|
摘要：
|
|
为提高锻件质量和成品率,有必要建立一种适合于实时控制的锻件成形过程模型。利用有限元模拟技术对涡轮盘的等温成形过程进行了虚拟正交试验,通过对成形过程的载荷——行程曲线的分析,建立了粉末高温合金涡轮盘件等温成形过程的人工神经网络(ANN)模型,并将其映射成模拟电路模型。以此模拟电路模型为参考模型,应用于模型参考自适应控制(MRAC)系统,对涡轮盘件等温成形过程进行控制。结果表明,所建立的ANN模型及其模拟电路模型对粉末高温合金涡轮盘件等温成形过程的拟合精度很高,且控制参数始终与模型输出相吻合,为实现盘件成形过程的实时控制奠定了基础。
|
|
In order to increase forging qualities and yield of powder metallurgical(P/M) turbine disc,it is necessary to establish a dynamic model fitting for real-time control of P/M turbine disc isothermal forging process.In the present work,the virtual orthogonal experimental(VOE) of P/M turbine disc isothermal forging was accomplished using finite element simulation.The deformation law of the material was analyzed.The artificial neural network model of P/M turbine disc isothermal forging process was obtained by using the data of VOE to train the BP(Back Propagation) network.Then the analog-circuit model of P/M turbine disc isothermal forging process was established by using the learning results of ANN.The analog-circuit model was applied to the reference model of model reference adaptive control system to realize real-time control.The results show that the ANN model and the analog-circuit model have high fit precision for P/M turbine disc isothermal forging process.The controlling parameters are always coincidence with the output of models.
|
|
基金项目：
|
|
天津大学青年教师基金资助项目(5110105);;
航空基础科学基金资助项目(03H53048)
|
|
作者简介：
|
|
|
|
参考文献：
|
[1]Alkan A D,Gulez K,Yilmaz H Design of a robust neural net-work structure for determining Initial Stability Particulars ofFishing Vessels[J].Ocean Engineering,2004,31(5-6):761-777.
[2]潘江峰,钟约先,袁朝龙.基于遗传算法和神经网络技术的板料拉深成形参数优化[J].锻压技术,2006,31(6):44-47.
[3]胡守仁,余少波,戴葵.神经网络导论[M].长沙:国防科技大学出版社:1993,10.
[4]Madadi Kojabadi H.Simulation and experimental studies ofmodel reference adaptive system for sensorless induction motordrive[J].Simulation Modelling Practice and Theory,2005,13(6):451-464.
[5]李久胜,石松伟,陈宇,等.充液拉深装备中压边压力闭环控制系统调节器的设计[J].锻压技术,2005,30(2):32-35.
[6]刘玉红,李付国,吴诗?.细晶态FGH96热成型时的流动行为研究[J].航空学报,2003,24(3):278-281.
[7]刘玉红.粉末高温合金等温成形模拟与控制[D].西安:西北工业大学,2004.11
[8]Cai Kai,Xia Juntao,Li Longtu,et al.Analysis of the electri-cal properties of PZT by a BP artificial neural network[J].Computational Materials Science,2005,34(2):166-172.
[9]Haque E Mohammed,Sudhakar K V.ANN back-propagationprediction model for fracture toughness in microalloy steel[J].International Journal of Fatigue,2002,24(9):1003-1010.
|
|
服务与反馈：
|
|
【文章下载】【加入收藏】
|
|
|
|
