The variable reconstruction method to establish the parameterized model of a certain kind of bearing-rotor system is proposed by integrating the multisource programs, parameterized modeling and collaborative process m...
详细信息
computersimulation use the computermodeling and simulation technology to realistically display the internal structure, working principle and related process flow of engineering technology, bioreactor, process equipm...
详细信息
The effects of a pandemic affect several areas of human activity. One of the key areas is production, which immediately reflects on economic growth. Through the years, theoretical and sound models have independently b...
详细信息
This paper presents techniques for building a modern corporative computer network. The paper focuses on a real architectural approach for implementing a corporative computer network based on the use of a hierarchical ...
详细信息
This paper presents techniques for building a modern corporative computer network. The paper focuses on a real architectural approach for implementing a corporative computer network based on the use of a hierarchical network connectivity model. Requirements for a private corporate computer network are defined. A network infrastructure model of a corporative computer network is proposed. An address scheme is proposed. A topology of a computer network with scalability based on modern modules and equipment of MikroTik is synthesized. Tests in simulation environment are made.
More and more researchers are trying to use deep reinforcement learning (DRL) for mobile robot tasks due to its powerful inference capability. However, deep reinforcement learning requires a large amount of data for D...
详细信息
ISBN:
(纸本)9781665481106
More and more researchers are trying to use deep reinforcement learning (DRL) for mobile robot tasks due to its powerful inference capability. However, deep reinforcement learning requires a large amount of data for DRL training in the pre-experimental stage, which hinders the application of the algorithm. On the other hand, the inconsistency between the ROS data interface and DRL GYM-Like data interface leads to a high cost of migration of the algorithm verification. This paper proposes a fast simulator generation method using linear approximate kinematics model and bake-based lidar rendering methods to generate a fast approximate simulator used in the pre-experiment stage to solve the problem of data cost. At the same time, an experimental system design scheme that converts the ROS interface into a GYM-like interface is also proposed to simplify the deployment process of deep reinforcement learning. We evaluate our proposed method on collision avoidance tasks in a variety of kinematics models and lidar scenarios. Our Method achieves about 14.2 times kinematics simulation speedup and 2.56 times lidar rendering speedup. We open-sourced our simulation environment and robot system software at https://***/efc-robot/MultiVehicleEnv and https://***/efc-robot/NICS_MultiRobot_Platform
We present a control architecture for real-time adaptation and tracking of trajectories generated using a terrain-aware trajectory optimization solver. This approach enables us to circumvent the computationally exhaus...
详细信息
ISBN:
(纸本)9781728190778
We present a control architecture for real-time adaptation and tracking of trajectories generated using a terrain-aware trajectory optimization solver. This approach enables us to circumvent the computationally exhaustive task of online trajectory optimization, and further introduces a control solution robust to systems modeled with approximated dynamics. We train a policy using deep reinforcement learning (RL) to introduce additive deviations to a reference trajectory in order to generate a feedback-based trajectory tracking system for a quadrupedal robot. We train this policy across a multitude of simulated terrains and ensure its generality by introducing training methods that avoid overfilling and convergence towards local optima. Additionally, in order to capture terrain information, we include a latent representation of the height maps in the observation space of the RL environment as a form of exteroceptive feedback. We test the performance of our trained policy by tracking the corrected set points using a model-based whole-body controller and compare it with the tracking behavior obtained without the corrective feedback in several simulation environments, and show that introducing the corrective feedback results in increase of the success rate from 72.7% to 92.4% for tracking precomputed dynamic long horizon trajectories on flat terrain and from 47.5% to 803% on a complex modular uneven terrain. We also show successful transfer of our training approach to the real physical system and further present cogent arguments in support of our framework.
The proceedings contain 20 papers. The special focus in this conference is on High Performance Computing for Advanced modeling and simulation in Nuclear Energy and Environmental Science. The topics include: Prospects ...
