matlab 外文翻译 外文文献 英文文献 MATALAB 混合仿真平台控制算法的概述

《matlab 外文翻译 外文文献 英文文献 MATALAB 混合仿真平台控制算法的概述》由会员分享，可在线阅读，更多相关《matlab 外文翻译 外文文献 英文文献 MATALAB 混合仿真平台控制算法的概述（15页珍藏版）》请在装配图网上搜索。

1、MATALAB 混合仿真平台控制算法的概述MATALB 混合仿真平台，即为将硬件引入到仿真回路里的半实物仿真系统，可用于过程控制器的开发与测试。平台提供了三种控制器的嵌入方法，尤其能用Matlab 语言编写，大大提高了平台的灵活性。为了建立过程控制混合仿真试验系统，必须解决PC 机作为虚拟控制器设计环境的实现和在Windows 操作系统中实时控制的实现这两个问题。我们先详细阐述过程控制混合仿真试验系统的实现原理；最后介绍平台控制算法的嵌入方法，并通过实验仿真验证平台的有效性。 过程控制混合仿真平台实现原理：(1)数值计算，MATLAB 提供了大约600 多个数学和工程上常用的函数。这些函数的数

2、值运算是针对矩阵操作优化过的，可以使用它来代替底层编程语言。在保持同样性能的情况下，编程工作量非常小，数值计算采用了LAPACK，BLAS，FFTW 等优秀数学函数库，使得计算效率得到进一步的提升。MATLAB 包含的主要数学函数有线性代数和矩阵运算、傅立叶变换和统计分析、微分方程求解、稀疏矩阵运算以及三角和其他初等数学运算等；除此之外，随着Matlab 的应用领域不断的扩大，补充了用于许多特定领域的函数。(2）算法开发，强大的计算能力，方便易用的编程语言和丰富的数学函数使MATLAB 最适于用于算法开发工作。典型的应用包括：数据分析，信号处理，图像处理，系统建模和高级算法研究等。不管用户是使

3、用已有的算法，还是自行开发，MATLAB 提供了一个通用的平台。使用MATLAB 进行算法开发就像平时书写数学表达式一样。将用户在MATLAB 中开发的算法结合到外部运行的系统中。一旦用户的算法和仿真经过了编写和调试，MATLAB Compiler 和C/C+ Math Library 会将MATLAB 应用自动转换成可移植C 和C+代码的工具。对于信号处理，控制系统设计和其他一些应用，MATLAB 工具箱提供了一系列先进的技术。工具箱远远超出了提供一些基本算法的范畴：他们提供了一个学习，研究，创新前沿理论和技术的舞台。提供的算法工具箱有Neural Network Toolbox、Optim

4、ization Toolbox、System Identification Toolbox、Robust Control Toolbox、Model Predictive Control Toolbox、Control System Toolbox，Fuzzy logic Toolbox 等。(3）数据分析与可视化，通过 MATLAB，用户可以分析所有类别的数据包括信号，图像，多项式，时间历程，多变量数据和线性系统等。从分析中总结出来的结果可以作为将来进一步的算法和模型开发的基础。此外，用户可以快速地将代码片段和知识转换成可以重复使用的自动分析例程，不需要变量声明和维数定义，可很快编写出程序。

5、MATLAB 提供了方便的数据访问工具。例如，Data Acquisition Toolbox 允许用户将实时的测量数据直接传送到MATLAB 进行分析，Database Toolbox 允许用户访问符合ODBC 和JDBC的数据库，而M 文件，C 和Fortran 程序中的处理文本和二进制文件的I/O 函数，则允许用户处理任何格式的数据。MATLAB 和相关的工具箱包含了科学计算中需要的专业图形功能。从2-D 原始数据的曲线图到带标记的等值线图和交互式的GUI，这些工具提供了模型可视化的能力，帮助用户理解复杂的系统。特别是MATLAB 提供了对3-D 标量和矢量可视化的能力，包括显示等值面和

