http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Li Cheng,Yu Ren,Yu Wenmin,Wang Tianshu 한국원자력학회 2022 Nuclear Engineering and Technology Vol.54 No.9
Based on the Deep Q-Network(DQN) algorithm of reinforcement learning, an active fault-tolerance method with incremental action is proposed for the control system with sensor faults of the oncethrough steam generator(OTSG). In this paper, we first establish the OTSG model as the interaction environment for the agent of reinforcement learning. The reinforcement learning agent chooses an action according to the system state obtained by the pressure sensor, the incremental action can gradually approach the optimal strategy for the current fault, and then the agent updates the network by different rewards obtained in the interaction process. In this way, we can transform the active fault tolerant control process of the OTSG to the reinforcement learning agent's decision-making process. The comparison experiments compared with the traditional reinforcement learning algorithm(RL) with fixed strategies show that the active fault-tolerant controller designed in this paper can accurately and rapidly control under sensor faults so that the pressure of the OTSG can be stabilized near the set-point value, and the OTSG can run normally and stably
Reinforcement learning-based control with application to the once-through steam generator system
Li Cheng,Yu Ren,Yu Wenmin,Wang Tianshu 한국원자력학회 2023 Nuclear Engineering and Technology Vol.55 No.10
A reinforcement learning framework is proposed for the control problem of outlet steam pressure of the once-through steam generator(OTSG) in this paper. The double-layer controller using Proximal Policy Optimization(PPO) algorithm is applied in the control structure of the OTSG. The PPO algorithm can train the neural networks continuously according to the process of interaction with the environment and then the trained controller can realize better control for the OTSG. Meanwhile, reinforcement learning has the characteristic of difficult application in real-world objects, this paper proposes an innovative pretraining method to solve this problem. The difficulty in the application of reinforcement learning lies in training. The optimal strategy of each step is summed up through trial and error, and the training cost is very high. In this paper, the LSTM model is adopted as the training environment for pretraining, which saves training time and improves efficiency. The experimental results show that this method can realize the self-adjustment of control parameters under various working conditions, and the control effect has the advantages of small overshoot, fast stabilization speed, and strong adaptive ability.
Simulating Strength Parameters and Size Effect of Stochastic Jointed Rock Mass using DEM Method
Chong Ma,Wenmin Yao,Yuan Yao,Jun Li 대한토목학회 2018 KSCE JOURNAL OF CIVIL ENGINEERING Vol.22 No.12
The strength parameters and the size effect of stochastic jointed limestone rock mass is investigated in this paper. Based on extensive statistics of joint parameters of rock mass in the research region, the probable distribution of geometric characteristic parameters of discontinuities are obtained by the probability graph method. Then the Monte-Carlo method is used for discontinuities network modeling. In addition, 3DEC software and its built-in FISH programming language are used to establish the stochastic jointed rock mass network model based on discrete element method. Triaxial numerical simulation tests under variable confining pressure are conducted with different model sizes and dip angles of bedding planes. The numerical simulation results indicate that the jointed rock mass exhibits weak anisotropy property and significant size effect when it is cut by stochastic discontinuities; the mechanical strength parameters of rock mass begins to fluctuate distinctly as the model size increases, and tend to be stable once the model size reaches or exceeds 4 m × 4 m × 8 m. Besides, the comprehensive mechanical parameters of rock mass in the research region are determined and failure modes of rock mass are analyzed as well based on the numerical simulation results.
A Fuzzy Identity-Based Signcryption Scheme from Lattices
( Xiuhua Lu ),( Qiaoyan Wen ),( Wenmin Li ),( Licheng Wang ),( Hua Zhang ) 한국인터넷정보학회 2014 KSII Transactions on Internet and Information Syst Vol.8 No.11
Fuzzy identity-based cryptography introduces the threshold structure into identity-based cryptography, changes the receiver of a ciphertext from exact one to dynamic many, makes a cryptographic scheme more efficient and flexible. In this paper, we propose the first fuzzy identity-based signcryption scheme in lattice-based cryptography. Firstly, we give a fuzzy identity-based signcryption scheme that is indistinguishable against chosen plaintext attack under selective identity model. Then we apply Fujisaki-Okamoto method to obtain a fuzzy identity-based signcryption scheme that is indistinguishable against adaptive chosen ciphertext attack under selective identity model. Thirdly, we prove our scheme is existentially unforgeable against chosen message attack under selective identity model. As far as we know, our scheme is the first fuzzy identity-based signcryption scheme that is secure even in the quantum environment.
