分析数据
R2026a确定延迟、反馈和激励水平等数据特征
在开始根据选定数据进行模型估计之前，必须检查数据中是否存在任何不良特征，以确定是否需要对数据进行预处理。您可以通过以下方式实现：
对数据进行绘图，以考察时域和频域的行为。
使用
advice命令分析数据中是否存在恒定偏移和趋势、延迟、可能的反馈以及信号激励水平。
函数
bode | 动态系统的波特频率响应 |
bodemag | Magnitude-only Bode plot of frequency response |
idplot | Plot input and output channels of estimation data (自 R2023a 起) |
advice | Analysis and recommendations for data or estimated linear models |
delayest | 根据数据估计时间延迟（死区时间） |
isreal | 判断模型参数或数据值是否真实 |
realdata | 确定 iddata 是否基于实值信号 |
checkFeedback | Identify possible feedback data (自 R2023a 起) |
pexcit | 输入信号的激励水平 |
impulseest | Nonparametric impulse response estimation |
etfe | Estimate empirical transfer functions and periodograms |
spa | Estimate frequency response with fixed frequency resolution using spectral analysis |
spafdr | 利用频谱分析估计频率响应与频谱，该分析具有频率依赖性分辨率 |
dataPlotOptions | Option set for idplot when plotting input/output estimation data
contained in a timetable, numeric matrices, or an iddata
object (自 R2023a 起) |
主题
- Is Your Data Ready for Modeling?
Before you start estimating models from data, you should check your data for the presence of any undesirable characteristics.
- How to Plot Data in the App
After importing data into the System Identification app, as described in 表示数据, you can plot the data.
- How to Plot Data at the Command Line
The following table summarizes the commands available for plotting time-domain, frequency-domain, and frequency-response data.
- Identify Delay Using Transient-Response Plots
You can use transient-response plots to estimate the input delay, or dead time, of linear systems.
- How to Analyze Data Using the advice Command
You can use the
advicecommand to analyze time- or frequency- domain data before estimating a model.
