>shuttle should be called to the star</span></span></span></span><span><span><span style="font-family:;" "="">t</span></span></span><span><span><span style="font-family:;" "="">-node. Path reading to determine the shortest</span></span></span><span><span><span style="font-family:;" "=""> <span>way to go from start-node to end-node. Finally, shuttles communicating,</span> which allow one shuttle to interact with the next shuttles so we have a loop of orders (shuttle 1 to shuttle 2;shuttle 2 to shuttle 3;etc</span></span></span><span><span><span style="font-family:;" "="">.</span></span></span><span><span><span style="font-family:;" "="">;shuttle n-1 to shuttle n). This proposes solution can help us to solve the huge numbers of shuttles <span>and stations in the system. The specific result of this study is applying</span> Dijkstra’s algorithm to propose an algorithm that allows handling a transportation system without caring about the number of shuttles as well as the number of stations for the closed-loop path. Several test problems are carried out in order to check the feasibility and the effectiveness of our purposed control algorithm.</span></span></span>
文摘大容量电池储能电站功率分配的决策变量多,且策略需考虑多个优化目标及能自动适应场景的不确定性。为此,提出了一种基于多智能体深度强化学习(multi-agent deep reinforcement learning,MADRL)的电池储能电站功率分配决策方法。首先,基于大容量电池储能电站结构及其功率分配特性构建基于MADRL的功率分配决策框架,每个储能单元设置一个功率分配智能体,多个智能体构成合作关系;然后,设计考虑储能电站有功功率损耗、荷电状态(state of charge,SOC)一致性和健康状态损失最小优化目标的功率分配智能体模型,采用深度确定性策略梯度(deep deterministic policy gradient,DDPG)算法去中心化训练各智能体网络参数,算法收敛后得到储能子系统充放电功率值。最后,算例验证了所提方法的有效性,能在有效提高储能子系统SOC均衡性的同时降低有功功率损耗、健康状态损失和充放电切换次数。
文摘国际月球科研站ILRS (International Lunar Research Station)的建立是中国顺应当前形势的重要太空战略部署。为研究基于多源数据的形貌分析在ILRS建设过程中的应用,本研究将“ILRS计划”分为选址和科考阶段,分析ILRS选址阶段需考虑的约束条件,简述正射影像、月表地形、微波辐射计、多/高光谱成像、测月雷达、伽马射线等多源数据在选址阶段的应用,利用嫦娥二号DOM、DEM、LRO Diviner辐射计、LOLA激光高度计以及地质制图等多源数据对月球南极地形特征、温度条件、光照条件、对地能见度和地质特征进行分析,提出在ILRS选址阶段应考虑的诸多因素和适合着陆选址的区域特点,并以Amundsen区域为例,选择了3个候选着陆点。最后,以Shackleton、Shoemaker、de Gerlache和Amundsen撞击坑为代表,基于多源数据分析结果为ILRS科考阶段巡视路线的规划、水冰探测的分析、观测基站的设立提供参考。综上,本研究基于多源数据的形貌分析结果对ILRS的选址和完成科学考察任务具有重要参考价值。
