9th Aug, 2022

Universiti Malaysia Sabah (UMS)

Question

Asked 12th Sep, 2021

I'm working on predicting solar power output using machine learning, but I can't find a public dabases of solar power output with 1 minute step.

I only find databases with 1 hour step, and an other databases of irradiance with 1 minute step. So I decided to use this databases to calculate the power solar output for each 1 minute, and then use the results to train my machine learning model...

I've seen that there's a linear model that we can use to calculate the power solar output using irradiance.

Does anyone know how to find it ? or another solution ?

Solar power output depends on your cell or module type, materials and its arrangement of array.

Normally solar irradiance determine efficiency of a panel;

Say, poly Si standard module of of power rating 125 W/m2 at standard condition 1.5 air mass but irradiation (av) 1200 w/m2 in any location

So efficiency of that location would be promoted as (125/1000)x (1200/1000) x 100% = 15%

But normally efficiency is (125/1000) x 100% = 12.5%

Total panel area required = 2500 /(125 x1.2) = 16.6 m2 ; dimension of each panel = 1.1mx 1.6 m =1.75 m2

So, total 16.6/1.75 = 10 panels are required

But normal standard 1000 w/m2 irradiance required area = 20 m2

Panel required 20/1.75 m=12 panels

1 Recommendation

Dear Saad Benslimane ,

welcome!

In order to get the output power given the input irradiance you need to multiply the input by the conversion efficiency. You can find data for the conversion efficiency

for different commertial solar cells as a function of T and the irradiance G.

It is assumed that Eta is more less independent on G and its dependence on temperature is known.

This could be a possible solution.

Best wishes

8 Recommendations

**Get help with your research**

Join ResearchGate to ask questions, get input, and advance your work.

The output power of a PV array depends on the available solar radiation (G) and the ambient temperature (T). The output power of a PV array increases linearly as the solar radiation increases, and decreases as the ambient temperature increases. Thus, the instantaneous output power of a PV array can be given by,

PPV(t) = Ppeak(G/Gstandard) – αT (Tc – Tstandard)

where Gstandard* *and Tstandard* *are the standard test conditions for solar radiation and cell temperature, respectively and αT is the temperature coefficient of the PV module power which can be obtained from the datasheet.

For more information on prediction PV power in term of solar radiation using dataset, check the following papers:

- Jabar Yousif, Hussein A. Kazem, John Boland, “Predictive Models for Photovoltaic Electricity Production in Hot Weather Conditions”, MDPI- Energies, Vol.10, Issue 7, September 2017, page. 971.
- Hussein A. Kazem, Jabar Yousif, “Comparison of prediction methods of photovoltaic power system production using a measured dataset”,
*Elsevier-Energy Conversion and Management*(ECM, ISSN: 0196-8904), UK, Vol. 148, September 2017, pp. 1070–1081. - Hussein A. Kazem, Tamer Khatib, and K. Sopian, “Sizing of a standalone photovoltaic/ battery system at minimum cost for remote housing electrification in Sohar, Oman”,
*Elsevier-Energy and Building*, Netherlands, 2013, Vol. 6C, pp. 108-115.

6 Recommendations

Dear Saad Benslimane

Follow these steps:

- Solar panel watts x average hours of sunlight x 75% = daily watt-hours. As an example, let's say you have 250-watt solar panels and live in a place where you get 5 hours of sunlight per day.
- 250 watts x 5 hours x .75 = 937.5 daily watt hours.
- 937.5 / 1000 = 0.937

Kind Regards

Hi Qamar,

There is a good reference that can help you. The reference is :

Modeling of Photovoltaic Systems Using MATLAB: Simplified Green Codes, Tamer Khatib, Wilfried Elmenreich

This reference can provide you with answers to your questions.

I hope this can help you

BR

Dear Saad Benslimane ,

welcome!

In order to get the output power given the input irradiance you need to multiply the input by the conversion efficiency. You can find data for the conversion efficiency

for different commertial solar cells as a function of T and the irradiance G.

It is assumed that Eta is more less independent on G and its dependence on temperature is known.

This could be a possible solution.

Best wishes

8 Recommendations

Hi **Saad Benslimane, **

You can model the PV power output at your desired time interval using the PVLIB package in python if you have information about the particular PV panel type that you are interested. You will need to give irradiance, temperature and wind as inputs to obtain the desired power output.

Yo can have a look at the the PVLIB package here, https://pvlib-python.readthedocs.io/en/stable/. I have added some previous literature that might be useful.

Best wishes!

The graphs show examples of changes in the photocurrent Isc with a rate of 1 minute on a 10-day interval and for the selected days.

The weekly average photocurrent values correlate with the cloudiness indices.

Details on the measurement conditions are contained in the book.

- 65.29 KB20200511-20.png
- 23.95 KB20200516.png
- 17.61 KB20200519.png

Solar power output depends on your cell or module type, materials and its arrangement of array.

Normally solar irradiance determine efficiency of a panel;

Say, poly Si standard module of of power rating 125 W/m2 at standard condition 1.5 air mass but irradiation (av) 1200 w/m2 in any location

So efficiency of that location would be promoted as (125/1000)x (1200/1000) x 100% = 15%

But normally efficiency is (125/1000) x 100% = 12.5%

Total panel area required = 2500 /(125 x1.2) = 16.6 m2 ; dimension of each panel = 1.1mx 1.6 m =1.75 m2

So, total 16.6/1.75 = 10 panels are required

But normal standard 1000 w/m2 irradiance required area = 20 m2

Panel required 20/1.75 m=12 panels

1 Recommendation

Conference Paper

- Oct 2021

Get high-quality answers from experts.