Massively Parallel Simulation Algorithms for Grid-Based Analog Signal Processors
一些专有名词的翻译
simulation 仿真；模拟simulation algorithm 仿真算法simulation algorithm libray 仿真算法库simulation block diagram 仿真（方）框图simulation centre 仿真中心simulation clock 仿真时钟simulation data base 仿真数据库simulation environment 仿真环境simulation equipment 仿真设备simulation evaluation 仿真评价simulation experiment 仿真实验simulation experiment modelibrary 仿真实验模式库simulation expert system 仿真专家系统simulation graphic library 仿真图形库simulation information library 仿真信息库simulation job 仿真作业simulation knowledge base 仿真知识库simulation laboratoryt 仿真实验室simulation language 仿真语言simulation methodology 仿真方法学simulation model 仿真模型simulation model library 仿真模型库simulation process 仿真过程simulation process time 仿真过程时间simulation program 仿真程序simulation result 仿真结果simulation run 仿真运行simulation software 仿真软件simulation support system 仿真支持系统simulation system 仿真系统simulation technique 仿真技术simulation type 仿真类型simulation velocity 仿真速度simulation work station 仿真工作站simulator 仿真器simultancous comparison method 同时比较法simultaneous technique 同时联用技术；同时并用技术sing around method 声环法sing-around velocimeter 环鸣声速仪single acting positioner 单作用定位器single arm measurement 单臂测量single beam spectrum radiator 单光束光谱辐射计single board microcomputer 单片微（型）计算机single bounce technique 一次反射法single-channel FSK system 单通道FSK系统single channel recorder 单通道记录仪single core typy current transformer 单铁心型电流互感器single ended transducer 单端换能器single field lens 单场透镜single focusing 单聚焦single-focusing mass spectrograph 单聚焦质谱仪single-focusing mass spectrometer 单聚焦质谱计single function(measuring)instrument 单功能（测量）仪表single gauge measurement 单计[片]测量single grain layer varistor 单颗粒层电压敏电阻器single-idler electronic belt conveyor scale 单托辊电子皮带秤single input single output control system;SISO control system 单输入单输出控制系统single-jet water meter 单流束水表single-junction SQUID 单结量子干涉器single junction temperature transducer [sensor] 单结温度传感器single level process 单级过程single loop control 单回路控制single loop control system 单回路控制系统single loop controller 单回路控制器single magnet galvanometer 单磁式振动子single-pass internal reflection element 单通内反射元件single-path diagonal-beam flowmeter 单声道斜束式流量计single-path ratio thermometer 单通道比色温度计single piston pressure-vacuum gauge 单活塞压力真空计single-plane (static) alancing 单面（静）平衡single-plane (static)balancing machine 单面（静）平衡机single probe technique 单探头法single range (measuring) instrument 单范围[量限]（测量）仪表single sheet apparatus for measuring specific total losses of magnetic sheet and strip 单片电工钢片[带]比总损耗测量装置single scintillation radioactive logger 单道闪烁辐射测井仪single-speed floating action 单速无定位作用single-speed floating controller 单速无定位控制器single-tube manometer 单管压力计single tube mercury manometer 单管水银压力表single value nonlinearity 单值非线性single variable control system 单变量控制系统sinker （浮子流量计）浮子sintered gas sensor 烧结式气敏元件sinusoidal quantity 简谐波siphon action 虹吸作用siphon barometer 虹吸气压表siphon pipe [tube] 虹吸管siphon rainfall recorder 虹吸式雨量计six component balance 六分力天平Six's thermometer 最高最低温度表；西克斯温度表skip distance 跨距点slant visibility 斜能见度slave operation 从动工作slave station 从站slaved system 受役系统slaving principle 役使原理silding vane rotary flowmeter 刮板流量计sling psychrometer 手摇干湿表sling thermometer 手摇温度表slip bezel ring 滑动盖环slope error over 10% 10%段的斜率误差slope factor 斜率slope over 10% 10%段斜率slope/temperature factor adjustment of pH meter pH计的斜率/温度系数校准器slurry packing 匀浆填充small focus X-ray tube 小焦点X射线管Smith-McIntyre mud sampler 史密斯—麦金太尔取泥器snaking motion value of specimen stage 样品台调节蛇行量snap ring 开口环snapper grab sampler 表层采样器snow density meter 积雪密度计snow measuring plate 积雪板snow-stake 测雪桩snow-storm 雪暴snow storm meter 雪暴测定仪snowfall 雪量snowfall totalizer 累计雪量计snowgauge 量雪器snowsampler 雪取样器；取雪器snubber 限制器soap-film burette 皂膜量管soft bearing balancing machine 软支承平衡机soft ionization 软电离soft keyboard 软键盘"soft" X-rays 软X射线software 软件software compatibility 软件兼容性software cost 软件成本software design procedure 软件设计过程software development library 软件开发库software development plan 软件开发计划software development process 软件开发过程software documentation 软件文件software engineering 软件工程software environment 软件环境software library 软件库software maintenance 软件维护software monitor 软件监督程序software package 软件包software package of computer aided disign 计算机辅助设计软件包software portability 软件可移植性software product 软件产品software psychology 软件心理学software quality 软件质量software reliability 软件可靠性software testing plan 软件测试计划software testing 软件测试software tool 软件工具soil evaporimeter 土壤蒸发仪soil moisture 土壤水分soil moisture content analyser 土壤水份测定仪soil oven 烘土箱soil thermometer 直管地温表solar constant 太阳常数solar radiation 太阳辐射solarigraph 总日射计solarimeter 总日射表solenoid 螺线管solenoid coil 电磁线圈solenoid valve 电磁阀solenoid valve for freon 氟里昂用电磁阀solenoid valve for gas 煤气电磁阀solenoid valve for steam 蒸气电磁阀solenoid valve for water 水用电磁阀solid electrolyte oxygen analyzer 固体电解质氧分析器solid front case with pressure relief at back 后部带泄压装置的前封式外壳solid scanning length measuring instrument 固体扫描式测长仪solid scanning transducer 固体扫描传感器solid scanning width meter [gauge] 固体扫描式宽度计solid-stage electrolyte gas transducer [sensor] 固体电解质气体传感器solid-state electrolyte humidity transducer [sensor] 固体电解质湿度传感器Solid-state electrolyte oin transducer [sensor] 固体电解质离子传感器solid-state (X-ray) detector 固态（X射线）检测器solid-stem liquid-in-glass thermometer 棒式玻璃温度计solo 单独布置；单独检测solvent removable dye penetrant testing method 溶剂去除着色渗透探伤法solvent removable penetrant 溶剂去除性渗透液sonar 声纳sonar dome 声纳导流罩sone 宋（响度单位）sonic [critical] Venturi nozzle 音速[临界]文丘里喷嘴sonic logger 声速测井仪sonic nozzle 音速喷嘴sound daffle 声障板sound energy density 声能量密度sound field 声场sound intensity 声强sound intensity level 声强级sound level 声级sound level calibrator 声级校准器sound level meter 声级计sound power 声功率sound power level 声功率计sound pressure 声压sound pressure level 声压级sound pressure transducer [sensor] 声压传感器sound radiation 声辐射sound ray tracking plotter 声线轨迹仪sound reflector 声反射器sound source 声源sound spectrum 声谱sounding 探测source language 源语言source of electron gun grid bias 电子枪栅偏压电源source of radiation 辐射源source program 源程序source slit 离子源狭缝space byte 空格字节space remote sensing 航天遥感space telemetry 航天遥测spacelab 太空实验库spacer 衬圈；垫片spaceship 宇宙飞船span 量程span calibration gas 量程校准气span drift 量程漂移span error 量程误差span of impact specimen supports 冲击试样支座跨距span shift 量程迁移[偏移]spark-proof instrument 安全火花型仪器spark source 火花电离源sparker 电火花震源spatial filter 空间滤波器spatial resolution 空间分辨率special simulation technique 特殊仿真技术specific acoustic impedance 声阻抗率specific gravity 比重specific humidity 比湿specific permeability 比渗透率specific resistance 电阻率specific retention volume 比保留体积specific service (pressure) gauge 特殊用途压力表specific viscosity 比粘specific weight 专用砝码specification 规格specified characteristic curve 规定特性曲线specified sensitivity 规定灵敏度specimen 试样；样品specimen chamber 样品室specimen cooling holder 致冷样品台specimen heating holder 加热样品台specimen holder 样品杯；样品杆；试样架specimen-holder assembly 样品（支持）器组件specimen rotating holder 旋转样品台specimen tensile holder 拉伸样品台spectral background 光谱背景spectral bandwidth 光谱带宽spectral characteristic curve 光谱特性曲线spectral density 谱密度spectral distribution curve 光谱分布曲线spectral distribution of energy 光谱能量分布spectral emissivity 光谱发射率spectral half width 光谱半宽度spectral line 光谱线spectral position 光谱位置spectral radiance 光谱辐射亮度spectral radiation exitance 光谱辐（射）出（射）度spectral range 光谱范围spectral resolution 光谱分辨率spectral slit width 光谱狭缝宽度spectro chemical analysis 光谱化学分析spectrofluorophotometer 荧光分光光度计spectrograph 摄谱仪spectrometer 光谱仪spectrometer channel 