In modern combat， nodes differ significantly from the single-function nodes of the past， evolving towards multi-function integration. To address issues such as the characterization of node function diversity， the dynamic reconfiguration of system models， and realistic simulation of enemy network analysis in current node importance evaluations of combat systems， a dynamic reconfiguration model based on unit attributes for the enemy’s network is proposed. The concept of tactical weight is introduced， and multi-function integration of nodes and edges is achieved through attribute induction and attribute assignment. A method and algorithm for calculating the dynamic reconfiguration of network effectiveness are presented. Taking into account the actual operational context and efficiency， the node attack cost is incorporated， and a node importance evaluation model based on the contribution rate to system network effectiveness is proposed. In simulation experiments， node importance is evaluated using seven different metrics， including node effectiveness index， degree centrality， and betweenness centrality， under various spatiotemporal conditions. The decline in network effectiveness after network strikes is also analyzed. The simulation results demonstrate that the proposed method enables dynamic reconfiguration of the network post-attack， reflecting the network’s strong robustness and confirming the validity of the proposed algorithm.