基于灰色物元的变压器多指标绝缘状态评估方法

为准确评估变压器绝缘状态,提出一种基于灰色物元分析法的变压器绝缘状态评估方法。首先,选取变压器的绝缘状态检测量和实际运行状态作为评估指标,建立多指标评估体系;然后,针对指标信息的模糊性,用灰色物元表示评估指标并将其特征值量化,从而能准确地对变压器进行定性和定量评估;同时,考虑到指标之间并不独立,基于物元理论构造关联函数,将多指标评估转化为单目标评估;但指标间相关性将导致信息重复,为解决该问题,将变异系数法和复相关系数法2种客观赋权法组合,根据指标在综合评估中的作用计算权重,该权重反映了各指标的实际状态;最后,应用改进方法计算灰色关联度,确定变压器所处的绝缘老化等级。实例证明该方法合理可行。

Multi-indices insulation condition assessment method for transformer based on grey matter-element method

In order to assess the insulation condition of transformer accurately,a comprehensive evaluation method is proposed on the basis of the grey matter element analysis method.First of all,measurement results of insula tion parameters and the actual operaling condition of transformers are selected as evaluation indices to establish a multi-indices evaluation system.Then,the grey matter elements are utilized to quantify eigenvalues of indices and then the condi-tion of transformer more can be evaluated qualitatively and quantitatively.Meanwhile,the independence among indices is taken into account and correlation functions are constructed based on the matter-element theory.The multi indices eval-uation problem is then transformed into a single objective assessment,However,the correlation between indices would bring duplicate information.To solve this problem,two objective weighting methods,which are the varia tion coefficient method and the multiple correlation cofficient method,are employed to calculate weights according to the importance of the indices.Finally,the improved method is applied to calculate the degree of grey correlation and the aging level of transformer insulation is evaluated.It is shown that this method is reasonable and feasible.