蒙特卡洛模拟用于系统的健壮性分析

最新推荐文章于 2021-03-21 06:15:09 发布

原创

最新推荐文章于 2021-03-21 06:15:09 发布 · 377 阅读

4 ·

CC 4.0 BY-SA版权

蒙特卡洛模拟是一种通过在统计约束下随机变化参数来探索复杂系统敏感性的方法。本文介绍了如何使用MATLAB和Simulink进行蒙特卡洛分析，包括在Simulink中创建动态模拟、并行加速分析以及通过稳健绘图和高级统计方法分析数据。同时，通过Simulink中的不确定系统分析示例展示了如何进行不确定性建模和分析。

Perform sensitivity analysis through random parameter variation

Monte Carlo simulation is a method for exploring the sensitivity of a complex system by varying parameters within statistical constraints. These systems can include financial, physical, and mathematical models that are simulated in a loop, with statistical uncertainty between simulations. The results from the simulation are analyzed to determine the characteristics of the system.

You can perform Monte Carlo analysis with MATLAB and Simulink. MATLAB and Statistics Toolbox let you vary uncertain parameters for your model. In Simulink, you can create dynamic simulations and alter parameters with statistical uncertainty. With both MATLAB and Simulink you can:

Create a Monte Carlo simulation to model a complex dynamic system
Distribute simulations between processor cores and individual PCs to speed analysis
Analyze data through robust plotting and advanced statistical methods

Robustness Analysis in Simulink

This example shows how to use Simulink blocks and helper functions provided by Robust Control Toolbox? to specify and analyze uncertain systems in Simulink and how to use these tools to perform Monte Carlo simulations of uncertain systems.

Introduction

The Simulink model usim_model consists of an uncertain plant in feedback with a sensor:

open_system('usim_model')

The plant is a first-order model with two sources of uncertainty:

Real pole whose location varies between -10 and -4
Unmodeled dynamics which amount to 25% relative uncertainty at low frequency rising to 100% uncertainty at 130 rad/s.

The feedback path has a cheap sensor which is modeled by a first-order filter at 20 rad/s and an uncertain gain ranging between 0.1 and 2. To specify the