分光波道spectrophotometer 分光光度计spectrophotometric titration 分光光度滴定法spectropolarimeter 旋光仪spectroscopy 看谱镜；能谱法spectrum 光谱；谱spectrum analyzer 频谱分析仪spectrum radiator 光谱辐射计speech recognition 语音识别speed characteristic 转速特性speed control system 调速系统speed effect 速度效应spherical aberration 球差spherical phranometer 球形总日射表spherical phrgeometer 球形地球辐射表spherical phrradiometer 球形全辐射表spin axis 旋转轴spin decoupling 自旋去耦spin test (of a current-meter) （流速计的）旋转试验spinning magnetometer 旋转磁力仪spinning sidebands 旋转边带spin-sin coupling constant （核磁共振）自旋—自旋耦合常数spirit level 气泡式水准仪split-body valve 分体阀split core type current transformer 钳式电流互感器split range opoeration 分程操作split-ranging 分程split screen 分区屏幕split stream injector 分流进样器splitter 分流器spot radiation source 点辐射源spot recorder 光点记录仪SPOT satellite 斯波特卫星spot scanning 点扫描spot size 目标尺寸spray method 喷雾方法spraying device 喷雾装置spring-loaded regulator 弹簧型自力式调节阀spring-loaded variable-head flowmeter 弹性加载可变压头流量计spring plate 弹簧盘spring testing machine 弹簧试验机spurious echo 楔内反射波spurious errors 疏忽误差square-edged orifice plate 直角边缘孔板square frame of magnetic needle 方框罗针square profile (pressure) gaege 矩形压力表square-wave polarogyaph 方波极谱仪stability 稳定性；稳定度stability analysis 稳定性分析stability condition 稳定（性）条件stability criterion 稳定（性）判据；稳定（性）准则stability error 稳定性误差stability limit 稳定（性）极限stability margin 稳定裕度；稳定裕量stability method 稳定法stability of towed body 拖曳体稳定性stability theory 稳定性理论stabilizability 可稳性；能稳性stabilization 镇定；稳定stabilized load characteristic 稳定负载特性stabilized supply apparatus 稳定电源stabilized voltage varistor 稳压电压敏电阻器stabilizing network 镇定网络stabilizing period 稳定过程stable region 稳定域stable system 稳定系统stable type gravimeter 稳定型重力仪stack 栈stacking test 堆码试验stadia line 视距线stadia rod 视距尺stadia wave gauge 视距测波仪staff tide gauge 验潮杆；水尺stagnation pressure 滞止压力stain sync 应变同步standard 标准standard acceleration transducer 标准加速度传感器standard accelerometer 标准加速度计standard buffer solution 标准缓冲溶液standard calorimeter 标准型热量计standard capacitor 标准电容器standard cell 标准电池standard deviation of a single measurement in a series of measurments 测量列中单次测量的标准（偏）差。
圣维南原理的有限元模拟
圣维南原理的有限元模拟圣维南原理是电子学中的一项基本原理，用于描述电导体中电流分布情况的方法，常用于有限元模拟中来解决电磁场问题。
有限元模拟是一种基于数值方法的工程分析技术，通过将连续的物理问题离散化为有限数量的元素，再利用数值计算方法对这些元素进行求解，以模拟实际问题的行为和物理特性。
以下是关于圣维南原理在有限元模拟中的详细介绍。
圣维南原理（Saint-Venant’s Principle）主要用于描述电导体中的电流分布情况。
它是基于电流连续性方程和欧姆定律的基本原理，即电流在导体内部的分布是均匀且沿导体表面方向渐变。
根据这个原理，在有限元模拟中可以通过离散化导体为一系列有限元素来近似描述电流的分布情况。
在有限元模拟中，首先需要将导体区域划分为小块，称为有限元。
每个有限元都有一组自由度，用于描述电场强度或电势的分布情况。
在圣维南原理的约束下，任意两个相邻的有限元之间，在其界面上，电场强度或电势需要满足一定的连续性条件。
这些连续性条件可以通过将不同有限元之间的界面进行连接，构建整个导体区域的有限元模型。
有限元模型构建完成后，利用数值方法求解模型中的电场强度或电势分布。
通常采用有限元法的变分形式，通过求解最小化电场强度或电势的能量泛函来得到电场方程的离散形式。
然后，通过数值求解方法（如有限差分法等）对离散的电场方程进行求解，得到电场强度或电势分布的近似解。
由于圣维南原理的应用，有限元模拟能够较准确地描述导体中电流的分布情况。
采用有限元模拟方法，可以更好地理解和分析各种电磁场问题，如电磁传感器中的电流分布、电源线中的电压降等。
有限元模拟结果可以帮助工程师优化设计和制造过程，提高电子设备的性能和可靠性。
总之，圣维南原理作为电导体中电流分布的基本原理，在有限元模拟中扮演着关键的角色。
通过有限元模拟，可以准确地描述电流在导体中的分布情况，帮助工程师解决电磁场问题，从而优化设计和制造过程，提高电子设备的性能和可靠性。
多体动力学仿真流程
多体动力学仿真流程英文回答：Multi-body dynamics simulation is a process used to study the motion and interaction of multiple bodies in a system. It is widely used in various fields such as robotics, biomechanics, automotive engineering, and aerospace engineering. The simulation allows us to analyze the behavior of the system under different conditions and make predictions about its performance.The general workflow of a multi-body dynamics simulation involves several steps. Firstly, we need to define the bodies in the system and their properties such as mass, geometry, and material properties. This can be done using a modeling software or by importing CAD models. Then, we define the constraints and connections between the bodies, such as joints, hinges, and contacts. These constraints determine how the bodies interact with each other.Once the system is defined, we need to set up theinitial conditions of the simulation. This includes specifying the initial positions, velocities, and accelerations of the bodies. We may also need to apply external forces or torques to the system to simulate real-world conditions.Next, we need to choose a suitable numericalintegration method to solve the equations of motion for the system. Common methods include the Euler method, Runge-Kutta methods, and the Verlet algorithm. The choice of integration method depends on the accuracy andcomputational efficiency required for the simulation.After setting up the simulation parameters, we canstart the simulation and observe the motion of the bodies over time. The simulation software calculates the positions, velocities, and accelerations of the bodies at each time step based on the applied forces and constraints. We can visualize the results using animations or plot graphs ofthe variables of interest.During the simulation, we can analyze the behavior of the system and extract relevant data such as forces, torques, and energy. This data can be used to evaluate the performance of the system and make design improvements if necessary.Once the simulation is complete, we can post-process the results by analyzing the data and generating reports or visualizations. This helps us to understand the behavior of the system in more detail and communicate the findings to others.In conclusion, the multi-body dynamics simulation process involves defining the system, setting up initial conditions, choosing an integration method, running the simulation, analyzing the results, and post-processing the data. It is a powerful tool for studying the motion and interaction of multiple bodies in various engineering and scientific applications.中文回答：多体动力学仿真流程是用于研究系统中多个物体的运动和相互作用的过程。
simulation 形容词
simulation 形容词simulation (形容词) - simulated or imitated closely according to models or patterns1. The flight simulator provided a realistic simulation of flying a fighter jet.飞行模拟器提供了逼真的战斗机飞行模拟体验。
2. The business simulation game allowed participants to experience running a virtual company.这个商业模拟游戏让参与者能够体验经营一个虚拟公司。
3. The virtual reality headset created a simulation of being underwater.这款虚拟现实头盔创造了一种水下的模拟体验。
4. The flight attendant training included a simulation of emergency situations.乘务员培训包括紧急情况的模拟。
5. The simulation exercise helped doctors practice performing surgeries before operating on real patients.模拟训练有助于医生在进行实际手术之前进行实践。
6. The video game provided a simulation of being a professional athlete on the soccer field.这个电子游戏提供了一个模拟身份成为职业足球运动员的体验。
7. The military used a simulation of a battlefield to train soldiers for combat scenarios.军方使用战场模拟训练士兵应对战斗场景。
The “Simulation Thing”“模拟物”
Particle Simulations
Particle Simulations
These particles can represent different entities depending on the simulation. This could be:
atoms molecules dust particles snooker balls asteroids planets galaxies
Particle Simulations
Once a collision has been detected the system must respond to the collision. For our hockey pucks, simply reverse the velocity in the direction of the collision.
Particle Simulations
• main loop – for all particles • ‘move’ particle • if ‘collision’ with boundary – respond to collision – for all particles • for all other particles – if ‘collision’ between particles » respond to collision
Particle Simulations
Second Law: The acceleration a of a body is parallel and directly proportional to the net force F acting on the body, is in the direction of the net force, and is inversely propotional to the mass m of the body.
