Traditionally, suboptimal diagnosis scheme would be evaluated only considering FDR and cost in order to achieving optimal di-agnosis one, but the method has some shortages. It needs to consider multi-attribute indexes and make a choice. A decision matrix is trans-ferred to a normal one using range transforming, weights of attributes can be created by a way of information entropy, and TOPSIS (Tech-nique for Order Performance by Similarity to Idea Solution) can serve as utility index, so the optimal diagnosis scheme model of multi -attrib-ute decision-making can be built using TOPSIS based on an objective entropy weight. In an example, seven attributes are considered, such as: cost, FDR, structure matching, weight matching, failure rate, expansibility and operating, and the result shows that the method is feasi-ble and effective.%为了在次优诊断方案中选出最优诊断方案,传统的依靠故障检测率和费用评价次优诊断方案存在明显不足；为了能够综合考虑多个属性指标进行决策,采用极差变换法对决策矩阵进行正规化变换,采用信息熵法产生各属性的权值,采用TOPSIS法作为综合效用判断指标,建立基于客观熵值权重的TOPSIS多属性最优诊断方案决策模型;示例中综合考虑7种属性进行决策,分别是费用、FDR、结构匹配、重量匹配、故障率、扩展性和操作性,结果显示该方法是可行的、有效的.
展开▼
