sizing microgrid using for and if statement

Hello,

I am currently doing a project where my task is to size the components in micro-grid and calculate the LCC. I extracted the data för one day. Data I have are: Battery voltage, Battery current, Inverter power( reprecent the consumption), Solar power (represent the production).I am a beginner in matlab and hope that some one can help me here. I want to write a code that loop through all the hours in the day and calculate the following:

for all hours

if (production > consumption) and (battery power <= 80%*Battery max capacity)

charge the battery= Production-consumption

elseif (production > consumption) and (battery power >= 80%*Battery max capacity)

wasted power= Production-consumption

elseif (consumption > production) and (battery power >= 20%*Battery max capacity)

use the battery=consumption-production

else (consumption > production) and (battery power <= 20%*Battery max capacity)

increase battery capacity = consumption-production

end

I am already stuck by looping through all the hours and dont know how to do it

so thankful for help

[Guillaume]

In all likelyhood you don't need a loop. Your test can probably be carried out at once on all the data.

For us to be able to give you an answer that works for you, it would be very useful to see an example of the data (is it a table, cell array, one or more matrix? How is the time stored? As a datetime? datenum? datestr?, etc.). So please attach a demo mat file to your question.

Answer 1

[Guillaume]

It sounds like you have several problems. The first one being the data generation which you currently do with some form of digitisation. I don't think it's something I can help with. Then you have the problem of processing your data once you've generated it.

I don't really understand what you're doing with your datenum, datevec, datestr. One thing for sure, it probably doesn't do what you want since at one point you're using hours/minutes/seconds in place of year/month/day. In any case, you shouldn't be using any of these types. Use datetime which is so much more practical. E.g. if you want to generate dates in a day in interval of 1 minutes:

d = datetime('2019-03-30 00:00:00') + minutes(0:24*60-1);

As for your data, I recommend you store it in a table or even better timetable, eg.

solardata = timetable(sometimevector, B_v(:), P_s(:), P_inv(:), 'VariableNames', {'BatteryVoltage', 'SolarPower', 'InverterPower'});

You could then retime your timetable to make it hourly if that's what you need.

After that you can store the state of your system (I think that's what you want to do, it's not clear) in a new column, eg.:

solardata.State(solardata.SolarPower > solardata.InverterPower & solardata.BatteryVoltage <= 0.8*SOCmax) = "charging";
solardata.State(solardata.SolarPower > solardata.InverterPower & solardata.BatteryVoltage > 0.8*SOCmax) = "wasting power";
solardata.State(solardata.SolarPower <= solardata.InverterPower & solardata.BatteryVoltage >= 0.2*SOCmax) = "discharging";
solardata.State(solardata.SolarPower <= solardata.InverterPower & solardata.BatteryVoltage < 0.2*SOCmax) = "low battery";

[Raoann]

Thank you so much for your detailed answer!

I went through my data again and try to follow your tips. I wonder, if I want to generate dates in not only one day, but let say 5 days in interval of 1 minute, how can I do so?

dates from 2018/07/01-2018/07/05

one more thing, I noticed when I tried to store the data in tabeltime, my data must have the same row numbers , which I do not have. is there anyway to fix this problem?

so thankful for your help

Regards

[Guillaume]

That last error you get is because the Time is not considered a variable in a timetable and therefore you should only specify four variable names.

Summer_week_data=timetable(interpt(:),interpBat_V(:),interpBat_C(:),interpSol_p(:),interpInv_P(:),'VariableNames',{Battery_volt','Battery_current','Solar_power','Inverter_power'});

It seems you have figured out how to resolve your different length issue.

[Guillaume]

Please comment on the answer rather than starting new answers.

I thought the way you interpolated your data was fine, so I'm not sure why you're no longer doing that. Another option is to start with one timetable per variable and then synchronize them all. Since all your variables have the same duration, create one timetable per variable with a linspace'd time vector:

starttime = datetime(2018,07,01,14,0,0);
endtime  = datetime(2018,07,05,14,0,0);


tVbat = timetable(linspace(starttime, endtime, numel(Bat_V)', Bat_V, 'VariableNames', {'Battery_volt'}))
tCbat = timetable(linspace(starttime, endtime, numel(Bat_C)', Bat_C, 'VariableNames', {'Battery_current'}))
tbat = synchronize(tVbat, tCbat, 'Uniform', 'Interval', minutes(1));
%... etc for other variables.

Answer 2

[Abm]

Thank you so much for responding !! I kept thinking in this problem from yesterday.

I will continue as you suggest for the rest of the solution.

Best regards