Abstract: To address the real-time damage computation requirements of integrated transmission system for tracked special vehicles under complex working conditions, a method for constructing a working condition-damage standard set based on real-vehicle data is proposed in this paper. First, dynamic modeling and rapid solving techniques are employed to address the issue of missing load information. Second, kinematic segments are divided and features are extracted using the variation characteristics of dynamic torque, followed by dimensionality reduction via principal component analysis (PCA). Subsequently, a parameter adaptive fuzzy C-means (PAFCM) clustering method is designed to achieve automatic working condition segmentation and dynamic determination of the number of clusters. On this basis, the rainflow counting method and Miner fatigue criterion are applied to analyze the load and damage characteristics of the fine-grained working conditions obtained through clustering, thereby constructing a working condition-damage standard set. The effectiveness and engineering applicability of the proposed method in real-time damage computation are validated. The results demonstrate that the proposed method overcomes the limitations of existing approaches in the application of integrated transmission system, providing robust support for degradation assessment and lifespan prediction of integrated transmissions.