电网故障诊断的决策表约简新算法

采用粗糙集理论的决策表方法对电网进行故障诊断，其关键过程就是决策表的简化过程(即约简)。为此，提出了基于粗糙集理论与二元逻辑运算相结合的属性约简算法以及改进的值约简算法，并将其应用于由断路器和保护作为条件属性、故障区域作为决策属性的诊断决策表的约简过程中，利用决策表的约简形成综合混合知识模型。整个算法简单、快速，可有效应用于电网的故障诊断中。运用Visual C 编程实现了对故障算例决策表的约简过程，结果证明了该约简算法的正确性。

NEW REDUCT ALGORITHM OF DECISION TABLE FOR ELECTRIC POWER GRID FAULT DIAGNOSIS

The decision table of rough sets theory is used for the fault diagnosis of electric power grids and the reduction process of the decision table is the key. Therefore, this paper presents a new attribute reduction algorithm and an improved value reduction algorithm based on the combination of rough sets theory and duality Boolean calculation. The algorithm is used to deal with the reduction process of the decision table, including all kinds of fault cases with the signals of protection relays and circuit breakers as condition attributes and fault sections as value attributes. Using the reduction of decision table, the general mixed knowledge model is formed. The whole algorithm is simple, fast and practical in electrical network fault diagnosis. Finally, the algorithm is programmed using VC and the results of an example are given.