Dear Ishaan Karnik ,

The ever growing demand and socio-economic development, the conventional energy sources have almost become obsolete for nurturing this demand. Thus opening new avenues to search for the non-conventional sources. Photovoltaic is inexhaustible, renewable and non-polluting, so it is worth saying that it is one of the prospective solution for this crisis. But, having these benefits does not solve the problem as the efficiency is dependent on various factors such as irradiance, cell temperature, dust, sunlight's spectral density.

Conference Paper Development of Fuzzy Logic based MPPT controller for PV syst...

Regards,

Shafagat

Hi Ishaan Karnik

You can designed a Mamdani Fuzzy System to track MPP, and I have plotted the Input–Output surface based on your FIS. Can you interpret the surface in tracking MPP? It will help understanding the behavior.

Did you actually design the fuzzy system specially for tracking the MPP? Or, you took the second-hand FIS from the existing boost convertor?

Hi Dr. Yew-Chung Chak

Sorry for the late replay.

Can you interpret the surface in tracking MPP?

Sir, I couldn't understand the question can you elaborate on the same.

Did you actually design the fuzzy system specially for tracking the MPP? Or, you took the second-hand FIS from the existing boost convertor?

I have designed the FIS specially for tracking the MPP and I have not taken second-hand FIS from the existing boost convertor. The FIS is yet to be optimized (especially the Fuzzification and Defuzzification MF).

I have attached required files for you review

Hi Ishaan Karnik

From the surface, I don't see any real difference from your previous design. I'm no MPPT expert, but do you really need those "flat regions" as circled in red? This is a very very common thing designers make in Mamdani-type systems. If you don't need those 'flatters', get rid of them.

Do you intentionally limit the change in duty cycle output to [–0.752, +0.752]? Is this amplitude sufficient?

By the way, we generally don't say 'optimization' at this stage because your FIS can't even track MPP in the first place (according to you). At this stage, your primary task is to design a FIS tracks the MPP as intended.

Do you really need neg/pos small and neg/pos medium MFs in the inputs? What roles do they serve? Are their roles really intelligently effective to begin with? As a Human MPPT Operator, you must be able to intelligently explain your Choice of Rules to improve the chances of designing a good FIS.

Hi Ishaan Karnik

Your documentation says that, if δP/δV = Error > 0, then the Duty Cycle should decrease. However, from your Surface, when Error = 1, and Change-in-Error = 0, the Duty Cycle is Positive (increasing). Can you explain this?

Hi Dr. Yew-Chung Chak

Thanks for your suggestion. I will look into it.

I will respond in a 1-2 days with my observation.

Yew-Chung Chak

As you suggested their might be problem with my rules.

I am trying to change the rules and verify if the algorithm tracks MPPT.

Till now I haven't been successful but I have few more ideas that I would like to try.

I will update you once I get some concrete result.

Hi Ishaan Karnik

Thanks for your reply. Since the mathematical model of your MPPT system is available, try designing the FIS through 'reasoning with numbers' as much as possible, instead of by feeling.

This is because the mathematical dynamics in your MPPT system DOES NOT CARE about human feeling. For example, just because the designer 'feels' that "there should be seven (7) MFs in each input" and "this Neg MF is placed here & that Pos MF is placed there" does not mean/guarantee that the MPPT will perform well.

In other words, every design element in your FIS shall be justifiable by numerical reckoning that satisfies the governing laws/theorems of the dynamics and stability. For example, one of your Rules says "if Error > 0, then the Duty Cycle should decrease." Now, how much should the Duty Cycle be decreased? (This value shall be calculated to satisfy some governing laws). Also, Error > 0 implies only one (1) MF to be placed at the positive side. What are the reasons to create more than 1 Pos MF?

Hi Ishaan Karnik

Found a good tutorial by Prof Essam Hamdi. Hope it helps. That's the starting point.

https://www.youtube.com/watch?v=Z919uopEmpo

Dr Yew-Chung Chak

Thanks. I will update you once I get positive result.