GH4169合金涡轮盘锻造成型的数值模拟和分析

利用Gleeble-3500型热模拟试验机研究GH4169合金在不同温度和变形速度下的热变形行为,建立该合金的高温流变应力模型.用Deform-3D对GH4169镍基高温合金涡轮盘锻造成型过程进行数值模拟,比较不同变形速度和不同变形温度下工件的变形行为.结果表明:相对于变形速度,变形温度对锻件性能的影响更加明显;较高的变形温度有利于材料的动态恢复和再结晶,使组织均匀,但过高的终锻温度会使晶粒尺寸变大,进而影响涡轮盘的机械性能.     

基于深度学习的GH4169合金组织表面缺陷检测

镍铁基高温合金GH4169合金由于其良好的综合性能被广泛应用在航空、石化、核能等行业,其冶炼工艺复杂,制备工艺路线较长,因此在铸造过程中会不可避免地产生大量组织缺陷,这些缺陷会对合金的性能造成重要的影响。为了消除合金组织表面缺陷,需要研究合金组织表面缺陷的分布和成因以此提高合金的冶炼技艺。但传统人工检测GH4169合金组织表面缺陷效率低、精度差,很难用于检测大棒材。因此,为了实现组织表面缺陷的自动检测,在RetinaNet网络结构的基础上提出了一种CA-RetinaNet网络结构用于GH4169合金组织表面缺陷检测,该方法主要增强了网络检测小缺陷的能力。首先,在特征提取网络中使用了CA-Resnet结构,引入轻型注意力机制对感兴趣目标进行特征权重增强,提高了含有目标通道的权重;然后对RetinaNet网络中的特征金字塔网络进行了优化,重新构建了特征金字塔网络的底层结构,以获取更大的特征图检测小缺陷。利用CA-RetinaNet网络模型在GH4169合金组织表面缺陷数据集上进行检测实验,取得了较高的准确率,相较于原始RetinaNet网络,mAP值提升了8.6%,极大地提升了网络的检测精...     

基于微粒群算法的转炉合金加入量多目标优化模型

针对转炉炼钢过程中钢水出钢环节脱氧合金剂加入量确定的问题,建立了合金加入量优化计算模型.首先,利用支持向量机建立钢包元素含量预测模型,将该模型的预测结果与元素规定含量的误差作为优化模型的一个目标函数,合金成本作为另一个目标函数;然后利用预测收得率方法计算合金加入量并作为优化模型的初始值,使用改进的动态邻域多目标微粒群算法对优化模型进行求解,有效地找出Pareto最优解集,并根据实际需要得出较好的优化结果.仿真实验表明了该优化模型的有效性.     

整体涡轮盘通道电火花加工方法的研究

本文在分析带冠弯扭叶片整体涡轮盘加工状况的基础上,提出了采用复杂成型电极靠模加工带冠弯扭叶片整体涡轮盘通道的加工方案,并讨论了该方案中整体涡轮盘通道几何造型,成型电极设计方案和加工工艺、成型电极进给路径确定等关键性技术问题。在给出该工艺方法的实验结果后,本文还讨论了该工艺方法产生误差的原因及其解决方案。 一、引言 广泛应用于航空、航天工业上的小型弯扭叶片高温合金整体涡轮,因其工件材料难于机械切削,工件加工空间又十分有限,所以很难采用传统的机械方式进行加工。而采用电火花加工的方法,则不受工件材料硬度…     

GH4169合金单/双层靶板高温高速冲击数值仿真

航空发动机机匣包容性是保障发动机安全、可靠工作的必要要求,涡轮机匣受到飞脱叶片撞击时处于高温环境下,因此如何考虑高温环境对机匣包容性的影响至关重要.针对涡轮机匣典型的GH4169高温合金材料,在LS-DYNA中采用Johnson-Cook本构模型和Thermal-constant关键字开展了单/双层靶板在常温与高温下抗...     

基于Cosserat连续体理论的疲劳寿命概率可靠性分析

基于Cosserat连续体理论建立考虑应变梯度效应的粉末高温合金弹塑性模型.该模型考虑微尺度长度效应,并能消除缺陷局部化问题求解时的网格依赖性.基于参变量变分原理提出一种高效的求解方法,该方法将原非线性问题转化为互补问题求解,可大大提高求解效率和收敛性.针对一种假想涡轮盘模型,将Cosserat尺度参数和Basquin斜率作为随机变量,分别采用蒙特卡洛法与二阶矩法进行疲劳寿命概率可靠度计算.结果表明二阶矩法在保证精度的情况下具有较高的效率.     

