传统的基于大量试验与经验的新材料研发过程效率低、周期长、成本高，而将高效试验技术与快速计算模拟预测协调融合的智能方法可极大缩短新材料研发与工程应用周期并降低成本。短切玄武岩纤维增强聚乳酸（BF/PLA）复合材料天然绿色环保可降解，作为汽车部分内外饰等零部件的理想替代材料之一，具有广阔发展空间。为研究不同纤维参数配比对BF/PLA复合材料热性能的影响机制，快速开发适宜的车用零部件材料，该文对多种纤维参数配比下BF/PLA复合材料的热性能分别进行试验，通过数据相关性分析，探讨了不同纤维参数配比对复合材料热性能的影响；利用三因素方差分析方法F值提出纤维参数拟中心化方法，建立玻璃化转变温度和结晶度与中心化变量间的拟合归元函数；基于拟合归元函数和多项式指导函数，将所拓展的新的纤维质量分数、直径和长度参数配比代入，得到玻璃化转变温度和结晶度在新的纤维质量分数、直径和长度参数配比下的中心化变量，并利用拟合归元函数得新参数配比下BF/PLA复合材料的玻璃化转变温度和结晶度，完成利用拟合归元函数预测更多纤维参数配比下复合材料热性能的预测，得到的决定系数分别为0.887 0和0.855 1，且预测精度在实际工程可接受范围内。对某车企实际车型车门内板的热性能进行有限元分析，将预测数据模拟结果与原车模拟结果进行对比，结果表明，利用该文所提方法所选较优配比复合材料的热性能略优，且研发效率显著提升，验证了所提方法的有效性，为未来车用复合材料的快速开发、降低成本、选择替代和绿色设计提供了重要理论指导和方法参考。

The traditional process of new material development, heavily reliant on extensive experimentation and empirical knowledge, suffers from low efficiency, prolonged cycles, and high costs. Integrating efficient experimental techniques with rapid computational simulation and prediction through intelligent methods can significantly shorten the R&D and engineering application cycle while reducing costs. Chopped basalt fiber-reinforced polylactic acid (BF/PLA) composites, being naturally green, environmentally friendly, and biodegradable, represent an ideal alternative material for automotive interior and exterior components with considerable development potential. In order to study the influence of different fiber parameter ratios on the thermal properties of BF/PLA composites and facilitate rapid development of suitable automotive component materials, this study first conducted experiments on the thermal properties of BF/PLA composites under various fiber parameter ratios. Through data correlation analysis, the effects of different fiber parameter ratios on composite thermal properties were examined. Using F-values from three-factor ANOVA, a fiber parameter quasi-centralization method was proposed, establishing fitted regression functions between glass transition temperature/crystallinity and centralized variables. Based on these regression functions and polynomial guiding functions, new ratios of fiber mass fraction, diameter, and length parameters were incorporated to obtain centralized variables for glass transition temperature and crystallinity under the new parameter ratios. The glass transition temperature and crystallinity of BF/PLA composites with new parameter ratios can be obtained by fitting regression function, and the thermal properties of composites with more fiber para-meter ratios can be predicted by fitting regression function, and the determination coefficients are 0.887 0 and 0.855 1 respectively, with prediction accuracy within practically acceptable engineering ranges. Finite element analysis of thermal performance for a vehicle door inner panel demonstrated slightly superior thermal properties using the optimized composite material selected through the proposed method, along with significantly improved R&D efficiency, validating the effectiveness of the approach. This provides important theoretical guidance and methodological reference for rapid development, cost reduction, material substitution, and green design of future automotive composites.