分子动力学加电场;lammps
分子动力学加电场；lammpsEnglish Response:Introduction.Molecular dynamics simulations with applied electric fields are widely used to investigate the behavior of charged materials, such as ions in electrolyte solutions or proteins in biological systems. LAMMPS (Large-scaleAtomic/Molecular Massively Parallel Simulator) is a popular molecular dynamics simulation package that offers a versatile platform for performing simulations with applied electric fields.Setting up the Simulation.To set up a molecular dynamics simulation with an applied electric field in LAMMPS, several key steps are involved:1. Define the System: The first step is to define the simulation system, including the molecular structure, atomic charges, and simulation box.2. Create the Input Script: An input script is created to specify the simulation parameters, such as the force field, timestep, and simulation length.3. Apply the Electric Field: An electric field is applied to the system using the "fix efield" command. This command specifies the magnitude and direction of the electric field.4. Run the Simulation: The simulation is run using the "run" command.Analysis of Results.Once the simulation is complete, the results can be analyzed to understand the effect of the electric field on the system. Some common analysis methods include:1. Particle Trajectories: The trajectories ofindividual particles can be tracked to observe their motion under the influence of the electric field.2. Radial Distribution Functions: Radial distribution functions can be calculated to analyze the distribution of particles around a central particle.3. Electric Potential: The electric potential distribution within the simulation box can be computed to visualize the effect of the electric field on the system.Example Input Script.Below is an example input script for a molecular dynamics simulation with an applied electric field in LAMMPS:units real.atom_style full.read_data mmps.fix efield all efield 0.0 0.0 1.0 1.0e5 v_global #Apply electric field along z-axis with magnitude 1e5 V/m.run 100000。
PARALLEL AND DISTRIBUTED SIMULATION SYSTEMS
Proceedings of the 2001 Winter Simulation ConferenceB. A. Peters, J. S. Smith, D. J. Medeiros, and M. W. Rohrer, eds.ABSTRACTOriginating from basic research conducted in the 1970’s and 1980’s, the parallel and distributed simulation field has ma-tured over the last few decades. Today, operational systems have been fielded for applications such as military training, analysis of communication networks, and air traffic control systems, to mention a few. This tutorial gives an overview of technologies to distribute the execution of simulation pro-grams over multiple computer systems. Particular emphasis is placed on synchronization (also called time management) algorithms as well as data distribution techniques.1 INTRODUCTIONParallel and distributed simulation is concerned with issues introduced by distributing the execution of a discrete event simulation program over multiple computers. Parallel dis-crete event simulation is concerned with execution on mul-tiprocessor computing platforms containing multiple cen-tral processing units (CPUs) that interact frequently, e.g., thousands of times per second. Distributed simulation is concerned with the execution of simulations on loosely coupled systems where interactions take much more time, e.g., milliseconds or more, and occur less often. It includes execution on geographically distributed computers inter-connected via a wide area network such as the Internet. In both cases the execution of a single simulation model, per-haps composed of several simulation programs, is distrib-uted over multiple computers.There are two principal categories of simulations of concern here. The first are simulations primarily used for analysis, e.g., to evaluate alternate designs or control poli-cies of a complex system, e.g., an air traffic network. Here the principal goal is to compute results of the simulation as quickly as possible in order to improve the effectiveness of the simulation tool. A related application is to use simula-tions to evaluate alternate courses of action, e.g., to evaluate different control actions imposed on an air traffic network in order to reduce delays induced by inclement weather in one portion of the air space. Use of simulations to manage on-going processes is referred to as on-line simulation.The second type of simulation of interest here are those used to create virtual environments into which hu-mans and/or hardware devices are embedded. Such envi-ronments are widely used for training, entertainment (e.g., video games), and test of evaluation of devices. Virtual environments have been used extensively to train military personnel because they provide a much safer, more cost effective, and environmentally friendlier approach to train-ing than field exercises.Parallel and distributed simulation systems can provide substantial benefit to these applications in several ways: • Execution times of analytic simulations can be re-duced by subdividing a large simulation computa-tion into many sub-computations that can executeconcurrently. One can reduce the execution timeby up to a factor equal to the number of proces-sors that are used. This may be important simplybecause the simulation takes a long time to exe-cute, e.g., simulations of communication networkscontaining tens of thousands of nodes may requiredays or weeks for a single run.• Very fast executions are needed for on-line simu-lations because there is often very little timeavailable to make important decisions. In manycases, simulation results must be produced in sec-onds in order for simulation results to be useful.Again, parallel simulation provides a means to re-duce execution time.• Simulations used for virtual environments must execute in real time, i.e., the simulator must beable to simulate a second of activity in a second ofwallclock time so that the virtual environment ap-pears realistic in that it evolves as rapidly as theactual system. Distributing the execution of thesimulation across multiple processors can help toachieve this property. Ideally, scalable executioncan be obtained whereby the distributed simula-tion continues to run in real time as the system be-PARALLEL AND DISTRIBUTED SIMULATION SYSTEMSRichard M. FujimotoCollege of ComputingGeorgia Institute of TechnologyAtlanta, GA 30332-0280, U.S.A.ing simulated and the number of processors areincreased in proportion.• Distributed simulation techniques can be used to create virtual environments that are geographi-cally distributed, enabling one to allow humansand/or devices to interact as if they were co-located. Such distributed virtual environmentshave obvious benefits in terms of convenienceand reduced travel costs.• Distributed simulation can simplify integrating simulators that execute on machines from differentmanufacturers. For example, flight simulators fordifferent types of aircraft may have been developedon different architectures. Rather than porting theseprograms to a single computer, it may be more costeffective to “hook together” the existing simulators,each executing on a different computer, to create anew virtual environment.• Another potential benefit of utilizing multiple processors is increased tolerance to failures. If oneprocessor fails, it may be possible for other proc-essors to continue the simulation provided criticalelements do not reside on the failed processors.Work in parallel and distributed simulation systems has taken place in three, largely separate research communities. The first is the high performance computing community which was concerned primarily with speeding up the execu-tion of simulation programs by distributing their execution over multiple CPUs. Early work in synchronization algo-rithms dates back to the late 1970’s with seminal work by Chandy and Misra (1978), and Bryant (1977) (among oth-ers) who are credited with first formulating the synchroniza-tion problem and developing the first algorithms to solve it. These algorithms are among a class of algorithms that are today referred to as conservative synchronization algo-rithms. A few years later, seminal work by Jefferson devel-oped the Time Warp algorithm (Jefferson 1985). Time Warp is important because it defined fundamental constructs widely used in a class of algorithms termed optimistic syn-chronization. Conservative and optimistic synchronization techniques form the core of a large body of work concerning parallel discrete event simulation techniques.The second community involved in the development of distributed simulation technology is the defense com-munity. While the high performance computing commu-nity was largely concerned with reducing execution time, the defense community was concerned with integrating separate training simulations in order to facilitate interop-erability and software reuse. The SIMNET (SIMulator NETworking) project (1983 to 1990) demonstrated the vi-ability of using distributed simulations to create virtual worlds for training soldiers in military engagements (Miller and Thorpe 1995). This lead to the development of a set of standards for interconnecting simulators known as the Distributed Interactive Simulation (DIS) standards (IEEE Std 1278.1-1995 1995). The 1990’s also saw the development of the Aggregate Level Simulation Protocol (ALSP) that applied the SIMNET concept of interoperabil-ity and model reuse to wargame simulations. ALSP and DIS have since been replaced by the High Level Architec-ture whose scope spans the broad range of defense simula-tions, including simulations for training, analysis, and test and evaluation of hardware components.A third track of research and development efforts arose from the Internet and computer gaming community. Work in this area can be traced back to a role-playing game called dungeons and dragons and a textual fantasy computer game called Adventure developed in the 1970’s. These soon gave way to MultiUser Dungeon (MUD) games in the 1980’s. Important additions such as sophisti-cated computer graphics helped created the video game in-dustry that is flourishing today.This paper is organized as follows. The next section is concerned with the execution of analytic simulations on parallel computers, with the principal goal of reducing execution time. Synchronization is a key problem that must be addressed. Section 3 is concerned with an ap-proach to parallel simulation known as time decomposi-tion. Section 4 is concerned with distributed virtual envi-ronments and issues such as data distribution that arise in that domain. This paper is an updated version of a previ-ous tutorial presented at this conference in (Fujimoto 1999b). A much more detailed treatment of this subject is presented in (Fujimoto 2000).2 TIME MANAGEMENTTime management is concerned with ensuring that the exe-cution of the parallel/distributed simulation is properly synchronized. This is particularly important in analytic simulations. Time management not only ensures that events are processed in a correct order, but also helps to ensure that repeated executions of a simulation with the same inputs produce exactly the same results. Currently, time management techniques such as those described here are typically not used in