检索结果-南通市图书馆

Deep reinforcement learning using least-squares truncated temporal-difference

在线全文

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2024年第2期9卷 425-439页

作者： Junkai Ren Yixing Lan Xin Xu Yichuan Zhang Qiang Fang Yujun Zeng College of Intelligence Science and Technology National University of Defense TechnologyChangshaChina State Key Laboratory of Astronautic dynamics Xi'an Satellite Control CenterXi'anChina

Policy evaluation(PE)is a critical sub-problem in reinforcement learning,which estimates the value function for a given policy and can be used for policy ***,there still exist some limitations in current PE methods,such as low sample efficiency and local convergence,especially on complex *** this study,a novel PE algorithm called Least-Squares Truncated Temporal-Difference learning(LST2D)is *** LST2D,an adaptive truncation mechanism is designed,which effectively takes advantage of the fast convergence property of Least-Squares Temporal Difference learning and the asymptotic convergence property of Temporal Difference learning(TD).Then,two feature pre-training methods are utilised to improve the approximation ability of ***,an Actor-Critic algorithm based on LST2D and pre-trained feature representations(ACLPF)is proposed,where LST2D is integrated into the critic network to improve learning-prediction *** simulation studies were conducted on four robotic tasks,and the corresponding results illustrate the effectiveness of *** proposed ACLPF algorithm outperformed DQN,ACER and PPO in terms of sample efficiency and stability,which demonstrated that LST2D can be applied to online learning control problems by incorporating it into the actor-critic architecture.

关键词： Deep reinforcement learning policy evaluation temporal difference value function approximation

Mechanical behavior of nanorubber reinforced epoxy over a wide strain rate loading

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nano Materials Science 2024年第1期6卷 106-114页

作者： Yinggang Miao Jianping Yin Wenxuan Du Lianyang Chen Joint International Research Laboratory of Impact dynamics and Its Engineering Applications School of AeronauticsNorthwestern Polytechnical UniversityXi'anShaanxi710072China State Key Laboratory for Strength and Vibration of Mechanical Structures Xi'an Jiaotong UniversityXi'anShaanxi710049China Shaanxi Key Laboratory of Impact dynamics and Its Engineering Application School of AeronauticsNorthwestern Polytechnical UniversityXi'anShaanxi710072China

Nanorubber/epoxy composites containing 0,2,6 and 10 wt%nanorubber are subjected to uniaxial compression over a wide range of strain rate from 8×10^(-4) s^(-1) to~2×10^(4) s^(-1).Unexpectedly,their strain rate sensitivity and strain hardening index increase with increasing nanorubber *** mechanisms are proposed based on numerical simulations using a unit cell *** increase in the strain rate sensitivity with increasing nanorubber content results from the fact that the nanorubber becomes less incompressible at high strain,generating a higher hydro-static *** shear localization starts to occur in the epoxy under a strain rate of 22,000 s^(-1) when the strain exceeds *** presence of nanorubber in the epoxy reduces adiabatic shear localization by preventing it from propagating.

关键词： Strain rate Strain hardening Nano rubber Epoxy Adiabatic shearing localization

Logical stochastic resonance in a cross-bifurcation non-smooth system

在线全文

学校读者我要写书评

暂无评论

Chinese Physics B 2024年第3期33卷 659-667页

作者：张宇青雷佑铭 School of Mathematics and Statistics Northwestern Polytechnical UniversityXi'an 710072China Ministry of Industry and Information Technology Key Laboratory of dynamics and Control of Complex Systems Northwestern Polytechnical UniversityXi'an 710072China

This paper investigates logical stochastic resonance(LSR)in a cross-bifurcation non-smooth system driven by Gaussian colored *** this system,a bifurcation parameter triggers a transition between monostability,bistability and *** using Novikov's theorem and the unified colored noise approximation method,the approximate Fokker-Planck equation is *** we derive the generalized potential function and the transition rates to analyze the LSR phenomenon using numerical *** simulate the logic operation of the system in the bistable and tristable regions *** assess the impact of Gaussian colored noise on the LSR and discover that the reliability of the logic response depends on the noise strength and the bifurcation ***,it is found that the bistable region has a more extensive parameter range to produce reliable logic operation compared with the tristable region,since the tristable region is more sensitive to noise than the bistable one.

