I have to do a project on MPPT (Perturb & observe) with a buck converter. I've read a lot of papers on it but I'm unable to understand a few things about it. I understand that in P&O algorithms, the voltage of the PV array is varied until MPPT is reached. But how is the voltage of the PV changed in the first place? Isn't it dependent on the sun entirely? I'm not sure on this but I think that is to be done with the help of a dc to dc converter. If that is so, then it is on the output of the dc to dc converter that MPPT should be applied. But in all the papers I've seen so far, the MPPT mechanism takes input directly from the PV module and then adjusts converter voltage. I don't understand the former step of the process. I mean, it is the converter's voltage that we should be playing around with to find the max power point, not the PV's direct output as we can't even change that. So shouldn't the MPPT mechanism read and compare the converter's current and previous power levels rather than using the PV's output readings directly?

I doubt I'm explaining myself very well, but I hope you understand me.

Thank you!

 

The job of the MPPT converter is to present a load to the Panel such that the instantaneous product of panel voltage and panel current is maximum.

At any given stable illumination level, the panel can be modeled as a current source shunted by a resistance. A small change in current drawn from the panel causes a large change in panel voltage. Tuning the MPPT requires that the converter change voltage to see what happens to current or change current to see what happens to voltage.

 

What is the MPPT driving? Typically say for a battery charger the current drawn from the panel is varried to find the max point of power, with the panel voltage being seen as a dependent variable.

 

Oh. So current drawn from the panel and in turn, the voltage, can be varied. I thought it was entirely dependent upon solar radiation..

Ok, so how is it done again? By varying the load impedance?

In my project, I have to use MPPT to perform PWM on the buck converter's switch to produce a stable, regulated dc output voltage. I wonder if the change in load impedance is brought about by changes in duty cycle of the switching pulses?

Could anyone explain to me the steps involved in the process i.e which parameters of which component need I be concerned about for MPPT? I don't know how correct or how wrong I am so far.

Thank you so much!

 

I'll try a short stab at it: Remember that for MPPT to operate, the load needs to be able to absorb full output power of SP. Assume the load is a partially discharged battery. With a variable load across SP, measure V & I to find max power. With V controlled Buck Converter connected from SP to battery, adjust BC output to draw down SP V to measured max power V of SP. Battery is now charging at MPPT of SP.

 

Assuming that the solar flux is constant, and assuming the load will accept as much power as the panel will generate, there is still the matter of "finding" the optimum operating point. The MPPT controller has to do this by trial-and-error. Say the independent variable is current into the buck converter. Then the dependent variable is the panel voltage, and only one combination of panel voltage and panel current results in peak power. If anything changes (sun get lower in the sky, load side changes) the peak power point has to be found again...

 

Assuming that the solar flux is constant, and assuming the load will accept as much power as the panel will generate, there is still the matter of "finding" the optimum operating point. The MPPT controller has to do this by trial-and-error. Say the independent variable is current into the buck converter. Then the dependent variable is the panel voltage, and only one combination of panel voltage and panel current results in peak power. If anything changes (sun get lower in the sky, load side changes) the peak power point has to be found again...

I understand that the MPPT will vary the SP current/voltage and compute power until a maximum is obtained. What I don't understand is, how will MPPT vary the panel's current or voltage? I don't know any way of changing the current or voltage of a PV module without changing environmental conditions like temperature, irradiation etc.

 

Connect an variable resistor (shown below) to a PV panel. At noon (constant illumination), vary the resistor, at each new resistance value, measure panel voltage and panel current. Plot current at each point vs voltage at that point. At each pair of points, multiply current by voltage, and plot power vs voltage (panel power) as a new plot. At what value of voltage does the power reach a maximum? With respect to the panel itself, it can be operated at an infinite number of possible voltages (the panel current changes based on the slope of the panels IV curve).

That is what the MPPT controller does while it is doing its thing...

 

Connect an variable resistor (shown below) to a PV panel. At noon (constant illumination), vary the resistor, at each new resistance value, measure panel voltage and panel current. Plot current at each point vs voltage at that point. At each pair of points, multiply current by voltage, and plot power vs voltage (panel power) as a new plot. At what value of voltage does the power reach a maximum? With respect to the panel itself, it can be operated at an infinite number of possible voltages (the panel current changes based on the slope of the panels IV curve).

That is what the MPPT controller does while it is doing its thing...

Thanks a lot. Feel kind of stupid now.

Now that that's settled, I have a follow-up question. If I want to use an algorithm like Perturb & Observe, then the process of varying panel voltage and current becomes part of the algorithm and would have to be automatically performed by the MPPT controller. How would that be achieved? Because with a variable resistor, the resistance adjustment would have to be manual.

