用自适应搜索算法拟合太阳电池I-V曲线

分析了序贯数论优化算法(SNTO)用于太阳电池I-V曲线拟合时的不足之处，结合遗传算法中自适应搜索的思想，用自适应搜索算法改进了SNTO算法．该算法的基本过程是：以太阳电池等效电路的数学模型为基础建立目标函数；在拟合时，利用自适应搜索算法来随机地确定收缩比，即用第t次拟合中的计算结果来确定第t 1次的最优参数可取值范围，使各步中目标函数的参数空间构成一个最优决策序列．拟合结果表明，该算法比直接采用序贯数论优化算法(SNTO)具有更少的计算量，更高的收敛性和鲁棒性。

Correlating the I-V Curve of Solar Cells with Adaptation Search Algorithm

The shortages of Sequential Number-Theoretic Optimization for correlating the I-V curve of solar cells are analysed. A adaptation search algorithm is proposed in combination with the adaptation search algorithm of genetic algorithm. The key steps are: Based on the mathematical model of solar cell's equivalent circuit, the objective function is established. During the correlating process, the shrink ratios are generated at random by the use of adaptation search algorithm, i.e., the correlation results in the t bout, are used for finding the optimization ranges of the parameters in the t+1 bout. Finally a optimal decision sequence of parameter space for objective function of every step is formed. Computed result shows that the accelerating correlating algorithm is more easier to compute, more convergent and robustious than Sequential Number-Theoretic Optimization.