training simulations, where incor-rect event orderings and non-repeatable simulation execu-tions can usually be tolerated.Time management algorithms usually assume the simulation consists of a collection of logical processes (LPs) that communicate by exchanging timestamped mes-sages or events. The goal of the synchronization mecha-nism is to ensure that each LP processes events in time-stamp order; this requirement is referred to as the local causality constraint. It can be shown that if each LP ad-heres to the local causality constraint, execution of the simulation program on a parallel computer will produce exactly the same results as an execution on a sequential computer. An important side effect of this property is thatit is straightforward to ensure that the execution of the simulation is repeatable.Each LP can be viewed as a sequential discrete event simulation. This means each LP maintains some local state and a list of time stamped events that have been scheduled for this LP (including local events within the LP that it has scheduled for itself), but have not yet been processed. This pending event list must also include events sent to this LP from other LPs. The main processing loop of the LP repeat-edly removes the smallest time stamped event and processes it. Thus, the computation performed by an LP can be viewed as a sequence of event computations. Processing an event means zero or more state variables within the LP may be modified, and the LP may schedule additional events for itself or other LPs. Each LP maintains a simulation time clock that indicates the time stamp of the most recent event processed by the LP. Any event scheduled by an LP must have a time stamp at least as large as the LP’s simulation time clock when the event was scheduled.Time management algorithms can be classified as be-ing either conservative or optimistic. Each of these are de-scribed next.2.1 Conservative SynchronizationThe first synchronization algorithms were based on conser-vative approaches. This means the synchronization algo-rithm takes precautions to avoid violating the local causality constraint. For example, suppose an LP is at simulation time 10, and it is ready to process its next event with time stamp 15. But how does the LP know it won’t later receive an event from another LP with time stamp (say) 12? The synchronization algorithm must ensure no event with time stamp less than 15 can be later received before it can allow the time stamp 15 event to be processed.Thus, the principal task of any conservative protocol is to determine when it is “safe” to process an event, i.e., when can one guarantee no event containing a smaller time stamp will be later received by this LP. An LP cannot process an event until it has been guaranteed to be safe.2.1.1 First Generation AlgorithmsThe algorithms described in (Bryant 1977, Chandy and Misra 1978) were perhaps the first synchronization algo-rithms to be developed. They assume the topology indicat-ing which LPs send messages to which others is fixed and known prior to execution. It is assumed each LP sends messages with non-decreasing time stamps, and the com-munication network ensures that messages are received in the same order that they were sent. This guarantees that messages arriving on each incoming link of an LP arrive in timestamp order. This implies that the timestamp of the last message received on a link is a lower bound on the timestamp of any subsequent message that will later be re-ceived on that link.Messages arriving on each incoming link are stored in first-in-first-out order, which is also timestamp order be-cause of the above restriction. Local events scheduled within the LP can be handled by having a queue within each LP that holds messages sent by an LP to itself. Each link has a clock that is equal to the timestamp of the mes-sage at the front of that link’s queue if the queue contains a message, or the timestamp of the last received message if the queue is empty. The process repeatedly selects the link with the smallest clock and, if there is a message in that link’s queue, processes it. If the selected queue is empty, the process blocks. The LP never blocks on the queue con-taining messages it schedules for itself, however. This pro-tocol guarantees that each process will only process events in non-decreasing timestamp order.Although this approach ensures the local causality constraint is never violated, it is prone to deadlock. A cycle of empty links with small link clock values (e.g., smaller than any unprocessed message in the simulator) can occur, resulting in each process waiting for the next process in the cycle. If there are relatively few unproc-essed event messages compared to the number of links in the network, or if the unprocessed events become clus-tered in one portion of the network, deadlock may occur very frequently.Null messages are used to avoid deadlock. A null message with timestamp T null sent from LP A to LP B is a promise by LP A that it will not later send a message to LP B carrying a timestamp smaller than T null. Null messages do not correspond to any activity in the simulated system; they are defined purely for avoiding deadlock situations. Processes send null messages on each outgoing link after processing each event. A null message provides the re-ceiver with additional information that may be used to de-termine that other events are safe to process.Null messages are processed by each LP just like ordi-nary non-null messages, except no activity is simulated by the processing of a null message. In particular, processing a null message advances the simulation clock of the LP to the time stamp of the null message. However, no state variables are modified and no non-null messages are sent as the result of processing a null message.How does a process determine the timestamps of the null messages it sends? The clock value of each incoming link provides a lower bound on the timestamp of the next event that will be removed from that link’s buffer. When coupled with knowledge of the simulation performed by the process, this bound can be used to determine a lower bound on the timestamp of the next outgoing message on each out-put link. For example, if a queue server has a minimum ser-vice time of T, then the timestamp of any future departure event must be at least T units of simulated time larger than any arrival event that will be received in the future.Whenever a process finishes processing a null or non-null message, it sends a new null message on each outgo-ing link. The receiver of the null message can then com-pute new bounds on its outgoing links, send this informa-tion on to its neighbors, and so on. It can be shown that this algorithm avoids deadlock (Chandy and Misra 1978).The null message algorithm introduced a key property utilized by virtually all conservative synchronization algo-rithms: lookahead. If an LP is at simulation time T, and it can guarantee that any message it will send in the future will have a time stamp of at least T+L regardless of what mes-sages it may later receive, the LP is said to have a lookahead of L. As we just saw, lookahead is used to generate the time stamps of null messages. One constraint of the null message algorithm is it requires that no cycle among LPs exist con-taining zero lookahead, i.e., it is impossible for a sequence of messages to traverse the cycle, with each message sched-uling a new message with the same time stamp.2.1.2 Second Generation AlgorithmsThe main drawback with the null message algorithm is it may generate an excessive number of null messages. Con-sider a simulation containing two LPs. Suppose both are blocked, each has reached simulation time 100, and each has a lookahead equal to 1. Suppose the next unprocessed event in the simulation has a time stamp of 200. The null message algorithm will result in null messages exchanged between the LPs with time stamp 101, 102, 103, and so on. This will continue until the LPs advance to simulation time 200, when the event with time stamp 200 can now be proc-essed. A hundred null messages must be sent and proc-essed between the two LPs before the non-null message can be processed. This is clearly very inefficient. The problem becomes even more severe if there are many LPs.The principal problem is the algorithm uses only the current simulation time of each LP and lookahead to pre-dict the minimum time stamp of messages it could generate in the future. To solve this problem, we observe that the key piece of information that is required is the time stamp of the next unprocessed event within each LP. If the LPs could collectively recognize that this event has time stamp 200, all of the LPs could immediately advance from simu-lation time 100 to time 200. Thus, the time of the next event across the entire simulation provides critical information that avoids the “time creeping” problem in the null message algorithm. This idea is exploited in more ad-vanced synchronization algorithms.Another problem with the null message algorithm concerns the case where each LP can send messages to many other LPs. In the worst case, the LP topology is fully connected meaning each LP could send a message to any other. In this case, each LP must broadcast a null mes-sage to every other LP after processing each event. This also results in an excessive number of null messages.One early approach to solving these problems is an al-ternate algorithm that allows the computation to deadlock, but then detects and breaks it (Chandy and Misra 1981). The deadlock can be broken by observing that the mes-sage(s) containing the smallest timestamp is (are) always safe to process. Alternatively, one may use a distributed computation to compute lower bound information (not unlike the distributed computation using null messages de-scribed above) to enlarge the set of safe messages.Many other approaches have been developed. Some protocols use a synchronous execution where the computa-tion cycles between (i) determining which events are “safe’” to process, and (ii) processing those events. It is clear that the key step is determining the events that are safe to process each cycle. Each LP must determine a lower bound on the time stamp (LBTS) of messages it might later receive from other LPs. This can be deter-mined from a snapshot of the distributed computation as the minimum among:• the simulation time of the next event within each LP if the LP is blocked, or the current time of theLP if it is not blocked, plus the LP’s lookaheadand• the time stamp of any transient messages, i.e., any message that has been sent but has not yet beenreceived at its destination.A barrier synchronization can be used to obtain the snapshot. Transient messages can be “flushed” out of the system in order to account for their time stamps. If first-in-first-out communication channels are used, null messages can be sent through the channels to flush the channels, though as noted earlier, this may result in many null mes-sages. Alternatively, each LP can maintain a counter of the number of messages it has sent, and the number if has re-ceived. When the sum of the send and receive counters across all of the LPs are the same, and each LP has reached the barrier point, it is guaranteed that there are no