GA优化的湖泊富营养化评价的普适公式探讨

Approach on a Universal Optimum Formula of Lake Eutrophic Evaluation Based on Genetic Algorithm

在适当设定富营养化指标的“本底值”情况下,当指标值用于对应“本底值”的相对值表示时,可采用S型曲线描述湖泊富营养化的发展程度,公式中的参数可视为与指标特性无关,采用遗传算法对公式参数优化,得到对多项指标均适用的富营养化程度的指数计算公式。并提出用广义模糊对比因子赋权新方法计算富营养化综合指数。该评价方法物理意义明确,计算简单,使用方便,具有普适性、可比性和实用性。

Based on the hypothesis of 'background values' for eutrophication indexes, an universal formula of S-type curve, in which parameters are independ of the characters of indexes, was presented for eutrophic state of lake, as the relative values of indexes were used instead of their monitoring values. Optimum parameters in formula were carried out using genetic algorithm to derive an universal formula of evaluation of lake eutrophication suited to multi-index. The comprehensive index of lake eutrophication can be calculated in the aid of the general fuzzy contrast weighting for the partial indexes. The results showed that this evaluation method has clearness of physical meaning, simplicity, generality, comparison and practicality.