Multiple Input Multiple Output Fuzzy Logic Controller

[natsmi]

I am designing an autotuning PID temperature control system whose PID gains are adjusted using a Multiple Input Multiple Output (MIMO) Fuzzy Controller. I am unsure how to set up specific inputs and how to use specific outputs when there are multiple inputs and outputs. Specifically, my Fuzzy system has two inputs and three outputs.

For the input, I have two 1D arrays of scalars (doubles) which I have used to build an array and concatanated the inputs to the array. How do I configure which values of this array are used as each specific input value to the fuzzy system?

Similarly, for the output, the fuzzy system gives three output values which are used as the PID gains. How do I configure these such that they are in the right order to be used as the PID Gains of the PID VI?

[AGJ_CR]

Hi natsmi,

Is there a special reason for you to have an array for every input? For all the examples that I have seen the operation is performed point by point, and with this, you would only concatenate single values to have a 1D array as the input.

Regards,

AGJ

[natsmi]

Hi agjcr, I believe that is what is happening in  practice, the 1D arrays used to make the input are just 1D arrays of single values. I have not seen any working examples of MIMO Fuzzy Systems, is there somewhere I can find these?

Thank you!

[AGJ_CR]

Hi natsmi,

You can find some examples by navigating to:

C:\Program Files (x86)\National Instruments\LabVIEW XXXX\examples\control\fuzzy

Some of them are at least MISO systems and might be helpful to clarify some ideas. You can also find them using the Example Finder from LabVIEW (By going to help on the top menu, and then Find Examples). There you can go to the search tab on the left and type in fuzzy.

Hope this helps.

Regards,

AGJ