About Developing DC-DC Converter Control in Simulink- from Vivek Raju-

Hi Everyone ,

I have basic question about Digital DC-DC Buck Boost converter from here :

Related video;

https://www.mathworks.com/videos/developing-dc-dc-converter-control-in-simulink-1533829367281.html?elqsid=1602349251309&potential_use=Commercial

and related files;

https://www.mathworks.com/matlabcentral/fileexchange/68536-digital-control-of-dc-dc-buck-boost-converter

I will briefly summarize my question.

At ‘Sepic_new_closedloop_tune’ matlab file is a model in the loop simulation at It has dicrete PI controller and its usage and values are shown below figures,

P(proportional)=1e-3

I(integral)=40

No problem so far.

As regards ‘DC_DC_LED_Implemenatation’ file , as you know , this is a hardware in the loop interface in order to build&deploy to C2000 microprocessor(target hardware) . In this code , to get rid of calcution complexity(multiplication) C2000 math operations are used as following figure ,

And integrator block as following

According to knob(varies) parameter , PI values are below;

Proportional=min:0 max :0.1

Integral= min:0 max :0.1

I know in hardware in the loop simulation fixed data point values have used in order to not waste time and microprocessor works efficiently. But,

I didn’t understand when model in the loop simulation P=1e-3 and I=40 why these values are different in the hardware simulation (P=0-0.1)(I=0-0.1) ?

How do they 40,1e-3 values transform to small range(0-0.1) .

And what is the convertion formula ?

Is anyone has idea?

Thanks in advance for your answers.

Burak

Answer 1

[Vasco Lenzi]

Hi Burak

there may have been an error in parametrizing the MOSFET in simulation, which led to different results between simulation and code generation with the real hardware, and so different PI parameters. We are working on uploading a working version with the right model so that you have the same values for simulation and code generation. Should come soon, stay tuned! I'll add a link when is live.