航空发动机涡轮盘模锻成形的时变特征与工艺轨迹规划方法研究

Die forging technology is an important approach to manufacture the aero-engine turbine disk. However, for the issue of time-variation and nonlinearity of die forging process，there are large deviations between the actual die forging trajectory and the predesigned one, which directly affects the forming precision and performance of turbine disk. Therefore, the trajectory planning method based on the time-varying die forging forming process will be proposed for the studied turbine disk, and the main work are carried out as follows: (1) The effects of time-varying deformation parameters of the hot plastic deformation and microstructural evolution behaviors are investigated for the typical nickel-based superalloy used for aero-engine turbine disk, and the constitutive models and material database are established to accurately describle the hot plastic deformation behaviors of the studied superalloy under the time-varying conditons; (2) The optimum preform shape for the turbine disk is investigated to guarantee the shape precision. Then, the soft sensor models based on the recursive neural network method are proposed to predict the microstructural evolution of the key deformation region on turbine disk; (3) The spatial-temporal segmentation method for the die forming process are studied. According to the obtiained microstructural evolution during each sub-forging step, the die forging parameters are optimized online, and the software system for the die forging trajectory planning is developed to achieve the collaborative of shape and property in die forging forming of turbine disk. The implementation of this project can provide the theoretical basis and technical support for the improvement of manufacturing level of high performance forgings with complicated shape.

模锻成形技术是航空发动机涡轮盘制造的重要途径。但是，由于涡轮盘模锻成形过程具有强时变性和非线性特征，致使实际模锻成形工艺轨迹与预规划轨迹之间存在较大偏差，直接影响着涡轮盘的成形精度与整体性能。因此，本项目拟建立基于时变模锻成形过程的工艺轨迹规划方法，开展如下研究：（1）研究时变工况参数对涡轮盘用镍基合金热塑性变形与微观组织演变行为的影响规律，建立可精确描述时变工况条件下材料流变规律的表征模型与数据库；（2）研究涡轮盘锻件预成形坯料形状以确保其精确成形的要求，采用递归神经元网络方法建立涡轮盘关键变形区域微观组织演变规律的软测量模型；（3）研究涡轮盘模锻成形过程的时空细分方法，针对每个模锻子过程的微观组织演变规律，实时在线优化模锻成形工艺参数，集成开发模锻成形工艺轨迹规划系统，实现涡轮盘模锻成形过程的形性协同制造。本项目研究成果可为提升复杂锻件的高性能成形制造水平提供理论基础与技术支撑。

模锻成形技术是航空发动机涡轮盘制造的重要途径。但是，由于涡轮盘模锻成形过程具有强时变性和非线性特征，致使实际模锻成形工艺轨迹与预规划轨迹之间存在较大偏差，直接影响着涡轮盘的成形精度与整体性能。因此，本项目深入研究了镍基合金涡轮盘的模锻成形规律，具体研究工作包括以下几方面：（1）研究模锻工况参数对涡轮盘用镍基合金热塑性变形与微观组织演变行为的影响规律，建立可精确描述时变工况条件下材料流变规律的表征模型，实现了镍基合金涡轮盘模锻成形过程流变行为的准确预测；（2）采用多因素正交实验设计和数值模拟方法确定了涡轮盘预成形最佳坯料形状，且实现了涡轮盘模锻成形形状精度的准确控制。在此基础上，采用递归神经元网络方法建立涡轮盘关键变形区域微观组织演变规律的软测量模型，该模型能够实现涡轮盘模锻成形过程关键区域围观组织的准确预测；（3）研究了涡轮盘模锻成形过程的时空细分方法，合理地将涡轮盘的模锻成形过程细分为三个子过程，针对每个模锻子过程的微观组织演变规律，在每个模锻子过程，根据感知的模锻工况信息，预测获得当前的晶粒组织信息，并且进一步预测出模锻结束时的晶粒组织信息，进而根据模锻结束时的晶粒组织信息和设定的目标晶粒组织之间的偏差，计算下一个模锻子过程的模锻成形工艺参数。该方法实现了模锻成形过程工艺参数的实时在线调控，确保了涡轮盘零件模锻成形过程的形性协同调控。项目研究成果为提升复杂锻件的高性能成形制造水平提供了理论基础与技术支撑。项目负责人发表学术论文6篇，其中SCI 论文4篇申请国家发明专利4项，授权2项。项目执行期间新增湖南省自然科学金优秀青年基金项目1项，湖南省自然科学基金面上项目1项目，长沙市杰出创新青年培养计划项目1项，培养研究硕士研究生3名（在读）。