more transient messages in the system. In practice, summing the counters can be combined with the computation for com-puting the global minimum value.To determine which events are safe, the distance be-tween LPs is sometimes used. This “distance” is the minimum amount of simulation time that must elapse for an event in one LP to directly or indirectly affect another LP, and can be used by an LP to determine bounds on the timestamp of future events it might receive from other LPs. This assumes it is known which LPs send messages to which other LPs. Full elaboration this technique is beyond the scope of the present discussion, however, these tech-niques and others are described in (Fujimoto 2000).Another thread of research in synchronization algo-rithms concerns relaxing ordering constraints in order to improve performance. Some approaches amount to simplyignoring out of order event processing (Sokol and Stucky 1990, Rao, et al. 1998). Use of time intervals, rather than precise time stamps, to encode uncertainty of temporal in-formation in order to improve the performance of time management algorithms have also been proposed (Fujimoto 1999a) (Beraldi and Nigro 2000). Use of causal order rather than time stamp order for distributed simula-tion applications has also been studied (Lee, et al. 2001).2.2 Optimistic SynchronizationIn contrast to conservative approaches that avoid violations of the local causality constraint, optimistic methods allow violations to occur, but are able to detect and recover from them. Optimistic approaches offer two important advan-tages over conservative techniques. First, they can exploit greater degrees of parallelism. If two events might affect each other, but the computations are such that they actually don’t, optimistic mechanisms can process the events con-currently, while conservative methods must sequentialize execution. Second, conservative mechanism generally rely on application specific information (e.g., distance between objects) in order to determine which events are safe to process. While optimistic mechanisms can execute more efficiently if they exploit such information, they are less reliant on such information for correct execution. This al-lows the synchronization mechanism to be more transpar-ent to the application program than conservative ap-proaches, simplifying software development. On the other hand, optimistic methods may require more overhead com-putations than conservative approaches, leading to certain performance degradations.The Time Warp mechanism (Jefferson 1985) is the most well known optimistic method. When an LP receives an event with timestamp smaller than one or more events it has already processed, it rolls back and reprocesses those events in timestamp order. Rolling back an event involves restoring the state of the LP to that which existed prior to processing the event (checkpoints are taken for this pur-pose), and “unsending” messages sent by the rolled back events. An elegant mechanism called anti-messages is provided to “unsend” messages.An anti-message is a duplicate copy of a previously sent message. Whenever an anti-message and its matching (positive) message are both stored in the same queue, the two are deleted (annihilated). To “unsend” a message, a process need only send the corresponding anti-message. If the matching positive message has already been processed, the receiver process is rolled back, possibly producing ad-ditional anti-messages. Using this recursive procedure all effects of the erroneous message will eventually be erased.Two problems remain to be solved before the above ap-proach can be viewed as a viable synchronization mecha-nism. First, certain computations, e.g., I/O operations, can-not be rolled back. Second, the computation will continually consume more and more memory resources because a his-tory (e.g., checkpoints) must be retained, even if no roll-backs occur; some mechanism is required to reclaim the memory used for this history information. Both problems are solved by global virtual time(GVT). GVT is a lower bound on the timestamp of any future rollback. GVT is computed by observing that rollbacks are caused by mes-sages arriving “in the past.” Therefore, the smallest time-stamp among unprocessed and partially processed messages gives a value for GVT. Once GVT has been computed, I/O operations occurring at simulated times older than GVT can be committed, and storage older than GVT (except one state vector for each LP) can be reclaimed.GVT computations are essentially the same as LBTS computations used in conservative algorithms. This is be-cause rollbacks result from receiving a message or anti-message in the LP’s past. Thus, GVT amounts to comput-ing a lower bound on the time stamp of future messages (or anti-messages) that may later be received.A pure Time Warp system can suffer from overly op-timistic execution, i.e., some LPs may advance too far ahead of others leading to excessive memory utilization and long rollbacks. Many other optimistic algorithms have been proposed to address these problems. Most attempt to limit the amount of optimism. An early technique involves using a sliding window of simulated time (Sokol and Stucky 1990). The window is defined as [GVT, GVT+W] where W is a user defined parameter. Only events with time stamp within this interval are eligible for processing. Another approach delays message sends until it is guaran-teed that the send will not be later rolled back, i.e., until GVT advances to the simulation time at which the event was scheduled. This eliminates the need for anti-messages and avoids cascaded rollbacks, i.e., a rollback resulting in the generation of additional rollbacks (Dickens and Rey-nolds 1990). A technique called direct cancellation is sometimes used to rapidly cancel incorrect messages, thereby helping to reduce overly optimistic execution (Fujimoto 1989, Zhang and Tropper 2001).Another problem with optimistic synchronization con-cerns the amount of memory that may be required to store history information. Several techniques have been devel-oped to address this problem. For example, one can roll back computations to reclaim memory resources (Jefferson 1990, Lin and Preiss 1991). State saving can be performed infrequently rather than after each event (Lin, et al. 1993, Palaniswamy and Wilsey 1993). The memory used by some state vectors can be reclaimed even though their time stamp is larger than GVT (Preiss and Loucks 1995).Early approaches to controlling Time Warp execution used user-defined parameters that had to be tuned to opti-mize performance. Later work has focused on adaptive approaches where the simulation executive automatically monitors the execution and adjusts control parameters to maximize performance. Examples of such adaptive controlmechanisms are described in (Ferscha 1995, Das and Fu-jimoto 1997), among others.Practical implementation of optimistic algorithms re-quires that one must be able to roll back all operations, or be able to postpone them until GVT advances past the simulation time of the operation. Care must be taken to ensure operations such as memory allocation and dealloca-tion are handled properly, e.g., one must be able to roll back these operations. Also, one must be able to roll back execution errors. This can be problematic in certain situa-tions, e.g., if an optimistic execution causes portions of the internal state of the Time Warp executive to be overwritten (Nicol and Liu 1997).Another approach to optimistic execution involves the use of reverse computation techniques rather than rollback (Carothers, et al. 1999). Undoing an event computation is accomplished by executing the inverse computation, e.g., to undo incrementing a state variable, the variable is in-stead decremented. The advantage of this technique is it avoids state saving, which may be both time consuming and require a large amount of memory. In (Carothers, et al. 1999) a reverse compiler is described to automatically generate inverse computations.2.3 Current State-of-the-ArtSynchronization is a well-studied area of research in the par-allel discrete event simulation field. There is no clear con-sensus concerning whether optimistic or conservative syn-chronization perform better; indeed, the optimal approach usually depends on the application. In general, if the appli-cation has good lookahead characteristics and programming the application to exploit this lookahead is not overly bur-densome, conservative approaches are the method of choice. Indeed, much research has been devoted to improving the lookahead of simulation applications, e.g., see (Deelman, et al. 2001). Otherwise, optimistic synchronization offers greater promise. Disadvantages of optimistic synchroniza-tion include the potentially large amount of memory that may be required, and the complexity of optimistic simula-tion executives. Techniques to reduce memory utilization further aggravate the complexity issue.Recently, synchronization algorithms have assumed an increased importance because of their use in the DoD High Level Architecture (HLA). Because the HLA is driven by the desire to reuse existing simulations, an important dis-advantage of optimistic synchronization in this context is the effort required to add state saving and other mechanism to enable the simulation to be rolled back.3 TIME PARALLEL SIMULATIONTime-parallel simulation methods have been developed for attacking specific simulation problems with well-defined objectives, e.g., measuring the loss rate of a finite capacity queue of an ATM multiplexer. Time-parallel algorithms divide the simulated time axis into intervals, and assign each interval to a different processor. This allows for mas-sively parallel execution because simulations often span long periods of simulated time.A central question that must be addressed by time-parallel simulators is ensuring the states computed at the “boundaries” of the time intervals match. Specifically, it is clear that the state computed at the end of the interval [T i-1,T i] must match the state at the beginning of interval [T i,T i+1]. Thus, this approach relies on being able to per-form the simulation corresponding to the ith interval with-out first completing the simulations of the preceding (i-1, i-2, ... 1) intervals.Because of the “state-matching” problem, time-parallel simulation is really more of a methodology for developing massively parallel algorithms for specific simulation prob-lems than a general approach for executing arbitrary dis-crete-event simulation models on parallel computers. Time-parallel algorithms are currently not as robust as space-parallel approaches because they rely on specific properties of the system being modeled, e.g., specification of the sys-tem’s behavior as recurrence equations and/or a relatively simple state descriptor. This approach is currently limited to a handful of applications, e.g., queuing networks, Petri nets, cache memories, and multiplexers in communication net-works. Space-parallel simulations offer greater flexibility and wider applicability, but concurrency is limited to the number of logical processes. In some cases, both time and space-parallelism can be used.One approach to solving the state matching problem is to have each processor guess the initial state of its simula-tion, and then simulate the system based on this guessed initial state (Lin and Lazowska 1991). In general, the initial state will not match the final state of the previous interval. After the interval simulators have completed, a “fix-up” computation is performed to account for the fact that the wrong initial state was used. This might be performed, for instance, by simply repeating the simulation, using the fi-nal state computed in the previous interval as the new ini-tial state. This “fix-up” process is repeated until the initial state of each interval matches the final state of the previous interval. In the worst case, N such iterations are required when there are N simulators. However, if the final state of each interval simulator is seldom dependent on the initial state, far fewer iterations will be needed.In (Heidelberger and Stone 1990) the above approach is proposed to simulate cache memories using a least-recently-used replacement policy. This approach is effec-tive for this application because the final state of the cache is not heavily dependent on the cache’s initial state. A variation on this approach devised in the context of simu-lating statistical multiplexers for asynchronous transfer mode (ATM) switches precomputes certain points in time where one can guarantee that a buffer overflow (full。