基于深度学习的空气动压轴承涡轮冷却器起浮压力预测

为提高对空气动压轴承涡轮冷却器起浮压力预测的准确性,提出基于深度学习的空气动压轴承涡轮冷却器起浮压力预测方法。先构建空气动压轴承涡轮冷却器起浮的动力学模型,然后采用非线性接触力神经网络模型获取起浮压力数据集参数,以剪应力、流体力学和速度梯度参数融合分析方法建立起浮压力预测的约束指标参数分配集,构建空气动压轴承涡轮冷却器的承载力、刚度及动态刚度拟合模型,最后基于深度学习完成冷却器起浮压力预测过程指标参数的寻优,以此提升空气动压轴承涡轮冷却器起浮压力的准确预测。仿真测试结果证实,该方法对空气动压轴承涡轮冷却器起浮压力预测的可靠性较好,预测误差较低,迭代100次以后误差仅为0.01 kPa,优于对比方法,具有较大的应用价值,为实现对空气动压轴承涡轮冷却器的承载设计奠定理论和模型基础。     

基于灰色线性回归组合模型铁路客运量预测

铁路客运量的准确预测是铁路旅客运输组织工作的重要基础和主要依据之一,准确预测铁路客运量是铁路运输企业面向市场、把握未来的重要保障.针对铁路客运系统是一个信息不完全的灰色系统,运用灰色预测理论构建灰色模型GM(1,1)与线性回归的组合模型,用于对未来五年内河南省铁路客运量进行预测.研究表明,模型改善了原线性回归模型中没有指数增长趋势和基本灰色预测模型中没有线性因素的不足,和单一模型相比,预测精度更高,预测结果更为可靠,具有较高的实际应用价值,也为铁路客运量预测研究提供了新的途径.     

改进在线最小二乘支持向量机的合金成分预测

针对硅锰合金埋弧冶炼过程中合金成分检测困难、离线化验滞后大、难以实时控制的问题,提出一种改进在线最小二乘支持向量机（IOLSSVM）的合金成分预测模型。该模型对每一个新增样本采用增量式学习,根据样本对模型贡献的不同删除样本集中对模型贡献最小的样本数据,利用递推计算增强模型的在线学习能力。将此模型应用于30MVA硅锰合金埋弧炉冶炼过程合金成分预测,实际生产运行数据表明了此方法的有效性。     

基于多算法优化的矿用掘进机刀盘扭矩的实时预测方法

本文提出了一种改进的矿用掘进机刀盘扭矩的多算法优化实时预测的方法和一种基于贝叶斯和交叉验证的模型超参数优化算法,开发了刀盘扭矩的实时预测模型,经验证,模型的平均绝对百分比误差为7.3%,表明所提出的模型适用于刀盘扭矩的实时预测。此外,还引入了基于上述基础模型的增量学习方法,以提高模型在掘进过程中的适应性,适用于刀盘扭矩的实时预测,可有效提高挖掘过程中模型的预测精度和增强泛化能力。     

GH4169晶粒尺寸的类圆映射超声评价方法

GH4169组织结构复杂,使用单一超声参数难以对其进行准确评价,而多超声参数评价方法存在参数集难以合理选择的问题。利用类圆映射提出一种新的全参数超声评价方法,将全部超声参数投影至二维圆内空间,构建投影多边形并提取具有全局超声信息的二阶特征;进而与晶粒尺寸进行高阶多项式拟合,将评价问题转化为以拟合误差最小为目标、类圆映射参数为设计变量的优化问题;最后利用狼群优化算法求解并获得最终的全参数超声评价模型。实验结果表明,相比其他超声评价方法,新方法具有精度高、鲁棒性好等特点。     

Optimal design of aeroengine turbine disc based on kriging surrogate models

A design optimization method based on kriging surrogate models is proposed and applied to the shape optimization of an aeroengine turbine disc. The kriging surrogate model is built to provide rapid approximations of time-consuming computations. For improving the accuracy of surrogate models without significantly increasing computational cost, a rigorous sample selection is employed to reduce additional design samples based on design of experiments over a sequential trust region. The minimum-mass shape design of turbine discs under thermal and mechanical loads has demonstrated the effectiveness and efficiency of the presented optimization approach.     

热加工工艺对GH738金相组织及超声特性的影响

针对不同热加工工艺下GH738锻件金相组织的无损评价问题,对经过固溶、稳定化、时效和钎焊处理的合金分别进行金相、超声检测,讨论热加工工艺变化对金相组织、超声检测特征值的影响,分析金相组织参数与超声参数之间的相关性。结果表明：晶粒直径变化对声速影响微弱,在热加工过程中,固溶处理后声速下降,稳定化与时效处理后声速不断上升,钎焊处理后声速开始下降且达到最小值。衰减系数、非线性系数与晶粒直径的变化规律趋于一致,随着热加工过程的进行,呈现出固溶阶段快速上升、稳定化后缓慢上升并趋于平缓的趋势,超声衰减与非线性系数可用于表征GH738合金的晶粒直径及织构不连续性。     

An interactive optimization procedure applied to the design of gas turbine discs

An interactive procedure based on the Complex method is applied to the design of a minimum weight gas turbine air-cooled disc. The disc profile is defined by spline interpolation and an interactive video display terminal is used to show the results of the thermal and stress analyses at each step of the optimization. The advantages of interactivity are shown in a practical example.     