关键词： logical stochastic resonance bifurcation mean first passage time

On fractional discrete financial system:Bifurcation,chaos,and control

在线全文

学校读者我要写书评

暂无评论

Chinese Physics B 2024年第10期33卷 129-140页

作者： Louiza Diabi Adel Ouannas Amel Hioual Shaher Momani Abderrahmane Abbes Laboratory of Dynamical Systems and Control University of Larbi Ben M’hidiOum El Bouaghi 04000Algeria Department of Mathematics and Computer Science University of Larbi Ben M’hidiOum El Bouaghi 04000Algeria Nonlinear dynamics Research Center Ajman UniversityAjman 346United Arab Emirates Department of Mathematics The University of JordanAmman 11942Jordan Laboratory of Mathematics Dynamics and ModelizationUniversity Badji MokhtarAnnaba 23000Algeria

The dynamic analysis of financial systems is a developing field that combines mathematics and economics to understand and explain fluctuations in financial *** paper introduces a new three-dimensional(3D)fractional financial map and we dissect its nonlinear dynamics system under commensurate and incommensurate *** such,we evaluate when the equilibrium points are stable or unstable at various fractional *** use many numerical methods,phase plots in 2D and 3D projections,bifurcation diagrams and the maximum Lyapunov *** techniques reveal that financial maps exhibit chaotic attractor *** study is grounded on the Caputo-like discrete operator,which is specifically influenced by the variance of the commensurate and incommensurate ***,we confirm the presence and measure the complexity of chaos in financial maps by the 0-1 test and the approximate entropy ***,we offer nonlinear-type controllers to stabilize the fractional financial *** numerical results of this study are obtained using MATLAB.

关键词： financial model stability chaos commensurate and incommensurate orders complexity

Differences in both the structure and interaction of the crust and mantle on the eastern and western sides of the Ordos Block

在线全文

学校读者我要写书评

暂无评论

Science China Earth Sciences 2024年

作者： Yong CHEN Yifang CHEN Jiuhui CHEN Biao GUO Yu LI Panpan ZHAO State Key Laboratory of Seismic dynamics Institute of GeologyChina Earthquake Administration National Field Scientific Observatory for Continental Rift Valley dynamics in Taiyuan Institute of Earthquake Forecasting China Earthquake AdministrationKey Laboratory for Earthquake Prediction

The Ordos Block, the western part of the North China Craton (NCC), has preserved most of its cratonic lithospheric roots during the destruction and modification of the NCC, and nowconnects two distinctly different tectonic units, i.e., the Tibetan Plateau (TP) to the west and the North China Plain to the east. It is important to studythe detailed crust-mantle structure of the Ordos Block and surroundings to understand the interactions between the Ordos and adjacent regions and the dynamics of the NCC evolution. In this study, the S-wave velocity structure at depths of 0–100 km along an E-W profile at around 36.5°N from the Qilian orogen (QLO) to the Trans-North China Orogen (TNCO) within the NCC was investigated by joint inversion of the receiver functions (RFs) and surface wave dispersion data from 104 broadband seismic stations deployed in the region under the ChinArray project (Phase Ⅱ and Phase Ⅲ). The image of discontinuity structures along the profile was further constructed by common conversion point stacking of RFs. The main results are as follows:(1) The depth of the Moho in the study region gradually decreases from west to east, being the deepest in the QLO, the second in the Ordos Block, and the shallowest in the TNCO. Local depression and uplift of the Moho are also identified beneath the Haiyuan fault zone and the Shanxi-Shaanxi rift system (SSR), respectively.(2) The crust-mantle structure displays obvious lateral heterogeneities among tectonic regions. A pronounced low-velocity anomaly is observed at depths of 20–40 km beneath the QLO and becomes weakened and thinned to the east, indicating that thickening of the mid-to-lower crust probably occurred in the QLO during the growth and lateral extrusion of the northeastern TP under the resistance of the NCC. The lithosphere-asthenosphere boundary beneath the TNCO is imaged at～90 km depth, which is significantly shallower than that beneath the Ordos *** observation suggests that the lithosphere of t

关键词：

Heat Recuperation for the Self-Condensing CO_(2)Transcritical Power Cycle

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Thermal Science 2024年第5期33卷 1630-1641页

作者： PAN Lisheng SHI Weixiu SUN Yin SUN Yiwei WEI Xiaolin State Key Laboratory of High-temperature Gas dynamics Institute of MechanicsChinese Academy of SciencesBeijing 100190China School of Environment and Energy Engineering Beijing University of Civil Engineering and ArchitectureBeijing 100044China

The supercritical CO_(2)Brayton cycle has potential to be used in electricity generation occasions with its advantages of high efficiency and compact *** on a so-called self-condensing CO_(2)transcritical power cycle,a model was established and four different layouts of heat recuperation process were analyzed,a without-recuperation cycle,a post-recuperation cycle,a pre-recuperation cycle and a re-recuperation *** results showed that the internal normal cycle's share of the whole cycle increases with increasing the cooling pressure and decreasing the final cooled *** load in the supercritical heater decreases with increasing the cooling *** perspective of performance,the re-recuperation cycle and the pre-recuperation cycle have similar thermal efficiency which is much higher than other two *** thermal efficiency and net power output have a maximum value with the cooling pressure,except in the condition with the final cooled temperature of 31℃.Considering both the complexity and the economy,the pre-recuperation cycle is more applicable than the other *** 35℃of the final cooled temperature,the thermal efficiency of the pre-recuperation cycle reaches the peak 0.34 with the cooling pressure of 8.4 MPa and the maximum net power output is 2355.24 kW at 8.2 MPa of the cooling pressure.