ISBN:
(纸本)9789813299863
The proceedings contain 20 papers. The special focus in this conference is on High Performance Computing for Advanced modeling and simulation in Nuclear Energy and Environmental Science. The topics include: Prospects for CVR-0: A prototype of china virtual reactor;Diffusion mechanism of small helium-vacancy clusters in BCC Fe and Fe-10%cr alloy;Effect of Ni and Mn on the interaction of an edge dislocation with cu-rich precipitates in BCC fe;The performance test and optimization of crystal-MD program on Tianhe-2;research on large scale parallel hydrological simulation;The study of parallelization of SWAT hydrology cycle;research on shared resource contention of cloud data center;a sequence anomaly detection approach based on isolation forest algorithm for time-series;Hourly day-ahead power forecasting for pv plant based on bidirectional LSTM;a reliability task scheduling for high-throughput computing platform;an edge computing platform for intelligent internet data center operational monitoring;topology layout technology of energy internet;big data analytics for water resources sustainability evaluation;exploring water resource changes of artificial reservoir using time-series remote sensing images from landsat sensors and in situ data;a deep learning based objection detection method for high resolution remote sensing image;a case study for user rating prediction on automobile recommendation system using mapreduce;diverse demands estimation and ranking based on user behaviors;energy aware edge computing: a survey.
The article is devoted to the description of the complex method to design granulation unit, which is based on the joint use of the computermodeling results on the simulation models and software modeling based on aut...
详细信息
As power modules based on (U)WBG materials gain attention for their potential to revolutionize high-voltage, high-power density applications, their optimal performance is hindered by current insulation materials. The ...
详细信息
ISBN:
(数字)9798350348538
ISBN:
(纸本)9798350348545
As power modules based on (U)WBG materials gain attention for their potential to revolutionize high-voltage, high-power density applications, their optimal performance is hindered by current insulation materials. The high electric field stress at triple points (TPs) and encapsulant's diminishing breakdown field strength at high temperatures significantly limit the application of power modules to below 200°C. Therefore, alternative encapsulation materials are essential for hightemperature applications. This paper proposes a novel solution involving replacing standard silicone gel (SG) encapsulant with high-temperature encapsulation composites created by integrating micro- and nano-sized fillers into a silicone elastomer matrix. Additionally, the ceramic structure is redesigned into a mesa configuration, featuring a trench at the base of the metal electrodes. The results from modeling and simulation of COMSOL Multiphysics demonstrate that this combined approach significantly reduces field stress at TPs. Although effective at lower frequencies, further design modifications, such as altering the metalized substrate into a protruding structure, are necessary for optimal electric field mitigation at frequencies above 10 kHz. This research addresses the critical issues of high electric field values at TPs and the limited temperature tolerance of current encapsulation materials, which are challenges exacerbated under high-frequency square pulse operation. By resolving these issues, this work represents a valuable advancement in utilizing highvoltage, high-power density (U)WBG power modules at demanding frequency environments.
Inverse problems describe the task of recovering an underlying signal of interest given observables. Typically, the observables are related via some non-linear forward model applied to the underlying unknown signal. I...
Inverse problems describe the task of recovering an underlying signal of interest given observables. Typically, the observables are related via some non-linear forward model applied to the underlying unknown signal. Inverting the non-linear forward model can be computationally expensive, as it often involves computing and inverting a linearization at a series of estimates. Rather than inverting the physics-based model, we instead train a surrogate forward model (emulator) and leverage modern auto-grad libraries to solve for the input within a classical optimization framework. Current methods to train emulators are done in a black box supervised machine learning fashion and fail to take advantage of any existing knowledge of the forward model. In this article, we propose a simple learned weighted average model that embeds linearizations of the forward model around various reference points into the model itself, explicitly incorporating known physics. Grounding the learned model with physics based linearizations improves the forward modeling accuracy and provides richer physics based gradient information during the inversion process leading to more accurate signal recovery. We demonstrate the efficacy on an ocean acoustic tomography (OAT) example that aims to recover ocean sound speed profile (SSP) variations from acoustic observations (e.g. eigenray arrival times) within simulation of ocean dynamics in the Gulf of Mexico.
暂无评论