6、流图。这个能力使科学家和工程师们能够对大量、复杂和多维的数据进行可视化。Matlab 实时仿真环境，RTW 是MATLAB 提供的一个实时开发环境，是MathWork 系列软件的重要组成部分。RTW 与MATLAB 其他组成软件的无缝连接，既满足了设计者在系统概念与方案设计等方面的需求，也为系统的技术实现或完成不同功能的系统实时操作实验提供了方便，并且为并行工程的实现创造了一个良好的环境。它能直接从Simulink 的模型中产生优化的、可移植的和个性化的代码，并根据目标配置自动生成多种环境下的程序，在硬件上运行动态系统模型，同时还支持基于模型的调试。使用RTW 进行实时硬件的设计测试，用户可以

7、缩短开发周期，降低成本。当用户在Simulink环境下建模，并得到较满意的仿真结果后，就可将RTW 与一个快速原型化目标（例如RTWT 目标）联合使用。该快速原型化目标与用户的物理系统连接在一起。用户可使用Simulink 模型作为连接物理目标的接口，完成对系统的测试。RTW 的实现机制是一个复杂的过程，这里仅从RTW 自动构建应用程序的过程这一方面进行分析。RTW 生成应用程序的过程图( modle 为建立的simulink 模型的名称)。RTW 构造应用程序的过程由一个M-file 的命令来控制，对于大多数目标，缺省命令是make.rtw。其过程如下：1.模型分析，首先分析 Simulin

8、k 模型，分析的过程包括以下一些主要任务（1）数值化仿真参数和模框参数；（2）传递信号宽度和采样时间；（3）确定模块中框图的执行次序；（4）计算工作向量的大小(主要是针对S-Functions 模框)。在这个过程中，RTW 读取模型文件modle.mdl，然后把它编译为模型的内部描述。这个描述存储是与语言无关的ASCII 文件，名称为model.rtw。我们可以把这个文件看作是下个过程的输入，它将会在代码生成后被自动删除。2.调用TLC 程序来生成C 代码，在这个阶段，TLC 把存储在modle.rtw 中的内部模型描述转换为特定的目标代码。TLC 是一种解释性的程序语言，设计这个程序语言的唯

9、一的目的就是把模型描述转变成代码。在编译过程中TLC 执行包含多个目标文件(TLC 脚本文件)和TLC 函数库的程序。目标文件分为两种：一种是系统目标文件，一种是模块目标文件。这些目标文件指定如何把 modle.rtw 用作输入，从modle 中生成代码。Real-Time Workshop 绑定了用于各种目标环境下的系统目标文件，图3.5 给出了所有可用的系统目标文件。在本系统中，我们的目标环境是windows，选定的快速原型化目标为RTWT，那么在构建程序前，指定rtwin.tlc 作为我们在编译过程用到的TLC 脚本文件。3.生成定制Makefile，在这一步中，将产生定制的模板文件(m

10、akefile )，文件名称是model.mk.。生成的makefi1e 用来指导make utility 编译和链接模型，从而生成源代码。RTW是从一个系统生成模板(system template makefile)中生成modle.mk 的，这个模板文件名称是system.tmf (syetem 是指被选择目标名称)，例如在上一步中，我们选择了rtwin.tlc 作为我们的系统目标文件，那么同时也选定了一个叫做rtwin.tmf 作为了我们的系统模板文件。模板文件(makefile )允许用户定制编译器、编译器的选项和程序建立期间其他的信息，如果所用的编译器不同，那么makefile 文件

11、也将不同，例如当目标为Windows 时，系统目标文件是rtwin.tlc，但是如果用visual CC+的编译器时，系统模板文件就是win_vc.tmf，而选用watcom 时，系统的模板文件就是win_watc.tmf。molde.mk 只是system.tmf 的一个拷贝，可以修改此文件来定制构建程序的过程。RTWT(Real-Time Windows Target)是MATLAB 提供和发行的一个基于RTW体系框架的附加产品，它可将PC 机转变为一个实时系统，其目的是引入一种快速原型设计的方法，用于控制器的实时测试和开发。在这个环境里，一台PC 机既作为宿主机，又作为目标机存在。对于R