关键词： CO_(2)transcritical power cycle heat recuperation process self-condensing

Global intensification of weak tropical cyclones in the past 30 years

在线全文

学校读者我要写书评

暂无评论

Science China Earth Sciences 2024年第2期67卷 639-640页

作者： Dake CHEN State Key Laboratory of Satellite Ocean Environment dynamics Second Institute of OceanographyMinistry of Natural ResourcesHangzhou 310012China Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai) Zhuhai 519000Chin

The response of tropical cyclones(TCs)to climate change is a hot topic in ocean and atmosphere *** theoretical analysis and numerical experiments indicated that the intensity of TCs may increase with global warming(Emanuel,1987;Knutson and Tuleya,2004),which however has not been convincingly validated by presently available observations(Landsea,2005;Pielke,2005).Using a huge amount of global ocean surface drifter data,Wang et al.

关键词： tropical ocean cyclones

Machine learning for predicting fatigue properties of additively manufactured materials

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2024年第4期37卷 1-22页

作者： Min YI Ming XUE Peihong CONG Yang SONG Haiyang ZHANG Lingfeng WANG Liucheng ZHOU Yinghong LI Wanlin GUO State Key Laboratory of Mechanics and Control for Aerospace Structures Nanjing University of Aeronautics and AstronauticsNanjing 210016China Institute for Frontier Science Nanjing University of Aeronautics and AstronauticsNanjing 210016China College of Aerospace Engineering Nanjing University of Aeronautics and AstronauticsNanjing 210016China Shenyang Engine Research Institute Shenyang 110015China Liaoning Key Laboratory of Impact dynamics on Aero Engine Shenyang 110015China Science and Technology on Plasma dynamics Laboratory Air Force Engineering UniversityXi’an 710038China

Fatigue properties of materials by Additive Manufacturing(AM) depend on many factors such as AM processing parameter, microstructure, residual stress, surface roughness, porosities, post-treatments, etc. Their evaluation inevitably requires these factors combined as many as possible, thus resulting in low efficiency and high cost. In recent years, their assessment by leveraging the power of Machine Learning(ML) has gained increasing attentions. A comprehensive overview on the state-of-the-art progress of applying ML strategies to predict fatigue properties of AM materials, as well as their dependence on AM processing and post-processing parameters such as laser power, scanning speed, layer height, hatch distance, built direction, post-heat temperature,etc., were presented. A few attempts in employing Feedforward Neural Network(FNN), Convolutional Neural Network(CNN), Adaptive Network-Based Fuzzy Inference System(ANFIS), Support Vector Machine(SVM) and Random forest(RF) to predict fatigue life and RF to predict fatigue crack growth rate are summarized. The ML models for predicting AM materials' fatigue properties are found intrinsically similar to the commonly used ones, but are modified to involve AM features. Finally, an outlook for challenges(i.e., small dataset, multifarious features,overfitting, low interpretability, and unable extension from AM material data to structure life) and potential solutions for the ML prediction of AM materials' fatigue properties is provided.

关键词： Additive manufacturing Machine learning Fatigue life Fatigue crack growth rate Prediction

Integrating remote sensing and 3-PG model to simulate the biomass and carbon stock of Larix olgensis plantation

在线全文

学校读者我要写书评

暂无评论

forest Ecosystems 2024年第4期11卷 543-555页

作者： Yu Bai Yong Pang Dan Kong Key Laboratory of National forestry and Grassland Administration on forestry Remote Sensing and Information System Institute of Forest Resource Information

Accurate estimations of biomass and its temporal dynamics are crucial for monitoring the carbon cycle in forest ecosystems and assessing forest carbon sequestration *** studies have shown that integrating process-based models(PBMs)with remote sensing data can enhance simulations from stand to regional scales,significantly improving the ability to simulate forest growth and carbon stock ***,the utilization of PBMs for large-scale simulation of larch carbon storage distribution is still *** this study,we applied the parameterized 3-PG(Physiological Principles Predicting Growth)model across the Mengjiagang forest Farm(MFF)to make broad-scale predictions of the biomass and carbon stocks of Larix olgensis *** model was used to simulate average diameter at breast height(DBH)and total biomass,which were later validated with a wide range of observation data including sample plot data,forest management inventory data,and airborne laser scanning *** results showed that the 3-PG model had relatively high accuracy for predicting both DBH and total biomass at stand and regional scale,with determination coefficients ranging from 0.78 to *** on the estimation of total biomass,we successfully produced a carbon stock map of the Larix olgensis plantation in MFF with a spatial resolution of 20 m,which helps with relevant management *** findings indicate that the integration of 3-PG model and remote sensing data can well predict the biomass and carbon stock at regional and even larger *** addition,this integration facilitates the evaluation of forest carbon sequestration capacity and the development of forest management plans.

关键词： 3-PG model Larch Biomass Carbon stock ALS