A three-fold approach for job shop problems: A divide-and-integrate strategy with immune algorithm

This paper presents a novel divide-and-integrate strategy based approach for solving large scale job-shop scheduling problems. The proposed approach works in three phases. First, in contrast to traditional job-shop scheduling approaches where optimization algorithms are used directly regardless of problem size, priority rules are deployed to decrease problem scale. These priority rules are developed with slack due dates and mean processing time of jobs. Thereafter, immune algorithm is applied to solve each small individual scheduling module. In last phase, integration scheme is employed to amalgamate the small modules to get gross schedule with minimum makespan. This integration is carried out in dynamic fashion by continuously checking the preceding module's machine ideal time and feasible slots (satisfying all the constraint). In this way, the proposed approach will increase the machine utilization and decrease the makespan of gross schedule. Efficacy of the proposed approach has been tested with extremely hard standard test instances of job-shop scheduling problems. Implementation results clearly show effectiveness of the proposed approach.     

基于数据挖掘的工艺参数优化研究

提出了基于数据挖掘的机械加工工艺参数寻优方法,根据已有的样本训练数据建立决策树分类器和神经网络模型,针对要求的加工目标,通过对决策树的分类规则进行提取,生成预测数据集,结合建立的神经网络模型,迅速准确的预测出对应的加工工艺参数,并通过磨削实验验证了此方法的可行性。     

Material phase transformation at high heating rate during grinding

The modeling of maraging steel phase transformation in grinding process is presented in this article. Specifically, heating rate and contact zone temperature are examined to quantitatively link material properties, wheel topography characteristics and process parameters to the kinetics of diffusion-controlled transformation and diffusionless transformation. Physics-based modeling and prediction for the volume fractions of phase transformation in continuous heating under anisothermal conditions are developed based upon the addition of volume fractions in sequential segmented isothermal processes of grinding. The predictive model is validated by 18Ni (250) maraging steel grinding experiments, X-ray diffraction measurements and regression analyses. Results are compared to the model predicted of martensite and ferrite phase volume fractions after grinding. The physics-based model is experimentally validated as viable to predict the occurrence and extent of phase transformation related to material properties, wheel topography and grinding thermal-mechanical loading. Finally, correlation analysis is used to quantify the importance of the input variables to both model-predicted and X-ray diffraction measured phase transformation results.     

数控工艺优化设计系统的研发与应用

系统以特征技术为基础,将CAPP技术引入到数控编程中,在三维CAD系统UG中,通过API直接开发CAPP系统,实现了CAD系统与CAPP系统的内嵌式集成,以及CAD系统到CAPP系统的数据信息内容的自动传递和CAPP数控加工工艺的自动生成,提高了数控编程的效率和质量。     

Predictive Modeling of Surface Roughness in Grinding of Ceramics

The surface roughness represents the quality of ground surface since irregularities on the surface may form nucleation for cracks or corrosion and thus degrade the mechanical properties of the component. The surface generation mechanism in grinding of ceramic materials could behave as a mixture of plastic flow and brittle fracture, while the extent of the mixture hinges upon certain process parameters and material properties. The resulting surface profile can be distinctively different from these two mechanisms. In this article, a physics-based model is proposed to predict the surface roughness in grinding of ceramic materials considering the combined effect of brittle and ductile material removal. The random distribution of cutting edges is first described by a Rayleigh probability function. Afterwards, surface profile generated by brittle mode grinding is characterized via indentation mechanics approach. Last, the surface roughness is modeled through a probabilistic analysis of ductile and brittle generated surface profile. The model expresses the surface finish as a function of the wheel microstructure, the process conditions, and the material properties. The predictions are compared with experimental results from grinding of silicon carbide and silicon nitride workpieces (SiC and Si₃N₄, respectively) using a diamond wheel.     

基于CAN总线和变结构方法的AGV控制器设计

针对一般AGV控制器中非柔顺转弯、相互之间无通讯联系以及跟踪误差在一定范围内有增加趋势的缺点,在考虑非完整约束基础上分析了柔顺运动模式下的AGV转弯问题,得出其状态空间和控制方程.同时,设计了上位机为PC机,下位机以AT89S52单片机为主控器的CAN总线无线通讯系统.在最优参数下进行了AGV路径跟踪仿真和实验,证明了柔顺运动方式下CAN总线和变结构控制方法的有效性.     

微进给系统在磨床加工中的应用与研究

根据磨床精密加工的需要,为补偿精密磨床砂轮的振动对加工精度的影响,在原有数控磨床的基础上设计了微进给工作台和基于信号处理及特殊算法的闭环控制系统.该闭环系统应用于对压电陶瓷致动器的控制,并且在此基础上进行了磨削加工实验.实验数据结果表明,零件加工精度在1 μm左右,满足零件精密加工的要求.     

基于独立成分分析算法的纺纱锭子噪声测试

纺纱锭子性能的优劣对生产效率和纱线质量具有决定性的影响.以纺纱锭子运转中的噪声信号和驱动电动机单独运转的噪声信号为考察对象,对纺纱锭子运转过程进行分析.利用频谱分析和快速独立成分分析( FastICA)算法将纺纱锭子运转噪声信号的分离信号与驱动电动机单独运转的噪声信号主频率进行对比,38.88％分离主频率与驱动电机单独...     

基于典型工艺的NC自动编程工具

提出了基于典型工艺的NC自动编程方法.利用UG提供的基本功能及其二次开发工具,开发出一套NC自动编程工具,实现了CAPP与NC编程的一体化.使用户能够基于自定义的典型零件工艺或典型工艺特征,借鉴已有的工艺知识和相关工艺参数,快速有效地进行NC编程.