12、TWT，Simulink 和所生成的代码都运行在同一个PC 机上，其运行界面可使用户PC 机的处理器运行在Windows NT 或者Windows 95/98/2000/XP 操作系统的同时采用RTW 生成的代码。RTWT 支持许多类型的I/O 设备板卡(包括ISA 和PCI 两种类型) 。用户只需要安装相关的软件、一个编译器和I/O 设备板卡，就可将一个PC 机用作实时系统并通过I/O 设备与外部设备进行连接。内核通过I/O driver 模块作为接口，与I/O 硬件进行通讯，并且检查I/O board 安装的正确性。Simulink与实时程序之间的通讯是通过Simulink的外部模式下的模

13、块来实现的。这个模块直接与实时内核建立通讯，来开始和终止可执行程序的运行。Simulink模型和实时应用程序之间保持着一个校验机制，实时内核使用这个校验机制来判断Simulink模型结构在代码生成的自动中是否和实时应用程序的结构保持一致。这就确保了在线修改模型参数的时候，Simulink模型的参数可以正确地映射到实时应用程序相应的参数上。RTWT 对I/0 board 的支持，在混合仿真试验系统中，数据的物理采集和物理控制输出是靠I/O板卡完成的，那么实时应用程序必须在软件上具备数据采集和控制输出的能力，也就说程序必须要和I/O板卡建立连接。这一功能的实现需要RTWT的支持。在Simulink

14、模型中加入输入输出模块，设置与实际板卡一致的参数，然后编译连接就可以了，需要说明的是，板卡的驱动程序并不真的参与编译，而只是在程序运行期间，需要采集数据或者输出数据时，进行动态连接。RTWT 支持标准的I/O boards。当Simulink 模型的运行期间，RTWT 从一个或多个输入通道获得采样数据作为模型的输入，然后快速的处理数据，再通过I/O boards 的输出通道输送到外部。RTWT 提供了一个通用的Simulink block library-I/O driver block library. 在Matlab 命令中输入rtwinlib 便可以查看，I/O driver block

15、 library 提供了RTWT 所支持的I/O board 的驱动程序。通过各个block 可以很方便的设置I/O boards。这些block 和其他的Simulink block 一样，都支持拖放操作。对于RTWT 不支持的板卡，可利用S-function 模块自行开发板卡的驱动程序。I/O boards 是由其厂商提供，往往通过Switches 或者Jumpers 和厂商提供的软件来设定based dresses，voltage levels 和unipolar or bipolar modes。在Real-Time Windows Target 也提供了参数设定的功能，这个功能的应用

16、与厂商提供的软件有相同的灵活性。Simulink 外部模式下实现参数传递的机制，在外部模式下Simutink 不再对框图表示的系统模型进行仿真，而是把当前的参数值下载到目标系统。在初始下载完成后，Simulink 保持在等待状态，只有在框图中的参数发生改变或者接收到来自目标机的参数才开始动作。当框图的参数发生改变，Simulink 调用一个外部接口Mex 文件，把新的参数值和其他一些信号传递给外部Mex 文件。外部接口Mex 文件执行Inter Process Communication (IPC)通道一端的代码，这个通道把Simulink 过程(Mex 文件执行的过程)和外部可执行性程序的过

17、程连接在一起。Mex 文件通过这个通道，把新的参数传递给外部程序。通道的另一端在外部程序中执行，这一端把新的参数值写到目标参数结构中。Simulink 通过发送一个含有参数信息的讯息初始化下载操作。在C/S 结构中，Simulink 作为客户端，外部程序为服务端。两个过程可以是远程的，也可以是当地的。当客户端和服务端是远程的时候，通过TCP/IP 方式传递数据，当客户端与服务端是当地的时候，采用共享内存的方式传送数据，在本系统中采用的就是后一种方式。过程控制算法的实现方法，一个通用的仿真平台，必须能够方便的让用户加入自己的控制算法或控制系统设计方案，此过程控制混合仿真平台提供了以下三种控制算法

18、的嵌入方法：应用Simulink 提供的模块搭建控制器模型，Simulink 提供了各种各样的模块集合，在Simulink 提供的图形用户界面上，只要进行简单拖拉操作，就可利用这些模块集合构造出复杂的控制器模型。此外Simulink 还提供了诸如模糊逻辑工具箱与神经网络工具箱之类的高级算法控制工具箱，利用这些工具箱，结合其他模块，可搭建出各种与之相关的控制算法模型，如模糊PID、神经网络PID 等，如以模糊PID 控制器模型。利用工具箱的图形用户界面编辑控制器，可方便地设计智能控制器的直观图式系统。利用 Simulink 模块搭建控制器模型灵活方便，它外表以方块图形式呈现，且采用分层结构。应用