Thank you so much for your constant support! I really appreciate it, guys.

 

Now that that's settled, I have a follow-up question. If I want to use an algorithm like Perturb & Observe, then the process of varying panel voltage and current becomes part of the algorithm and would have to be automatically performed by the MPPT controller. How would that be achieved? Because with a variable resistor, the resistance adjustment would have to be manual.

I wish you would stop saying "the process of varying panel voltage and current" because you do not vary the panel voltage! True the panel voltage will change in response to current as you draw different values of current but you do not directly vary this voltage

Again, what are you driving? What is using this power? If you are intending to provide a fixed voltage output then the whole concept of MPPT does not apply as you are not opperating at a maximum power point. All you have is a fixed voltage which may take up to a maximum current for a given condition of the panel and it's illumination (and temperaturte and cleanliness and...).

 

I wish you would stop saying "the process of varying panel voltage and current" because you do not vary the panel voltage! ....

At any given illumination level, temperature, etc, the panel has a characterist I vs V curve. If you vary voltage, the current is determined. If you vary current, the voltage is determined. They are not independent variables.

 

At any given illumination level, temperature, etc, the sole way to vary the terminal voltage is to vary the current being drawn.

 

If you are intending to provide a fixed voltage output then the whole concept of MPPT does not apply as you are not opperating at a maximum power point. All you have is a fixed voltage which may take up to a maximum current for a given condition of the panel and it's illumination (and temperaturte and cleanliness and...).

Thank you! That was exactly my confusion. And yet I am required to take a 48 volt input, convert it to 28 volt (and hold it there) while applying MPPT. It makes no sense. I can either have a fixed voltage, or the max. power.

But what if I vary the input to the dc-dc converter i.e the PV module output in such a way that the converter's output always keeps adjusting to provide 28 V, while tracking the MPP? Could the max power be transferred in this way? Or would the action of the converter change the power level by the time it reaches the output?

Once again, thank you so much all of you!

 

For the moment let's assume the converter works at 100% efficiency and is designed to produce a steady 28VDC at whatever current the load is demanding. Also assume the panel has a max power point of Pmax.

As long as Pout<Pmax everything works great, just some potential power is being lost.

When Pout=Pmax then yeah, you are operating at the max power point and not an erg is wasted. But the converter is not controlling the power, just outputting the same voltage it always did.

When Pout>Pmax the panel can no longer support the load on it, the voltage falls back, and the converter starts trying to draw more power then it can, running at as close to a 100% duty cycle as it can, and the voltage will probably drop until the output is as high as can be.

MPPT was developed as a way to draw the largest power to supply a load where the converter controls the output current in an application such as a battery charger: the output voltage stays near constant but the delivered current is varied to keep it at the maximum possible level. A constant voltage output has no such control over the load.

A variable voltage output into a fixed load could be operated at a max power point, but the voltage will not be constant.

 

So now we have a 48 V, OC, output from SP connected to a fixed 28 V output, DC-DC converter. Supose that we connect to the output a voltage controlled electronic load . Now the MPPT ckt. will cause the load to decrease resistance, increasing load on SP causes an increase in current & lowering SP V out untill V = calculated MPPT V. System back in equalibrium with SP output V around 36V but delivering 28 V output to E Load @ max power.
E Load maybe FETs on a heatsink.

 

the output voltage stays near constant but the delivered current is varied to keep it at the maximum possible level.

So it is possible to have a constant voltage output along with MPPT, right?

 

Not really. A battery charger is at base a current source. The (nearly) constant voltage is determined by the battery. The controller is varying the current supplied to the battery to charge it. Thus the controller is determining the power it is delivering.

A voltage source is only controlling the voltage, not the power.

 

It seems to me now that the dc-dc converter is pointless. MPPT can just be achieved by tuning the PV module to max power point and then driving a load.

 

You have that exactly backwards. Attempting MPPT is pointless. The switching DC-DC converter will allow the maximum possible power to be transfered.

How else would you accomplish the voltage conversion? A linear supply will be converting up to and over 40% of the available useful power to heat.

 

You have that exactly backwards. Attempting MPPT is pointless. The switching DC-DC converter will allow the maximum possible power to be transfered.

That would be true for a particular PV voltage and current. But if current, and in turn, the voltage, is moved over a range of values until a point is found where the power level is greater than that achieved at any other current and voltage value, then max power would be reached and this has to be done with an MPPT controller making dc-dc converter redundant.