工业类英语词汇_60
mass stability,质量稳定性mass storage,器mass-to-charge ratio,质荷比master file,主文卷master/slave discrimination,副鉴别master station,主站master viscometer,标准粘度计material measure,实体量器material processibility,材料工艺性能material tesing machine,材料试验机mathematical model,数学模型mathematical similarity,数学相似mathematical simulation,数字仿真matrix correction,基本修正matrix effect,基体效应matrix printer,Mattauch-Herzog geometry(mass spectrograph),马-赫型双聚焦质谱法max allowable continuous working current,最大允许连续工作电流max deflection of linearity,最大线性偏转maximum acceleration,最大加速度maximum allowde deviation,最大允许扁差maximum ballistic scanning,最大冲击拂掠maximum capacity,最大称量maximum cyclic load,最大循环负荷maximum cyclic stress,最大循环应力maximum displacement,最大位移maximum excitation,最大激励maximum floating voltage,最大浮置电压maximum flow-rate,最大流量maximum load of the test,最大试验负荷maximum load of the testing machine,试验机最大负荷maximum operating pressure differential,最大工作压差maximum operationg water depth,最大工深度maximum output inductance,最大输出电感maximum output resistance,最大输出电阻maximum overshoot,最大超调量maximum peneration power,最大穿透力maximum power supply voltage,最高电源电压maximum principle,原理maximum profit programming,最大利润规划maximum rated circumferential magnetizing current, 额定maximum rated force under sinusoidal conditions,正弦态最大激振力maximum revolutions of output shaft,输出轴最大转数maximum scale value,标度终点值maximum sound pressure level of microphone,传声器最高声压级maximum strain,最大应变maximum temperature,最高温度maximum thermometer,高温度表maximum transverse load,最大横向负荷maximum velocity,最大速度maximum wind speed,最大风速maximum working pressure(MWP),最大工作压力McLeod vacuum gauge,麦氏真空计mean availability,平均轴就流体速度mean dynamic pressure in a cross-section,横截面内的平均动压(mean) effective mavelength,效波长mean flow-rate,平均流量mean life,平均寿命mean linear velocity of mobile phase,流动相平均线速mean load,平均负荷mean repair time(MRT),平均修理时间mean squared spectral density,均方谱密度mean strain,平均应变mean stress,平均应力mean time between failures(MTBF),平均失效间隔时间mean time to failure(MTTF),平均失效前时间mean time to restoration,平均体膨胀系数meantime auto-spectrometer,同时式自动光谱仪(measurable)quantity,(measurand,被测量measured object,被测对象measured quantiry,被测量(measure)target,measured value,被测值measured variable,被测变量measurement,测量measurement hardware,测量硬件measurement of directional response pattern,指向性响应图案测量measurement of exciting force,激振力的测量measurement of vibration quantity,振动量的测量measurement procedure,测量步骤measurement signal,测量信号measurement standand,measurement time,测量时间measuring amplifier,测量放大器measuring bridge,测量电桥measuring current transformer,测量用电流互感器measuring distance,测量距离measuring element(of an electro-mechanical measuring in strument), -机构measuring equipment,备measuring hole,测量孔measuring indication system,测量指示装置measuring instrument,measuring instrument with circuit control device,带有电路控制器件的测量仪表measuring junction,测量端区measuring microphone,测试传声器measuring plane,测量平面measuring point for the humidity,湿度测定点measuring point for the temperature,温度测定点(measuring)potentiometer,差计measuring range,测量范围measuring range higher limit,测量范围上限值measuring range lower limit,测量范围下限值measuring section,测量段measuring spark gap,测量球隙measuring system,测量系统measuring terminal,测量端measuring time,测量时间(measuring)transducer,measuring transducer(with electrical output),量变换器measuring voltage transformer,测量用电压互感器mechanical bathythermograph(MBT),机械式深温计mechanical hygrometer,机械湿度计mechanical impedance,机械阻抗mechanical properties,机械性能mechanical quantity,机械量mechanical quantity transducer[sensor],力学量传感器mechanical regulator,机械稳速器mechanical resonance,机械共振mechanical resonance frequency of the moving element,运动部件机械共振频率mechanical resonance frequency of the moving element suspension, 运动部件悬挂机械共振频率mechanical runout,机械脱出mechanical sensor,力敏元件mechanical shock,机械冲击mechanical strain,机械应变mechanical structure type transducer[sensor],结构型传感器mechanical test,机械性能试验mechanical testing machine,机械式试验机mechanical top-loading balance,机械式上皿天平mechanical vibration,(mechanical vibrator,机械振动器mechanical vibraometer,机械测振仪mechanical zero,机械零位mechanical zero adjuxter,机械零位调节器mechanism model,机理模型medium temperature strain gauge,中温应mel,melted quartz cqpacitor,熔融石英电容器melting heat,熔解热melting point,熔解点melting point type disposable fever thermeometer,熔点型消耗式温度计memory,存储器memory protection,存储保护Mendeleev weighing,门捷列夫称量法meniscus,弯月面menu selection mode,选择式meroury barometer,水银气压表mercury drop amplitude,汞滴振幅mercury motor meter,水银电机式仪表mercury pool electrode,示池电极mercury thermoneter,水银温度表message,报文message mode,报文方式message switching,报文交换messenger,使锤metal base indicated electrode,金属基指示电极metal-ceramic X-ray tube,金属陶瓷X射线管metal-insoluble salt indicated electrode,金属-难溶盐指示电极metal-oxide gas transducer[sensor],金属氧化物气体传感器metal-oxide humidity transducer[sensor],金属氧化物湿度传感器metal-spring gravimeter,金属弹簧重力仪metallic material testing machine,金属材料试验机metallurgical automation,冶金自动化metastable dceomposition,亚稳分解metastable defocussing,亚稳去聚焦metastable ion,亚稳离子metastable scanning,亚稳扫描meteorograph,气象计meteorological instrument,气象仪器meteorological observation,气象观测meteorological radar,气象雷达meteorological rocket,气象火箭meteorological satellite,气象卫星meteorological tower,气象塔meter,。
大学物理下册常用英文单词
大学物理下册常用英文单词及短语Mechanical Oscillation（vibration）机械振动Electromagnetic Oscillation 电磁振荡Simple Harmonic Motion (SHM) simple harmonic oscillator 简谐振动Superposition 合成、叠加Spring 弹簧Period 周期frequency Angular frequency 频率、角频率Phase initial phase 相位、初相rotating vector 旋转矢量vertical directions 垂直方向Mechanical Wave 机械波Plane Harmonic Waves 平面简谐波Propagation 传播Huygen’s Principle 惠更斯原理Interference of Waves 波的干涉Standing Wave (Stationary Waves) 驻波traveling waveDoppler Effect 多普勒效应Medium 介质Transverse waves 横波Longitudinal waves 纵波:Wave Front：(波前)Wavelength 波长Crest 波峰trough 波谷Wave speed 波速Randomly 随机Position 位置Displacement 位移wave equation 波动方程oscillating curve 振动曲线kinetic, potential and total energy 动能、势能和总能量energy flow density 能流密度spherical waves 球面波Principle of Independent Propagation 波的独立传播原理Principle of Superposition of Waves 波的叠加原理Coherent wave 相干波phase difference 相位差wave path difference 波程差constructive 干涉相长destructive 干涉相消interface draft(干涉图样)Node 波节Antinode 波幅important features: 重要性质Half-Wave Loss 半波损失wavy thinner medium 波疏介质wavy denser medium 波密介质toward 向着far away 背离Optics 光学Visible light可见光light vector 光矢量Luminescence 发光Interference of Light 光的干涉Coherence Light 相干光incoherent superposition 非相干叠加coherent superposition 相干叠加Optical Distance 光程Optical Path Difference 光程差Film Interference 薄膜干涉Reflection反射Refraction折射Diffraction 衍射Slit 缝Fraunhofer 夫朗和费Grating 光栅Polarized Light 偏振光stable interference pattern 稳定杆射花样light intensity 光强Young’s Interference 杨氏干涉central bright fringe 中央明纹first-order bright fringes 第一级明纹adjacent bright (or dark) fringes 相邻明暗纹Lloyd’s mirror experiment 络唉镜实验the refractive index 折射率Thin Lens 薄透镜parallel light 平行光in-phase 同相位primary optical axis 主光轴secondary optical axis 副光轴primary focus 主焦点focal plane 焦平面optical center 光心Equal-inclination interference, isoclinal interference 等倾干涉Film 薄膜Equal-thickness interference(等厚干涉)Wedge Interference 劈尖干涉bright or dark fringes 明暗纹distance 距离Newton’s ring 牛顿环Michelson’s Interferometer 迈克尔逊干涉仪Huygens-Fresnel’s Principle 惠更斯菲涅尔原理incident lights 入射光Fresnel Diffraction 菲涅尔衍射Vertical 垂直的center bright fringe 中央明纹Circular Hole Diffraction 圆孔衍射Ariy disk 艾丽斑Rayleigh criterion 瑞丽判据Ruling 刻痕glass plate 平板玻璃grating constant 光栅常数diffraction angle 衍射角grating equation 光栅方程Primary maximum fringe 主机大条纹spectrum line 谱线visible frings 可见条纹incident normally 垂直入射missing order 缺级Dark fringe 暗纹equal to 等于Grating spectrum 光栅光谱Oblique incidence 斜入射Nature Light 自然光Polarized Light 偏振光Linearly Polarized Light 线偏振光plane polarized light 平面偏振光Partially Polarized Light 部分偏振光Polarizer 起偏、偏振片Analyser 检偏The polarized direction 偏振方向Malus Law 马吕斯定律Extinction消光Snell’s Law (refraction law)死聂耳折射定律Brewster’s Law 布如斯特定律full polarization angle全偏振角or Brewster angle布儒斯特角the Kinetic Theory of Gases 气体动理论Equilibrium State 平衡态Ideal Gas 理想气体Equipartition Theory of Energy 能量均分理论Internal energy 内能Maxwell Speed Distribution 麦克斯韦速率分布Boltzmann Distribution 波尔滋蔓速率分布The Mean Free Path 平均自由程vast number of molecular 大量分子Thermodynamics 热力学Thermology 热学macroscopic quantities 宏观态microcosmic quantities 微观态Avogadro’s constant 阿伏伽德罗常量Brownian Motion 布朗运动Attractive 吸引Repulsive 排斥isolated System 孤立系统State Parameters 状态参量Ideal Gas Equation of State 理想气体状态方程geometric parameter 几何参数mechanical parameter 力学参数thermal parameter 热学参数fixed 固定的The root-mean-square(RMS) speed 方均根速率The degrees of freedom of molecule 分子自由度monatomic gas 单原子分子diatomic gas双原子分子polyatomic gas 多原子分子Energy Equipartition theorem 能量均分定理degree of translational freedom 平动自由度translational kinetic energy 平动动能average kinetic energy 平均动能rotational kinetic energy 转动动能thermodynamic temperature 热力学温度speed distribution function 速率分布函数the mean value 平均值collisions 碰撞Normalization归一化Curve 曲线the most probable speed 最可几速率average speed 平均速度The mean free path 平均自由程The First Law of Thermodynamics 热力学第一定律Adiabatic Process绝热过程Carnot Cycle 卡诺循环Cyclical Processes 循环过程The Second Law of Thermodynamics热力学第二定律Reversible可逆的& Irreversible不可逆的ProcessStatistical Meaning 统计意义Entropy 熵Quasistataic process 准静态过程Work功、Heat热、Internal Energy 内能Piston 活塞Integration 积分Isochoric process等容过程Isobaric process等压过程Isothermal Processes等温过程Heat Capacities 热容Specific Heat capacity比热容molar heat capacity 摩尔热熔Mayer’s formular 迈耶公式Adiabatic equation 绝热过程adiabatic curve 绝热曲线isotherm curve 等温曲线heat engine 热机p-V diagram p-V图original state 初态Clockwise 正循环Anticlockwise 逆循环Refrigerator 电冰箱The efficiency of heat engines 热机效率refrigeration coefficient 制冷系数Carnot cycle 卡诺循环isothermal expansion 等温膨胀adiabatic expansion 绝热膨胀compression 压缩Kelvin statement开尔文说法Clausius statement克劳修斯说法work transfers into heat 功热转换Carnot Theorem 卡诺定理Statistical Meaning 统计平均。
随机模拟(仿真)-simulation
计算结果如下：
>> a=2;b=3;n=30000; >> I=jifen1(a,b,n) I = 2.6912
syms x >> int((1+x^2)^(1/2),2,3) ans = 3/2*10^(1/2)-1/2*log(-3+10^(1/2))-5^(1/2)1/2*log(2+5^(1/2)) >> double(a编写M文件： function p=fangzhenguji(N) x1=unifrnd(2,5,N,1); x2=exprnd(3,N,1); x3=normrnd(3,2,N,1); x4=normrnd(1,1,N,1); n=0; for k=1:N y1=x1(k)+x2(k)^2; y2=x3(k)+x4(k)^2; if y1>=3&y2<=9 n=n+1; else n=n; end End p=n/N;