19、Simulink 模型作为控制器的可视化界面，不仅能让用户知道控制器具体环节的动态细节，而且能让用户清晰地了解控制器中各器件、各子系统的信息交换，掌握各部分间的交互影响，为进一步分析改善控制器提供了便利条件。编写C 语言的S-函数实现控制算法，在实际应用中，通常会发现有些算法用普通的Simulink 模块不容易搭建，对此可以使用Simulink支持的S-函数格式，S-函数作为Simulink 的扩展工具有固定的程序格式，可采用MATLAB、C/C+、Fortran 等语言编写。但由于在实时仿真过程中需要将Simulink 模型转化为C 代码并生成独立文件，只有C/C+语言编写的S-函数支持这一

20、功能，因此在算法扩展中必须采用C/C+语言编写的S-函数。S-函数的工作原理和Simulink的仿真原理基本类似，每一个Simulink模块都具有输入u、输出y、状态x三个向量和其他一些对应仿真各阶段的方法。Simulink通过循环调用模型中的各模块的特定方法来完成诸如计算输出值、更新离散状态值、计算连续状态微分等任务。在S-函数中提供了与Simulink仿真不同阶段相对应的回调函数。初始化阶段包括mdlInitializeSizes ， mdlInitializeSampleTimes 和mdlInitializeCondi 模块它们分别起初始化系统输入维数输出维数、状态变量个数，定义仿真采

21、样时间和初始化状态变量的作用。输出阶段mdlOutputs模块计算系统输出。循环仿真阶段mdlUpdate模块计算更新系统当前状态变量的值。C语言S函数不仅可以将具有C语言描述功能的代码嵌入Simulink模块，以便于在Simulink中仿真，而且可以很容易的将C语言S-函数模块功能代码转换成纯C代码。14MATALAB hybrid simulation platform control algorithm overviewMATLAB Hybrid simulation platform, that is, the hardware put into the simulation loop

22、 in the loop simulation system for process control of the development and testing. Platform embedded controller provides three methods, in particular, can use MATLAB language has greatly enhanced the flexibility of the platform. In order to build simulation system for process control systems must be

23、 addressed as a virtual controller PC, the environment and in the Windows operating system to achieve real-time control to achieve these two problems. We first elaborate simulation system for process control system principle, and finally introduced the embedded platform control algorithm method and

24、the experiment results verify the effectiveness of the platform. Hybrid simulation platform process control theory: (1) numerical calculation, based on more than 600 MATLAB provides some commonly used mathematical and engineering functions. The numerical computation of these functions are optimized

25、for matrix operations, you can use it to replace the underlying programming language. Maintain the same performance in the case, the programming effort is very small, value calculated using the LAPACK and BLAS, FFTW other excellent math library, so the calculation efficiency is further improved. MAT

26、LAB contains the main mathematical functions are linear algebra and matrix operations, Fourier transform and statistical analysis, solving differential equations, sparse matrix operations, and trigonometric and other elementary mathematics, etc. In addition, use of MATLAB applications with continuou

27、s expanded to add a function for a number of specific areas. (2) Algorithm development, computing power, easy to use programming language and rich mathematical functions to the most suitable for MATLAB for algorithm development. Typical applications include: data analysis, signal processing, image p

28、rocessing, system modeling and advanced algorithm research. Whether the user is using the existing algorithms or developing their own, based on MATLAB provides a common platform. MATLAB for algorithm development using the mathematical expression of the same writing as usual User-based algorithms dev

29、eloped in MATLAB integration to external systems running. Once the users algorithms and simulation through the writing and debugging, MATLAB compiler and the C / C + + Math Library MATLAB application will be automatically converted to portable and C + + code tool. For signal processing, control syst

