应用GS-FCM复合聚类的配网项目造价分析研究

为有效控制城乡配电网不停电作业下台区新建工程项目造价问题，考虑基于不停电作业角度开展工程项目造价评价及聚类预测指标建立。利用GS-FCM复合聚类模型开展配电网项目指标聚类及评价预测工作，提高模型对复杂高维度评价数据处理能力，通过寻优初始聚类数及合理优化聚类中心点以提高模型聚类预测精度，基于该套模型实现不同地域项目、投资差异及作业难度的项目聚类。通过GS-FCM复合聚类模型实现同类型项目同质性特点搜集及模糊整合评价，并且基于既有典型案例成功经验开展属类项目成本分析及造价管控，提出适用于同属类项目成本控制策略。

Research on the Cost Analysis of Distribution Network ProjectsUsing GS-FCM Composite Clustering

To effectively control the cost of new projects in urban and rural distribution networks without power outage， this paperestablishes the cluster prediction index system and evaluates the project cost based on the perspective of power outage. The GS-FCMcomposite clustering model is used to conduct the index clustering， evaluation and prediction of the distribution network projects.The processing ability of the model for complex high-dimensional evaluation data as well as the clustering and prediction accuracyof the model are improved by optimizing the initial clustering number and the clustering center point. The project clustering ofprojects in different regions， different investment, and different operation difficulties are realized based on the proposed models.Through the GS-FCM composite clustering model， the homogeneity characteristics collection and fuzzy integration evaluation of thesame type of projects are realized， the cost analysis of generic projects are carried out based on the successful experience of existingtypical cases， and the cost control strategy suitable for the same type of projects is put forward.