#TITLE_ALTERNATIVE#
Correlation analysis shows a strong correlation between solar activity and cosmic ray flux and solar constant. A higher correlations (but with opposite sign) are found between solar constant variations and sunspot number variations than between variations in cosmic ray flux and solar constant. It wa...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/14574 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Correlation analysis shows a strong correlation between solar activity and cosmic ray flux and solar constant. A higher correlations (but with opposite sign) are found between solar constant variations and sunspot number variations than between variations in cosmic ray flux and solar constant. It was found a positive correlation between solar constant and sunspot number, with correlation coefficient about 0.89 and 0.96 for monthly and yearly data, respectively. In other hand, a negative correlation between solar constant and cosmic ray flux, i.e. -0.65 and -0.69. It was found a negative correlation also between solar activity and cosmic rays flux, i.e. -0.73 (monthly) and -0.77 (yearly). When solar activities decrease until minima condition, the cloud cover rate increase due to secondary ions produced by cosmic rays. The increasing of the cloud cover rate cause the decreasing of solar constant value and solar radiation on the earth surface.<p>Monthly solar radiation prediction for 14 locations in Indonesian region using Adaptive Neuro-Fuzzy Inferrence System (ANFIS) model has been done. Sunshine duration and solar radiation measurement of period 1994-2003 are used as input data. Generally, prediction using ANFIS method give a good result with low Root Mean Square Error (RMSE) relatively. Prediction time-length varies of 3 to 9 months with error prediction less than 10%, depends on characteristic and data length. In addition, prediction result has been validated using ground data and satelite data from NASA SSE website with error validation less than 10%.<p>In addition, the ANFIS outputs were used for designing a solar water pumping system by using Lost of Energy Probability (LOEP) method for application purpose. Using this method size of the system, i.e. area and number of PV module and battery capacity can be calculated. It was found 4 match locations for this system, i.e. Makassar, Pontianak, Padang and Bengkulu because of their higher dirrect radiation than diffuse radiation component. |
|---|