Keywords and Phrases
Correlation; Expecting future; Linear Regression; Meinel and Meinel Model; SOLAR ARRAY; STOCHASTIC MODEL
Statistical approaches are often used in time series analysis, for example, to predict the future trend of a time series. Trend forecasting can be applied in many time related parameters such as: solar radiation, generation of electricity and other variables related to time series to improve the efficiency and to set the design requirements. Since the design of any solar energy system requires knowledge of the availability of solar radiation data at the location of interest. Therefore, this research seeks the application of a statistical model to fit the solar radiation time series and predict the future values. There are various methods used to estimate the hourly global solar radiation on the earth surface. However; in this research Meinel and Meinel model was used based on its fit accuracy relaying on mean bias error (MBE) and root mean square error (RMSE) tests. The study concerns to two main goals: First, predicting the future produced power of a given solar panel in a series-parallel configuration based on the present data and weather condition in order to improve the performance of the solar panel. Second, there was an attempt to relate all 24 sensors that located on a solar panel so that we can estimate the sun radiation at each part of a hypothetical solar array using one sensor's reading only. In addition, as the availability of the solar radiation related with the climate conditions, an attempt has been made to correct the predicted data under different climate conditions. Linear regression method was used for the purpose of fitting the next point in predicting of the solar radiation, while the covariance, correlation factors and slopes among all sensors are used to relate the whole parts of the panel. The proposed technique yields an acceptable result within an average mean squared error (MSE) of 2.6%.
Kimball, Jonathan W.
Electrical and Computer Engineering
M.S. in Electrical Engineering
Saudi Arabian Cultural Mission to the United States
Missouri University of Science and Technology
ix, 75 pages
© 2012 Faris Alfaris, All rights reserved.
Thesis - Open Access
Electronic OCLC #
Alfaris, Faris, "Stochastic model for solar sensor array data" (2012). Masters Theses. 6947.