SignalTracking
|
Page in progress. |
Creates an object with requirements for tracking the signal.
| Library |
|
Description
Function SignalTracking creates an object EngeeDesignOptimization.SignalTracking with requirements for tracking the signal relative to the reference signal in the time domain. You can set a requirement in the form of equality, as well as a restriction from above or below.
Syntax
Function call
-
track_req = SignalTracking(Name=Value)— creates an objecttrack_reqand sets its properties using one or more arguments of the type «name-value». Unspecified properties retain their default values.
Arguments
Input arguments «name-value»
Specify optional argument pairs as Name=Value, where Name — the name of the argument, and Value — the appropriate value. The order of type arguments «name-value» It doesn’t matter.
# AbsTol — absolute tolerance
+
0 (by default) | scalar
Details
The absolute tolerance used to define the boundaries as the signal approaches the reference signal, set as a non-negative scalar.
The boundaries for the reference signal are defined as follows:
where — the value of the reference signal at a certain point in time, and and — the upper and lower limits of the tolerance corresponding to this point in time.
| Типы данных |
|
# Description — description of the requirement
+
nothing (by default) | line
Details
Description of the requirement, set as a string.
| Типы данных |
|
# InterpolationTimes — time points used when comparing the points of the monitored and reference signals
+
"Reference only" (by default) | "Testpoint only" | "Reference and Testpoint"
Details
The time points used when comparing the points of the monitored and reference signals.
They are set to one of the following values:
-
"Reference only"— comparison of signals only at time points corresponding to the reference signal; -
"Testpoint only"— comparing signals only at time points corresponding to the monitored signal; -
"Reference and Testpoint"— comparison of signals at time points corresponding to both the reference signal and the monitored one.
Linear interpolation is used to compare signals at the same time points.
| Типы данных |
|
# Method is an algorithm for evaluating the fulfillment of a requirement
+
"SSE" (by default) | "SAE" | "Residuals"
Details
The algorithm for calculating the fulfillment of a requirement, set by one of the following values:
-
"SSE" -
"SAE" -
"Residuals"
When evaluating a requirement, the error between the reference and monitored signals is calculated. This property defines how functions evalRequirement the error signal should be handled .
| Типы данных |
|
# Name — the name of the requirement
+
nothing (by default) | line
Details
The name of the requirement, set as a string.
| Типы данных |
|
# Normalize — option to enable normalization
+
true (by default) | false
Details
The option to enable normalization, set as true or false. The maximum absolute value of the reference signal is used for normalization.
| Типы данных |
|
# ReferenceSignal — reference signal
+
nothing (by default) | the matrix | tabular data
Details
A reference signal defined as a matrix or table, the first column of which contains the time, and the second contains the real values of the signal.
| Типы данных |
|
# RelTol — relative tolerance
+
0 (by default) | scalar
Details
The relative tolerance used to define the boundaries as the signal approaches the reference signal, set as a non-negative scalar.
The boundaries for the reference signal are defined as follows:
where — the value of the reference signal at a certain point in time, and and — the upper and lower limits of the tolerance corresponding to this point in time.
| Типы данных |
|
# RobustCost — option to enable sustainable emissions treatment
+
true (by default) | false
Details
The option to enable sustainable emissions treatment when calculating the requirement, set as true to enable sustainable emission management or false otherwise.
If sustainable outlier treatment is enabled, then when calling the function evalRequirement The Huber loss function is used to estimate the cost (penalty amount) associated with tracking error emissions. The tracking error is calculated as . Statistical error characteristics are used to identify outliers.
The specific type of cost function used depends on the method of assessing the requirement.
| The evaluation algorithm | Cost function for non-outlier counts | Cost function for emissions |
|---|---|---|
|
where — this is a set of counts that are not outliers. |
where — linear weight, — lots of outliers. |
|
where — this is a set of counts that are not outliers. |
where — constant value, — lots of outliers. |
|
Emissions are not removed. where |
|
| Типы данных |
|
# Type — type of requirement
+
"==" (default) | "⇐" | ">="
Details
The type of signal tracking requirement specified by one of the following values:
-
"=="—It’s a coincidence; -
"⇐"— upper limit; -
">="— the lower limit.
| Типы данных |
|
# Weights — the weights used to calculate the error
+
nothing (by default) | vector
Details
The weighting coefficients used in calculating the error between the reference signal and the monitored signal, specified as a vector, the number of elements of which coincides with the number of signal values in the property ReferenceSignal.
| Типы данных |
|
Output arguments
# track_req — an object with requirements
+
object
Details
An object with tracking requirements. The object contains the following properties: ReferenceSignal, Weights, Type, Method, InterpolationTimes, Normalize, RobustCost, RelTol, AbsTol, Name, Description.
| Типы данных |
|