模拟的一般步骤
• 明确问题，建立模型：正确描述研究的问题，明确规定模 拟的目标和任务，确定衡量系统性能或模拟输出结果的目 标函数，然后根据系统的结构及作业规则，分析系统各状 态变量之间的关系，以次为基础建立所研究的系统模型； • 收集和整理数据资料：模拟的实现往往离不开大量数据的 输入，且需要确定随机因素的概率分布特性，并以此为抽 样的根据； • 编制程序：模拟运行，选择适当的计算机语言，按照系统 数学、逻辑模型编写计算机程序。 • 分析模拟输出结果：一般包括如下几个方面 • （1）模拟结果的统计特性：均值、方差以及置信区间； • （2）灵敏度分析； • （3）根据确定的目标函数，在众多的实现方案中选取最 优方案。
2、先到先服务；
仿真步骤：
Contact of single asperities with varying adhesion
a r X i v :c o n d -m a t /0606588v 1 [c o n d -m a t .m t r l -s c i ] 22 J u n 2006Contact of Single Asperities with Varying Adhesion:ComparingContinuum Mechanics to Atomistic SimulationsBinquan Luan and Mark O.Robbins Department of Physics and Astronomy,The Johns Hopkins University,3400N.Charles Street,Baltimore,Maryland 21218(Dated:March 22,2006)Abstract Atomistic simulations are used to test the equations of continuum contact mechanics in nanome-ter scale contacts.Nominally spherical tips,made by bending crystals or cutting crystalline or amorphous solids,are pressed into a flat,elastic substrate.The normal displacement,contact radius,stress distribution,friction and lateral stiffness are examined as a function of load and adhesion.The atomic scale roughness present on any tip made of discrete atoms is shown to have profound effects on the results.Contact areas,local stresses,and the work of adhesion change by factors of two to four,and the friction and lateral stiffness vary by orders of magnitude.The microscopic factors responsible for these changes are discussed.The results are also used to test methods for analyzing experimental data with continuum theory to determine information,such as contact area,that can not be measured directly in nanometer scale contacts.Even when the data appear to be fit by continuum theory,extracted quantities can differ substantially from their true values.PACS numbers:81.40.Pq 68.35.Np 62.20.Dc 68.37.PsI.INTRODUCTIONThere has been rapidly growing interest in the behavior of materials at nanometer scales [1].One motivation is to construct ever smaller machines [2],and a second is to improve material properties by controlling their structure at nanometer scales [3].For example,decreasingcrystallite size may increase yield strength by suppressing dislocation plasticity,and material properties may be altered near free interfaces or grain boundaries.To make progress,this research area requires experimental tools for characterizing nanoscale prop-erties.Theoretical models are also needed both to interpret experiments and to allow new ideas to be evaluated.One common approach for measuring local properties is to press tips with characteristic radii of 10to 1000nm into surfaces using an atomic force microscope (AFM)or nanoindenter[4,5,6,7,8,9,10,11,12,13,14,15].Mechanical properties are then extracted from the measured forces and displacements using classic results from continuum mechanics [16].A potential problem with this approach is that continuum theories make two key assumptions that must fail as the size of contacting regions approaches atomic dimensions.One is to replace the atomic structure in the bulk of the solid bodies by a continuous medium with internal stresses determined by a continuously varying strain field.The second is to model interfaces by continuous,differentiable surface heights with interactions depending only on the surface separation.Most authors go further and approximate the contacting bodies by smooth spheres.In a recent paper [17],we analyzed the limits of continuum mechanics in describing nanometer scale contacts between mental probes.As in studies of other bulk could be described by continuum atomic diameters.However,the atomic structure of surfaces had profound consequences for much larger contacts.In particular,atomic-scale changes in the configuration of atoms on nominally cylindrical or spherical surfaces produced factor of two changes in the width of the contacting region and the stress needed to produce plastic yield,and order of magnitude changes in friction and stiffness.In this paper we briefly revisit non-adhesive contacts with an emphasis on the role of surface roughness.We then extend our atomistic studies to the more common case of通常 AFM 的针尖直径在10nm~1000nm 之间。
simulation的形容词
simulation的形容词simulation simulationsSimulation, as an adjective, refers to the imitation or representation of a situation, process, or system in order to gain insight into its functioning or behavior. This powerful tool has become increasingly popular across various fields and industries, enabling practitioners to explore and experiment with different scenarios without the associated risks and costs of real-world implementation. By replicating the complexities of the real world in a controlled environment, simulations allow for analysis, learning, and optimization, ultimately leading to improved decision-making, planning, and problem-solving.One of the defining characteristics of simulations is their ability to accurately replicate real-world processes and systems. These simulations strive for high fidelity, meaning that they closely resemble their real-world counterparts in terms of behavior, physics, and interactions. Through sophisticated mathematical algorithms and computational models, simulations can accurately reproduce the intricate dynamics and relationships of complex systems, such as economies, traffic networks, or natural environments. This high level of realism allows practitioners to study and understand the underlying mechanisms and emergent patterns that drive these systems, enabling them to make informed decisions and design effective interventions.In addition to accuracy, simulations also offer the advantage of flexibility and adaptability. Unlike real-world experiments or trials, simulations can be easily modified, adjusted, and repeated toexplore different scenarios and hypotheses. This flexibility allows practitioners to test a wide range of variables and parameters, examine the influence of various factors, and uncover hidden patterns or relationships that may not be immediately apparent. Furthermore, simulations can simulate time at an accelerated rate, enabling practitioners to observe long-term trends, assess the impact of policy changes, or predict future outcomes. This temporal compression allows for a more efficient and comprehensive analysis, leading to more effective strategies, policies, and interventions.Simulations are particularly valuable in situations where conducting real-world experiments is impractical, costly, or unethical. In fields such as aerospace engineering, for example, simulations are extensively used to model and predict the behavior of complex aircraft systems, saving time, resources, and minimizing risks. Similarly, in medicine, simulations play a crucial role in training surgeons, allowing them to practice and refine their techniques in a safe and controlled environment. By providing a realistic yet risk-free platform for learning and experimentation, simulations contribute to improving performance, enhancing skills, and increasing confidence among practitioners.Moreover, simulations have proven to be invaluable in the field of disaster preparedness and response. By recreating different disaster scenarios, such as earthquakes, floods, or pandemics, simulations enable emergency management teams to develop and evaluate strategies, test response plans, and identify areas of improvement. These simulations can simulate the behavior of populations, infrastructure, and supply chains, helping decision-makersanticipate potential bottlenecks, allocate resources efficiently, and coordinate response efforts effectively. Furthermore, simulations can also be used to educate and raise awareness among the public, providing them with a firsthand experience of the potential impacts of a disaster and empowering them to make informed decisions to mitigate risks and protect themselves.Beyond engineering and emergency management, simulations are widely used in business, finance, and economics to model and analyze complex market dynamics, investment strategies, and economic policies. These simulations enable market analysts, financial institutions, and policymakers to study the behavior of markets, predict trends, and optimize investment portfolios. By simulating various economic scenarios and policy alternatives, simulations can help identify potential risks, optimize resource allocation, and inform the formulation of effective regulatory measures. This application of simulations in economics and finance is particularly valuable in a rapidly changing and interconnected global economy, where accurate predictions and informed decisions are paramount.Simulations are also transforming the field of education by providing immersive and interactive learning experiences. Virtual reality (VR) simulations, for example, offer students the opportunity to explore historical events, travel to distant countries, or conduct scientific experiments in a virtual environment. This hands-on approach enables students to actively engage with the subject matter, enhance their understanding, and develop critical thinking and problem-solving skills. Additionally, simulations can simulate complex systems, such as ecosystems or climate models,allowing students to observe and experiment with the impacts of different variables, fostering a deep understanding of environmental dynamics and sustainability challenges.While simulations offer numerous benefits, it is important to acknowledge their limitations and potential pitfalls. Simulations are only as accurate and reliable as the data and assumptions they are based on. Inaccurate or incomplete data can lead to biased or misleading results, compromising the validity of the simulation. Furthermore, simulations are inherently simplified representations of complex reality, making it crucial to carefully select and parameterize the simulation model. The quality of the simulation model and the expertise of the practitioners using it are key determinants of the simulation's effectiveness and insights it can provide.In conclusion, simulation is a versatile and powerful tool that has revolutionized the way we study, analyze, and understand complex systems and phenomena. With their ability to accurately replicate real-world processes, simulations allow practitioners to explore different scenarios, test hypotheses, and optimize decision-making. Through their flexibility and adaptability, simulations enable a wide range of applications across various fields, from engineering and emergency management to economics and education. However, it is important to approach simulations with caution, ensuring the accuracy and validity of the underlying data and assumptions. By leveraging the full potential of simulations and continuously improving their accuracy and reliability, we can harness their transformative power to drive innovation, enhance learning, and solve complex challenges facing our world.。
SimulationXpress基本概念
SimulationXpress 基本概念1、屈曲以承受高压应力的结构单元忽然失效为特征的一种破坏形式，在这种破坏中真实的压应力小于材料能承受的最大压应力。
这种失效模式也被描述为由于弹性失稳而造成的失效。
2、热膨胀系数义为温度每变化1℃所引起的单位长度的改变（即每变化1℃所引起的法向应变的变化）。
3、蠕变述固体材料在应力的作用下慢慢移动或逐渐变形的趋势，作为长期承受低水平应力的结果出现。
该应力低于材料的屈服应力或极限应力。
长期受热并且接近熔点的材料蠕变更为严重。
4、自由度组用来完全确定物体或系统平移或变形方位的独立平移量和旋转量。
