Fractal dimension of effective parameters on solar flat plate collector's performance

The solar radiation data play a very important role for understanding internal structure of solar systems. Due to weather conditions, solar radiation has high fluctuation. Analysis of radiation data help to optimizing these systems. Fractal dimension is one of the methods to analyze time series such as radiation. In this paper some methods such as Sevcik, Katz, MRL, PetrosianD, PetrosianC and Higuchi are used to estimate fractal dimension. These methods were coding by MATLAB software and then the data was analyzed. The result of ANOVA showed that temperature on the wet basis in the MRL method was significant at 1% and relative humidity in the PetrosianC was significant at 5%. According to results, Higuchi method had maximum mean value of fractal dimension and Petrosian method had minimum mean value of fractal dimension. Also fractal dimension of Petrosian method was less than1, therefore, Petrosian method is not reliable(because fractal dimension is a number between 1 and 2). Also, Sevcik and Higuchi was the suggested method for future researches.