基于神经网络和遗传算法的锭子弹性管性能优化

为得到减振弹性管对下锭胆的支承弹性和锭子高速运动下的稳定性等性能的最优匹配效率,依据减振弹性管的等效抗弯刚度及底部等效刚度系数公式,利用MatLab数值分析软件构建弹性管抗弯刚度和底部挠度数学模型。首先,结合Isight优化软件基于径向基神经网络构建其近似模型,且使精度达到可接受水平,并以模型的关键结构参数弹性模量、螺距、槽宽、壁厚为设计变量,结合遗传算法对弹性管抗弯刚度和底部挠度进行多目标优化设计,得到Pareto最优解集和Pareto前沿图,确定出减振弹性管结构工艺参数的优化方案。通过对优化数据进行分析发现,该方案在保证减振弹性管弹性的同时,其底部振幅明显减弱。     

Investigations on grinding process of woven ceramic matrix composite based on reinforced fiber orientations

Fiber orientations play the decisive role in grinding process of woven ceramic matrix composites, but the influence of woven fibers in grinding process is not clear. This paper studies the surface quality and grinding force by comparing different woven surfaces. Through a series of experiments in optimized sampling conditions, we analyze characteristics of the material surface topography height, wave distribution and surface support properties in details. And we find some outstanding characteristics of the surface microstructure. We also study the influence of grinding processing parameters on surface microstructure. The results show that machining surface which contains more parallel fibers is rougher and more keenness than gauss surface. Grinding wheel speed and depth of cut have great influence on surface topography and surface support properties. And it is discovered that grinding forces are also highly dependent on fiber orientations. The mechanism of the grinding phenomena is also analyzed in this paper according to knowledge of fracture mechanics and mechanical damage phenomenology. The research obtained will be an important technical support on improving the processing quality of woven ceramic matrix composites.     

Classification and regression models of audio and vibration signals for machine state monitoring in precision machining systems

We present a data-driven method for monitoring machine status in manufacturing processes. Audio and vibration data from precision machining are used for inference in two operating scenarios: (a) variable machine health states (anomaly detection); and (b) settings of machine operation (state estimation). Audio and vibration signals are first processed through Fast Fourier Transform and Principal Component Analysis to extract transformed and informative features. These features are then used in the training of classification and regression models for machine state monitoring. Specifically, three classifiers (K-nearest neighbors, convolutional neural networks and support vector machines) and two regressors (support vector regression and neural network regression) were explored, in terms of their accuracy in machine state prediction. It is shown that the audio and vibration signals are sufficiently rich in information about the machine that 100% state classification accuracy could be accomplished. Data fusion was also explored, showing overall superior accuracy of data-driven regression models.     

The influence of plastic properties on chip formation

A two-dimensional finite-element model of the chip formation process is used to study the influence of the material law determining plastic flow on chip formation of titanium alloys at high cutting speeds. For simplicity, friction and thermo-mechanical properties of the tool are neglected in the simulations. Titanium chips are strongly segmented and therefore especially suitable to study the role of segmentation for the cutting force and other cutting parameters. Of special interest is the influence of the thermal softening parameter in the material law. Increasing thermal softening leads to a drastic decrease in the cutting force and a corresponding increase in chip segmentation. A variation of the hardening exponent is also performed. It is shown that the simulation results can be understood by using adiabatic flow stress curves. The variation of the hardening exponent leads to strongly differing chip shapes, although the cutting force stays nearly constant. This shows that mean cutting forces should not be used as simulation verification parameters.     

Investigation into polishing process of CVD diamond films

A new technique used for polishing chemical vapor deposition (CVD) diamond films has been investigated, by which rough polishing of the CVD diamond films can be achieved efficiently. A CVD diamond film is coated with a thin layer of electrically conductive material in advance, and then electro-discharge machining (EDM) is used to machine the coated surface. As a result, peaks on the surface of the diamond film are removed rapidly. During machining, graphitization of diamond enables the EDM process to continue. The single pulse discharge shows that the material of the coated layer evidently affects removal behavior of the CVD diamond films. Compared with the machining of ordinary metal materials, the process of EDM CVD diamond films possesses a quite different characteristic. The removal mechanism of the CVD diamond films is discussed.     

β21s高强度钛合金的铣削试验研究

针对一种新型的难加工钛合金材料β21s,进行了刀具磨损试验和铣削力的测量试验,对材料的切削加工性进行评价,并且给出优选结果,为实际生产提供参考。     

LASER ENHANCED ELECTROCHEMICAL MACHINING PROCESS

An experimental apparatus was designed and built for the hybrid process of laser enhanced electrochemical machining (ECM). Comparison experiments had been done between solely ECM and laser enhanced ECM. The machining performance of hybrid process was investigated and compared with solely ECM. The principle of the hybrid process was exposed by analyzing the microstructure of machining surface. The results show that the hybrid process has advantages in machining accuracy, surface quality, and machining efficiency.     

铝合金薄板尖边缘缺口试样的机械加工方法及缺口敏感性测试

介绍高强度铝合金薄板的尖边缘缺口拉伸试样的机械加工方法，用该方法制备的仿美７０７５、美制７０７５、国产ＬＣ４、ＬＣ９、ＬＹ１２、ＬＢ７３３等６种铝合金１６种热处理状态下的ＥＮ试样，其内率半径Ｒ满足了ＡＳＴＭＥ３３８－６８标准方法中Ｒｍａｘ０．０１８毫米的要求，并且达到Ｒ＜０．０１０毫米的特高精度。同时，还对仿美７０７５、美制７０７５、国产ＬＣ４的６种热处理状态下的缺口敏感性和宏观断口、以及选择仿美７０７５、ＬＹ１２对扫描断口形态进行了研究。