Abstract
A solar PV array system is comprised of the following components - solar cells, panel modules, and an array system. Thus, overall optimal design of a solar PV system involves the optimal design of the components at three levels - solar cell, panel module, and array. The conversion efficiency, power output, and incident solar energy pertaining to the requirements of seasonal demands are to be considered in the process. At the solar cell level, cell performance depends on solar cell structure, top contact design, and cell size. The correlations between cell structure, cell size and top contact design are investigated. At the PV panel module level, the optimization of a PV panel module is investigated based on the optimal design of individual solar cells for maximizing the power output. The role of the PV panel module is interactive between solar cells and the array system and is composed of a number of solar cells and panel modules. In designing a solar PV array system with cost considerations, the performance of a solar PV array system is investigated based on the performance of its subsystems - the solar cell and the panel module – as well as the cost of the array system. The optimal design of an array system is considered by formulating six single-objective optimization problems – the maximization of the conversion efficiency of the cells, power output of the arrays, annual monthly average incident solar energy, lowest month’s and highest month’s incident solar energy and minimization of cost. Multi-objective optimum designs of a solar cell, flat plate solar PV array system and compound parabolic concentrator (CPC) PV collector system are also considered by using mathematical techniques. Game theory and fuzzy set theory methodologies are used for finding the solution of multi-objective optimization problems derived from the results of single-objective problems using genetic algorithms of ga (program MATLAB). For a solar cell, the multi-objective optimization is constructed using two objectives – the maximization of the conversion efficiency and power output. The resulting multi-objective optimization (of a solar cell) is investigated with varying intensities of sunlight and by placing constraints on the minimum permissible conversion efficiency while maximizing efficiency and power output. Multilevel system optimization problems are solved using game theory and fuzzy set theory for finding a compromise solution of the six-objective optimization problems which are related to conversion efficiency, power output, annual incident solar energy, winter incident solar energy, summer incident solar energy and total cost of the PV array system. In the case of a solar CPC collector system, there are three single-objective problems: annual monthly average of incident solar energy, lowest month’s incident solar energy and cost. Game theory methodology is used for finding a compromise solution in the process of constraints stated. The aim of uncertainty analysis is to predict the performance of a component or system in the presence of uncertain parameters. Uncertainty analyses of a solar cell, flat plate PV array system and CPC PV collector system are considered using probabilistic and fuzzy analysis methodologies. In probabilistic analysis, the random variables of a solar cell and solar PV array system include geometric design variables (except for integer values) and uncertain design parameters of top metallic contact. The solar cell and solar PV array system have been investigated by varying the values of the weight of mean and coefficient variations and illustrations by applying the parametric study related to the probabilistic efficiency of a solar cell and solar PV array system. The fuzzy membership functions are used for modeling the uncertain or imprecise design parameters of a solar PV system. Triangular membership functions are used to represent the uncertain parameters as fuzzy quantities. Fuzzy arithmetic operations and extension principles are used for finding the membership functions of the fuzzy response parameters of the system. In the case of a solar cell, the deviations of solar cell performance including the conversion efficiency and power output from the crisp value are investigated by varying α-cut interval levels and uncertain input parameters of different fuzzy confidence intervals. In the case of a CPC PV collector system, the responses from applying uncertain input parameters of different fuzzy confidence interval levels are investigated by using the crisp values of annual monthly average incident solar energy, lowest month incident solar energy, and cost. Also, the variations of three single-objective problems are represented by using a triangular shape with respect to various fuzzy interval confidence levels.