30、em design and other applications, MATLAB toolbox providing a range of advanced technology. Toolbox far beyond the scope of provision of some basic algorithms: they provide a learning, research, innovation and cutting-edge theory and technology arena. Algorithm toolbox to provide a neural network too

31、lbox, optimization toolbox, system identification toolbox, robust control toolbox, model predictive control toolbox, the control system toolbox, fuzzy logic toolbox. (3) Data analysis and visualization, through the MATLAB software, users can analyze all types of data, including signal, image, polyno

32、mials, time course, data and multi-variable linear systems. From the analysis results can be summed up as a model for the further development of the algorithm and the base. In addition, users can quickly convert the code fragment and knowledge can be reused in automated analysis routines, no variabl

33、e declarations and the definition of dimension, can write programs quickly. Based on MATLAB provides a convenient data access tools. For example, the data acquisition toolbox allows users to send real-time measurement data directly into MATLAB for analysis, database toolkit to allow users to access

34、consistent with the ODBC and JDBC in the database, while M c and the Fortran language documents and procedures in handling text and binary file I / O function, then allow the user to handle any format. And related MATLAB toolbox contains a scientific computing need of professional graphics. The raw

35、data from the two-dimensional curve to the contour map with markers and interactive graphical user interface, these tools provide a model visualized capabilities to help users understand complex systems. In particular, provides a MATLAB-based three-dimensional scalar and vector visualization capabil

36、ities, including display equivalence face and flow diagram. This capability enables scientists and engineers to a large number of complex and multidimensional data visualization. MATLAB real-time simulation environment, RTW is to provide a real-time based on MATLAB other components of the software f

37、or seamless connection, not only to meet the designers in the system concept and program design needs, the technology for the system or perform different functions in real-time operating system provides the convenience of experiments, and for implementation of concurrent engineering to create a good

38、 environment. It directly from the SIMULINK simulation model produces optimized, portable and personalized code, and automatically generate a variety of configurations depending on the target under the program environment, the hardware running dynamic system model also supports model-based debug.Rea

39、l-time using the RTW of the design of test hardware, users can shorten the development cycle and reduce costs. When the user model in SIMULINK environment and get satisfactory simulation results, can be a RTW with the goal of rapid prototyping (such as RTWT target) joint use. The goal of rapid proto

40、typing of physical systems and users connected. Users can use the SIMULINK model as the interface to connect the physical target to complete system testing.RTW implementation mechanism is a complex process, where only from the RTW process automated building applications that were analyzed. RTW gener

41、ated application process diagram (model for the establishment of the SIMULINK name). RTW construction application process by an M-file in order to control for most of the target, the default command is make.rtw. The process is as follows: 1. model, first analyzes the SIMULINK model to analyze the pr

42、ocess includes the following main tasks (1) Numerical simulation parameters and molding parameters; (2)Passing the signal width and sampling time; (3) determine the module block diagram of the implementation of the order; (4) Calculation of the size of the vector (mainly for S-Functions module box).

43、 In this process, RTW read the model file modle.mdl, and then compile it as a model of internal description. This description is stored in ASCII files and language-independent, name model.rtw. We can use that file as input for the next process, it will be automatically deleted after the code generat

44、ion. 2. call TLC program to generate C code, at this stage, TLC stored in modle.rtw in the internal model description into a specific object code. TLC is an interpreted programming language, programming language design that the only purpose is to model description into the code. TLC in the compilati

45、on process of the implementation of multiple object files (TLC script file), and TLC library procedures. Object file is divided into two types: one is the target file system, one is the module object file. The target file to specify how to modle.rtw used as an input, generate code from modle. Real-T

46、ime Workshop target binding environment for a variety of target file system, Figure 3.5 shows all the available system target file. In this system, our target environment is windows, the selected target rapid prototyping RTWT, then the building process before, we specify rtwin.tlc as TLC used in the

47、 compilation process script file. 3.Generate customized M-file, in this step, will have a custom template file (M-file), the file name is model.mk. Generated makefi1e used to guide the make utility to compile and link model, which generates source code. RTW is a system-generated template (system tem

48、plate M-file) generated modle.mk, this template file name is system.tmf (syetem is the target name is selected), such as in the previous step, we selected rtwin.tlc as Our system target file, also called rtwin.tmf selected our system as a template file. Template file (M-file) allows users to customi