兆瓦级风力发电机组轮毂强度数值分析

基于GL规范,采用有限元法对某兆瓦级风力发电机组轮毂进行极限强度分析.对轮毂S N曲线进行修正,分析疲劳计算需要考虑的载荷工况,使用GH Bladed软件仿真得到疲劳计算所需的载荷时间序列.使用ANSYS/nCode软件对轮毂进行疲劳寿命分析,结果表明轮毂的极限强度和疲劳强度均满足设计规范的要求.分析结果可以为兆瓦级风力发电机组轮毂的结构设计提供参考.     

Heat transfer in cylindrical shrouded cavities

The flow field, temperature field and the heat transfer rates in cylindrical shrouded cavities with rotation, recirculation and coolant through-flow have been analyzed numerically. Two cavity configurations are considered. In the first configuration, a heated cylindrical shroud is enclosed by a stationary insulated stator disc and a rotating insulated rotor disc. The coolant air enters the cavity by a central opening in the rotor and exits through an annular gap at the rim of the rotor. The second configuration studies the heat transfer from an air cooled gas turbine disc using the model of a plane disc rotating close to an insulated shrouded stator. The coolant enters centrally through the stator disc and exits radially through a gap between the shroud and the rotor. The flow field and heat transfer rates are computed for several values of coolant flow rate, the rotor swirl speed, the cavity aspect ratio and the exit gap width in the two cavity configurations. The swirl of the rotor changes immensely the flow pattern, recirculating zones and isotherms inside such cavities. In general, increasing coolant flow rate, decreasing swirl and decreasing aspect ratio enhances the heat transfer from the shroud in the first cavity configuration. For the second cavity configuration, the heat transfer rates increase with increasing coolant flow rate, increasing swirl of the rotor, increasing size of the cavity and decreasing exit gap width between the stator and the rotor.     

Strength Analysis of Steam Turbine Diaphragm with a Crack

The diaphragm of steam turbine is one of the main parts of the passage unit for air flowing and the diaphragm of steam turbine is generally made by welding, but during the course of welding, some cracks often have happened, these cracks threaten safety production on the spot, so it is necessary to analyze the strength of steam turbine diaphragm with a crack. For studying its working condition and performance with a crack, the three-dimensional finite element model(FEM) with the software ANSYS 10.0 is established according to the actual working parameters. With the model, the deformation results and stress field results are analyzed and the maximum displacement and stress value are calculated, meanwhile, the maximum stress change with crack depth of steam turbine diaphragm is analyzed. It provides a good foundation for solving the problem in production.     

Modelling the vibration response of a gas turbine using machine learning

This work deals with modelling the vibration response of a gas turbine obtained during the start-up process until reaching the nominal speed for power generation. Analysing the vibrations of a complex systems like a gas turbine is useful for the diagnostic of faults or damages in the internal mechanical components of the different stages that integrate a turbine. This work focuses on the study of the shaft vibrations of the bearing radial type mounted between the shaft and the bearing compressor associated with the speed of the turbine. This relationship is studied using experimental data collected from a particular gas turbine model. In particular, we propose a methodology to synthesize a computational model following a supervised learning approach implemented through different machine learning techniques, including a multi-layers perceptron network, support vector machine (SVM), random forest (RF) and genetic programming (GP) with local search. Results show that SVM, RF and GP perform very well in this task, producing accurate predictive models. Moreover, there are some interesting trade-offs between the methods, regarding generalization error, overfitting and model interpretability that are relevant for future applications and research.     

Disc cutters plane layout design of the full-face rock tunnel boring machine (TBM) based on different layout patterns

The quality of plane layout design of the disc cutters for the full-face rock tunnel boring machine (TBM) directly affects the balance of force distribution on the cutter head during the excavating. Various layout patterns have been adopted in practice during the layout design of the disc cutters. Considering the engineering technical requirements and the corresponding structure design requirements of the cutter head, this study formulates a nonlinear multi-objective mathematical model with multiple constraints for the disc cutters plane layout design, and analyses the characteristics of a multi-spiral layout pattern, a dynamic star layout pattern and a stochastic layout pattern. And then a genetic algorithm is employed to solve a disc cutters’ multi-spiral layout problem, and a cooperative co-evolutionary genetic algorithm (CCGA) is utilized to solve a disc cutters’ star or stochastic layout problems. The emphasis was put on the study of superiority of three different layout patterns. Finally, an instance of the disc cutters’ plane layout design was solved by the proposed methods using three different kinds of layout patterns. Experimental results showed the effectiveness of the method of combining the mathematical model with the algorithms, and the pros and cons of the three layout patterns.