舰船成本预测系统中的自变量约简方法

针对舰船成本预测系统中自变量众多且相互关系错综复杂的特点,提出了用模糊粗糙集方法对舰船成本预测系统中的自变量进行约简。在模糊粗糙集方法的框架下,给出了模糊不可分辨关系和相对约简的定义,构建了基于模糊粗糙集的舰船成本预测系统自变量约简模型,并且针对实际的舰船成本预测系统一般是混合数据系统的问题,对广义差别矩阵的定义进行了改进,设计了相应的启发式约简算法。实例分析结果表明:文中所提的方法不仅能使约简的自变量大为减少,而且约简的效果较传统方法更好,且计算过程也更简单。

An independent variables reduction method used in ship cost forecasting system

According to a lot of independent variables and their complex relations in the ship cost forecasting system,a fuzzy rough set method is proposed for independent variables reduction.On this basis,the definitions of fuzzy discernibility power and relative reduction are presented so as to establish an independent variables reduction model.Considering that the actual ship cost forecasting system is generally a hybrid data system,an improvement in the definition of general discernibility matrix leads to the construction of a new heuristic algorithm for reduction.The practical experiment shows that compared with the traditional method,the method proposed has better performance,and it can also reduce the number of independent variables and simplize the calculation process.