Simulink Sensitivity Analyzer - What do the output results mean?

I'm using the Simulink analyzer with two requirements:

1. Maximize final value of Signal A
2. Minimize final value of Signal B

I'm trying to see how varying a parameter changes the final value of Signals A and B. I chose Maximize and Minimize because for my implementation, a bigger final A is better and a smaller final B is better. At the end of the sensitivity analysis, I get resulting values of 0 to 200 for the minimize Signal B requirement, and 0 to -0.15 for maximize signal A requirement. What do these numbers mean?

Also, is it possible to circumvent the trouble of figuring out what the resulting numbers mean, and just get the sensitivity analyzer to output the final value of the signal?

Answer 1

[Ayush]

Hi Kadhir,

I understand that you want to know how varying the parameters changes the values of Signals A and B and the interpretation of their sensitivity ranges after the analysis.

Here is a possible interpretation of the value ranges provided by you:

1. For Signal A, the values 0 to -0.15 means that the varying parameters leads to decrease in the signal. Since your requirement is to maximise the final value, the closer the value is to 0 or less negative the better as per the varying parameter.
2. For Signal B, the values 0 to 200 means that the sensitivity of this signal ranges between these two numbers with a varying parameter. Since your requirement is to minimize its final value, the close it is to 0 the better as per the varying parameter

For your second question, you can try leveraging the “Response Optimizer” to view and output the final values of the signals directly without you having to interpret the ranges. Please refer to the below documentation to know more about the “Response Optimizer”:

https://www.mathworks.com/help/sldo/ref/responseoptimizer-app.html

To use this app efficiently you would need to do the following:

1. Specify design requirements: In your case, it would be to setup the objective to maximize Signal A and minimize Signal B.
2. Incorporate parameter uncertainty: With the sensitivity analysis you can set different ranges for different parameters.
3. Optimise varying parameters: Decide on the parameters that influence or affect the most.

Hope it helps,

Regards,

Ayush Misra