Haikuan Zhang,Changdong Li,Wenmin Yao,Jingjing Long 대한토목학회 2019 KSCE JOURNAL OF CIVIL ENGINEERING Vol.23 No.9
It is reported that there are many colluvial landslides with multilayered sliding masses; however, previous studies of the pile spacing of stabilizing piles mainly focus on the single-layered sliding mass, which may lead to design errors for pile spacing. Consequently, the paper presents a novel method to determine the pile spacing with considering interactions of multilayered sliding masses in colluvial landslides. Based on a generalized landslide model, equations for calculating stability coefficients of multilayered sliding masses were improved by examining the interactions among sliding masses. An accordingly colluvial landslide model with double-layered sliding masses was established by the finite differential method. The distribution of vertical landslide driving force and horizontal loading between adjacent piles were studied based on the colluvial landslide. A novel method of calculating the maximum pile spacing and minimum pile spacing was deduced by considering the soil arching effect and the interactions among multilayered sliding masses. The reasonable pile spacing was obtained considering cost and performance of stabilizing piles. The calculational process, which determines optimal pile spacing in multilayered masses, were shown based on the Bazimen landslide. The variations in pile spacing affected by various soil-layer sequences was illustrated by employing the Bazimen landslide model. The calculation results indicate that the pile spacing is positively correlated with the depth of soil with the maximum resistance sliding force. Effectiveness and significance of the presented method were proved through verify the calculational results by using numerical modeling approaches.
Identity Based Proxy Re-encryption Scheme under LWE
( Wei Yin ),( Qiaoyan Wen ),( Wenmin Li ),( Hua Zhang ),( Zheng Ping Jin ) 한국인터넷정보학회 2017 KSII Transactions on Internet and Information Syst Vol.11 No.12
The proxy re-encryption allows an intermediate proxy to convert a ciphertext for Alice into a ciphertext for Bob without seeing the original message and leaking out relevant information. Unlike many prior identity based proxy re-encryption schemes which are based on the number theoretic assumptions such as large integer factorization and discrete logarithm problem. In this paper, we first propose a novel identity based proxy re-encryption scheme which is based on the hardness of standard Learning With Error(LWE) problem and is CPA secure in the standard model. This scheme can be reduced to the worst-case lattice hard problem that is able to resist attacks from quantum algorithm. The key step in our construction is that the challenger how to answer the private query under a known trapdoor matrix. Our scheme enjoys properties of the non-interactivity, unidirectionality, anonymous and so on. In this paper, we utilize primitives include G-trapdoor for lattice and sample algorithms to realize simple and efficient re-encryption.
Wen, Min,Zheng, Jin Hai,Choi, Jin Myung,Pei, Jian,Li, Chun-Hao,Li, Song-Yuan,Kim, In-Young,Lim, Sa-Hoe,Jung, Tae-Young,Moon, Kyung-Sub,Min, Jung-Joon,Jung, Shin Elsevier 2018 Cancer letters Vol.433 No.-
<P><B>Abstract</B></P> <P>Glioma is one of the most devastating and refractory cancers. The main factors underlying therapeutic failure include extremely invasive characteristics and lack of effective methods for drug delivery. Attenuated <I>Salmonella</I> strains presented a high concentration of tumor targets in various types of cancer models, suggesting a role as potential vectors for drug delivery. In this study, we genetically engineered an attenuated strain of <I>Salmonella</I> as an anti-invasive vector for the targeted delivery and expression of tissue inhibitor of metalloproteinases 2 (TIMP-2) in an orthotopic nude mouse model of glioma. The bioluminescence signals related to tumor size significantly declined in the TIMP-2-expressing <I>Salmonella</I> (SLpTIMP-2)-treated group compared with the control group. Compared with the control group with a survival rate of an average of 33 days, the SLpTIMP-2 group showed an extended survival rate by nearly 60% and lasted an average period of 53 days with TIMP-2 induction. These results indicated the promising therapeutic potential of <I>S. typhimurium</I> for targeted delivery and secretion of TIMP-2 in glioma.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Intracranial injection of <I>Salmonella</I> has been demonstrated to be a more effective than tail vein injection. </LI> <LI> Treatment with TIMP-2-expressing bacteria showed down regulation of MMP-2 in orthotopic glioma. </LI> <LI> TIMP-2-expressing bacteria significantly inhibited tumor growth and elongated animal survive. </LI> </UL> </P>