49、ze the compiler, compiler options and other information during the process to establish, if the compiler used in different, then the makefile file will be different, such as when the target is Windows, the system target file is rtwin . tlc, but with the visual C C + + compiler, the system template f

50、ile is win_vc.tmf, and choose watcom, the system of the template file is win_watc.tmf. molde.mk only system.tmf a copy, you can modify this file to customize the build process of the process. RTWT (Real-Time Windows Target) is to provide and distribute a MATLAB RTW system framework based on the addi

51、tional products are shipping 将 turn its machines into as a PC system with the aim of rapid prototyping one kind of methods for Kong real-time testing and development. In this environment, a PC, both as a host, but also as the target machine there. For RTWT, SIMULINK and the generated code running on

52、 the same PC, the interface allows users to run PC-processor running on Windows NT or Windows 95/98/2000/XP operating system, while the code generated by RTW . RTWT supports many types of I / O device boards (including the two types of ISA and PCI). Users only need to install related software, a com

53、piler and I / O device card, a PC machine can be used for real-time system and through I / O devices and external devices to connect. Core through the I / O driver module as the interface and I / O hardware, communication, and to check I / O board installation is correct. SIMULINK and Real-time comm

54、unication between processes is through the SIMULINK external mode, the module to achieve. This module is to establish communication directly with the real-time kernel to start and terminate the operation of an executable program. SIMULINK models and real-time applications have maintained a checking

55、mechanism to check real-time kernel to use this mechanism to determine the SIMULINK model structure in the automatic code generation and real-time whether the structure of the application consistent. This ensures that the line changes when the model parameters, SIMULINK model parameters can be corre

56、ctly mapped to the real-time application, the corresponding parameters. RTWT on the I / 0 boards support, in a mixed simulation systems, data collection and physical control of the physical output by I / O board to complete, then the real-time applications must have the software output data acquisit

57、ion and control capability , also said that procedures have to be, and I / O board to establish a connection. The realization of this function requires RTWT support.SIMULINK model, adding input and output modules, setting and actual card Yi Di , and then build the connection is available, and need t

58、o be explained that the boards procedure does not it is involved in compilation, but only during the program, need to data or output data collected when the dynamic connection. RTWT support standard I / O boards. When the SIMULINK model to run time, RTWT from one or more input channels were sampled

59、data as model input, and then process the data quickly, and then through the I / O boards to the external output channel delivery. RTWT provides a generic SIMULINK block library-I / O driver block library. In the MATLAB command input can see, I / O driver block library provides RTWT supported I / O

60、board driver. Each block can be easily by setting I / O boards. The block and, like other SIMULINK block, support drag and drop. Do not support the board for RTWT can use S-function modules developed by board driver. I / O boards is provided by companies, often through the Switches or Jumpers and so

61、ftware vendors to set based dresses, voltage levels, and bipolar modes. In Real-Time Windows Target also offers the functions of the parameters set, the function of application and software vendors have the same flexibility. SIMULINK external mode, parameter passing mechanisms to achieve in the exte

62、rnal mode SIMULINK no longer a system block diagram model simulation, but the current value of the parameter download to the target system. After the initial download is complete, SIMULINK to keep waiting for the state, only in the block diagram of the parameters change or receive parameters from th

63、e target machine began to move. When the parameters change diagram, SIMULINK calls an external interface M-file, the new parameter values and some other signal to an external M-file. External interface M-file execution Inter Process Communication (IPC) channel at one end of the code, the access to S

64、IMULINK process (M-file execution process) and external procedures executable process together. M-file through this channel, the new parameter to the external program. Channel the other end of the external program execution, the client writes the new value of the parameter object parameter structure

65、. SIMULINK information by sending a message containing the initialization parameter download operation. In the C / S structure, SIMULINK as a client, server-side external procedures. Two processes can be remote, it can be local. When the client and server is remote, when, through the TCP / IP mode t

66、ransmission of data, when the client and the server is local time, use shared memory, transmission of data, the system is used in the latter way. Implementation of process control algorithms, a common simulation platform, to Jean-users to easily add their own control algorithm or control system design, process control hybrid s