在机械工程、航空工程、机器人技术、结构工程等领域，它是关系运动物体系统的基本概念。
物体有6个自由度：3个平移和3个旋转。
5、密度位体积的质量。
密度的单位在英制中式3/in lb ，在公制中式㎏/m ³。
密度用在静态、非线性、频率、动态、屈曲和热分析中。
静态和屈曲分析中只有在定义重力（或离心力）时才用这个概念。
6、延展性述材料在发生断裂之前延伸变形成棒状能力的机械特性。
高延展性材料包括银、金、铜和铝等。
钢的延展性随合金成分而变化。
增加含碳量能降低延展性（比如钢变脆）。
7、弹性模量于弹性材料，弹性模量是在材料中产生单位应变所加的应力，即应力除以应变。
弹性模量常称为杨氏模量。
8、疲劳材料承受循环载荷时出现的逐渐的、局部的结构破坏。
最大应力应低于材料的拉伸极限，还应低于材料的屈服应力。
9、固定约束于物体来说，这种约束把所有平移自由度都设置为0。
对于壳和梁，平移和旋转自由度均设置为0。
对于桁架连接点，设置平移自由度为0。
使用这种约束，不需要几何参考体。
10、力于与另一物体的相互作用，而在一个物体上产生的推拉效果。
只要有两个物体间的相互作用，力就在每个物体上产生。
相互作用停止，两个物体不再承受力。
力只能作为相互作用的结果存在。
例如，如果选择了3个面并指定了50lb的力，SimulationXpress一共施加了150lb 的力（每个面上为50lb）。
lammps原子数量 -回复
lammps原子数量-回复"lammps原子数量"是指使用分子模拟软件LAMMPS进行分子动力学模拟时，在系统中所包含的原子的数量。
原子数量的选择对于模拟结果的准确性和计算资源的运用都有重要影响。
本文将逐步解答关于LAMMPS 原子数量的相关问题。
第一步：什么是LAMMPS？LAMMPS（Large-scale Atomic/Molecular Massively Parallel Simulator）是一个基于分子动力学的开源软件包，用于模拟原子、分子和离子在固体和液体材料中的动力学行为。
第二步：原子数量对模拟结果的影响如何？原子数量对于分子动力学模拟的模拟结果有着重要的影响。
在许多情况下，模拟结果在更大的原子数量下更加准确，因为更多的原子数可以提供更精确的平均值和统计力学的近似。
更多的原子数量也能够更好地反映系统的热力学性质，例如温度、压力和自由能等。
第三步：如何确定合适的原子数量？确定合适的原子数量需要考虑多个因素，其中包括系统的大小、所研究问题的性质以及计算资源的可用性。
以下是一些常见的方法和准则：1. 与实验数据对比：在进行模拟之前，首先应与实验数据进行对比，以确定所研究的系统中的原子数量范围。
通过与实验数据的比较，可以估计所需的原子数量，以保持模拟结果的准确性。
2. 系统的特性：系统的性质对原子数量的选择也有影响。
例如，如果研究的是局部结构和表面性质，较小的原子数量可能就足够了。
相反，如果研究的是宏观性质，如热导率和热膨胀等，较大的原子数量可能会更准确。
3. 计算资源：原子数量对所需的计算资源有直接影响。
较大的原子数量需要更多的内存和计算时间。
因此，通常需要在可用的计算资源和模拟所需的准确性之间找到一个平衡。
第四步：常见的原子数量范围是多少？常见的原子数量范围取决于所研究的系统和相应的问题。
一般而言，原子数量可以从几百到几百万不等。
以下是一些示例：1. 小型系统：对于小型系统，如表面吸附或小分子的动力学行为研究，通常使用几百到几千个原子。
simulation中质量速度
simulation中质量速度模拟（simulation）是通过建立数学模型来模拟实际系统的运行过程，以便预测和理解系统行为的方法。
在模拟中，质量（mass）和速度（velocity）是两个重要的概念。
质量是物体所拥有的惯性和引力特性的属性，速度则表示物体在某个时间段内的位移程度。
在本文中，将探讨质量和速度在模拟中的作用和意义，并阐述它们在不同领域中的应用。
首先，质量在模拟中起到了至关重要的作用。
质量是物体所具有的一种物理属性，它决定了物体对力的响应。
在数学模型中，质量通常以标量的形式表示，并用单位千克（kg）进行量化。
在模拟过程中，质量可以被看作是系统的一个参数，它直接影响到系统的动态响应。
在一些复杂的模拟中，质量可能会以分布的形式出现，这使得模拟结果更加真实和准确。
其次，速度也是模拟中一个非常重要的概念。
速度是物体在某一时刻的位移程度，通常用矢量的形式表示，并用单位米每秒（m/s）进行量化。
在数学模型中，速度通常与时间有关，可以通过对位移与时间的导数来求得。
速度的概念在模拟中与质量密切相关，它们共同决定了物体的运动状态和变化速率。
在一些复杂的模拟中，速度可能会以随时间变化的函数的形式出现，这使得模拟结果更加丰富和多样化。
将质量和速度结合起来，可以更好地理解模拟中的物理现象和系统运行。
例如，在电气工程中，质量和速度的概念可以应用于模拟电路中的电流和电压变化。
通过建立电路的数学模型，可以推导出电路中元件的电流和电压随时间的变化规律。
质量可以被理解为电容器或感应器的参数，它决定了电流和电压的响应速度。
速度则表示电流和电压的变化速率，通过分析模拟结果，可以预测和改进电路的工作性能。
在工程力学中，质量和速度的概念可以应用于模拟材料的变形和应力分布。
通过建立材料的数学模型，可以推导出应力和应变随时间和空间的变化规律。
质量可以被理解为材料的密度或惯性系数，它决定了材料的应力响应能力。
速度表示材料的变形速率，通过模拟可以分析材料的破坏机制和优化设计。
ADAMS中的装配、静态、运动学、动力学仿真
About Adjusting Your Model Before SimulationBefore you begin your simulation, you may want to do one or more preliminary operations to help ensure a better simulation. You can do any of the following:• Check to see if you have the expected number of movable parts and the expected number and type of constraints in your model.• Determine the total number of system degrees of freedom (DOF) and which, if any, constraint equations are redundant. Learn more .• Check to see if any constraints are broken or incorrectly defined and, if so, perform an initial conditions simulation on your model to try to correct these broken joints. Learn more .• Perform a static simulation to move your model into an equilibrium configuration immediately before performing a dynamic simulation to reduce some of the initial, transient system response.• Calculate the natural frequencies of your model as linearized about a particular operating configuration. Learn morePerforming Initial Conditions SimulationYou can perform an initial conditions simulation to check for any inconsistencies in your model. The initial conditions simulation is often referred to as an assemble model operation. An initial conditions simulation tries to reconcile any positioninginconsistencies that exist in your model at its design configuration and make it suitable for performing a nonlinear or linear simulation. Most importantly, the initialconditions simulation tries to ensure that all joint connections are defined properly. For example, for a revolute joint to be defined properly, the origins of the markers that define the joint must be coincident throughout a simulation. If the markers are not coincident, the joint is broken and needs to be repaired. In this example, the initial conditions simulation helps repair the broken revolute joint by moving the origins of the two markers until they are coincident, as shown in the following figure.Consistent Gears that Become InconsistentIn the case of the door with two hinges, Adams/Solver ignores five of the constraint equations that it finds redundant. You do not know which equations Adams/Solver ignores, however. If Adams/Solver ignores all of the equations corresponding to one of the hinges, then all the reaction forces are concentrated at the other hinge in the Adams/Solver solution. Adams/Solver arbitrarily sets the reaction forces to zero at the redundant hinge. But Adams/Solver might not discard all the equations for one hinge and retain all the equations from the other. It might just as easily retain one or more equations from each, and discard one or more from each.Although Adams/Solver still provides the physically correct solution, the simulation may require extra computational effort to constrain the motion when all of the constraint forces and torques are concentrated at one end of the door. Consequently, it is always a good idea to carefully select your constraints and define models without any redundancies. For example, you can construct the model of the door with a spherical joint and a parallel-axes constraint instead of the single revolute joint.Door Frame with Spherical and Parallel-axes ConstraintsWhen you verify your model or run a simulation, Adams/Solver tells you which constraints are redundant. To solve the redundancy, try replacing a redundant idealized joint with a joint primitive. You may also want to replace redundant constraints with approximately equivalent flexible connections.Adams/Solver does not always check the initial conditions set for a constraint when it performs overconstraint checking. If you apply a motion on one joint and initial conditions on another joint, check to ensure that they are not redundant because Adams/Solver does not check them for redundancy and your model may lock up when simulation begins. As a general rule, do not specify more initial conditions than the number of DOF in your model. For more on initial conditions for joints, see Setting Initial Conditions.Examples of Redundant Constraint MessagesThe following sections provide examples of redundant constraint messages and ways to avoid the redundancies:• Example 1 - Converting a Revolute to a Spherical• Example 2 - Converting a Translation to an Inline• Example 3 - Removing Redundancies from Fourbar MechanismExample 1 - Converting a Revolute to a SphericalIf in your model, Joint_7 is a revolute joint, and Adams/View gives you the following warning messages, then you have two redundant constraint equations:Joint_7 unnecessarily removes Rotation Between Zi and XjJoint_7 unnecessarily removes Rotation Between Zi and YjThese messages indicate that the rotational constraint equations 4 and 5 that the revolute joint introduces are not needed. Therefore, you could replace the revolute joint with a spherical joint since it does not use these equations.Example 2 - Converting a Translation to an InlineIf in your model, Joint_29 is a translational joint, and Adams/View displays the following warning messages, then you could change Joint_29 from a translational joint to an inline joint to remove the redundancies:Joint_29 unnecessarily removes Rotation Between Zi and XjJoint_29 unnecessarily removes Rotation Between Zi and YjJoint_29 unnecessarily removes Rotation Between Xi and YjExample 3 - Removing Redundancies from Fourbar MechanismIf you build a fourbar mechanism with four revolute joints, Adams/View displays messages similar to the following:Joint_1 unnecessarily removes Rotation Between Zi and XjJoint_1 unnecessarily removes Rotation Between Zi and YjJoint_3 unnecessarily removes Rotation Between Zi and XjThese messages indicate that you could change Joint_1 from a revolute joint to a spherical joint, and change Joint_3 from a revolute joint to a universal or Hooke joint. By changing the joint types, you eliminate the redundant constraint warnings and possibly improve the performance of your solution.Alternatively, you could also remove the redundancies by changing just one of the revolute joints to an inline joint. There is almost always more than one way to remove redundant constraints. The best way is to select joint types so they match the way your physical system can move. Some of the possible configurations are shown in the figure below.Alternative Configurations for Fourbar MechanismRemember that Adams/Solver does not calculate joint reaction forces in any directions associated with redundant constraint equations because it automatically removes these equations when it performs a simulation. Therefore, you may also want to select your joint types based on where you want to measure joint reaction forces.Performing Static Equilibrium SimulationsWhen you perform a static equilibrium simulation on your model, Adams/Solver iteratively repositions all parts in an attempt to balance all the forces for one particular point in time.To learn more:• About Performing Static Equilibrium Simulations• Finding Static Equilibrium for Your Model• About Performing Dynamic Simulations to Find Static EquilibriumPerforming Initial Conditions SimulationYou can perform an initial conditions simulation to check for any inconsistencies in your model. The initial conditions simulation is often referred to as an assemble model operation. An initial conditions simulation tries to reconcile any positioning inconsistencies that exist in your model at its design configuration and make it suitable for performing a nonlinear or linear simulation. Most importantly, the initial conditions simulation tries to ensure that all joint connections are defined properly. For example, for a revolute joint to be defined properly, the origins of the markers that define the joint must be coincident throughout a simulation. If the markers are not coincident, the joint is broken and needs to be repaired. In this example, the initial conditions simulation helps repair the broken revolute joint by moving the origins of the two markers until they are coincident, as shown in the following figure.Repaired Revolute JointYou can also use the initial conditions simulation if you are creating parts in exploded view. Exploded view is simply creating the individual parts separately and then assembling them together into a model. You might find this convenient if you have several complicated parts that you want to create individually without seeing how they work together until much later. Adams/View provides options for specifying that you are creating your model in exploded view as you create constraints.To perform an initial conditions simulation:• From the Simulation Controls dialog box, select the Initial Conditions tool. Adams/View tells you when it has assembled your model properly. You can revert back to your original design configuration or you can save your assembled model as the new design configuration for your model. For more information on how to do this, see Saving a Simulation Frame.相关主题PopupPopup另请参阅Popup（注：可编辑下载，若有不当之处，请指正，谢谢!）。