基于三角面-三角形相交检测的五轴数控加工碰撞干涉检测算法研究

碰撞干涉检测速度是五轴数控加工的关键难题之一。为了提高碰撞干涉检测速度,提出一种新的基于三角面片的碰撞干涉检查算法。使用分层OBB和八叉树相结合的算法快速检测出被检测面上可能存在干涉的碰撞叶子节点,将碰撞叶子节点用三角面片表示出来;在包含碰撞干涉叶子节点的三角面片和OBB的三角面片之间,通过三角面片的矢量判别法实现工件表面和加工轴之间的碰撞干涉检测。该算法克服了传统算法速度慢、效率低等问题,提高了碰撞干涉检测的准确性和精度。     

一个粗加工用的八叉树的方法

本文介绍一种数控加工中心刀具路径自动生成的八叉树分割基本算法。被加工的零件及其坯料都是由实体模型描述的。方法的大意是在满足刀具强度和零件几何形状的条件下分层进行加工。这种方法先将一个三维问题化为一个二维刀具路径生成问题,而后再采用平面上的一个四叉树表示方法来求解。这种方法还可很容易地扩展到需要处理的夹具及实体模型是由其它方法表示的场合。     

数控机床实时碰撞检测算法的研究与实现

碰撞检测是数控机床加工过程中的一个重要功能模块.碰撞检测可以有效地避免刀具与部件发生碰撞,同时可以提高加工的安全性.在研究了大量碰撞检测算法的基础上,提出了一种实时性较好的碰撞检测算法,从而避免了仿真软件检测缺陷.此算法首先在预处理阶段采用轴向包围盒AABB对部件和刀具进行保护区的建立,对碰撞检测进行初步检测.然后采用分离轴理论和长方体相交理论对保护区进行实时碰撞检测.最后对提出的算法进行了仿真验证,验证结果表明该算法可实时完成对碰撞体保护区域的检测.     

非球面刀具数控加工模具型腔的无干涉路径规划

在球面刀具干涉处理算法的基础上 ,对模具型腔自由曲面NC三轴加工中平头刀的刀具轨迹的干涉处理技术进行了分析。将干涉处理从三维空间问题分解为2个二维空间问题 ,并提出相应的干涉处理方法。该算法具有干涉处理彻底、处理效率高等优点     

虚拟装配环境中的一种碰撞干涉检测方法

提出了在虚拟装配过程中零件与装配体碰撞干涉检测的一种方法。用投影轴算法检测出装配对象的碰撞,用坐标跟踪检测方法确定干涉余量值,并用连续包围盒对复杂和不规则的模型进行分割,通过这三种方法的配合使用,虚拟装配系统对干涉体进行了有效的、实时的、精确的检测。     

快速成型中散乱数据点云的自适应分层算法研究

提出了一种快速成型中散乱数据点云的自适应分层算法,该算法根据分层厚度阈值对散乱数据点云进行空间栅格划分,采用八叉树组织其空间拓扑结构,基于该结构获取层面邻域数据,通过其曲率分布特征自适应调整分层厚度,根据层面邻域数据与分层平面的交点获取分层数据.实验证明,该算法数据适应性强,分层数据获取准确性高,可快速有效地实现散乱数据点云的自适应分层处理.     

基于八叉树的铸造前处理网格剖分算法研究

基于八叉树数据结构,开发了一种自适应网格剖分算法。该算法以STL文件模型为根节点进行空间分割,通过递归计算并设定剖分停止临界条件产生非平衡树,从而能对复杂模型的复杂形状部位进行自适应的深度剖分。开发出的相应程序对几何模型文件进行计算和数据分析,最终能够产生质量较高的八叉树非均匀网格。     

基于刀具姿态控制的指令点插补算法

为克服五轴数控加工中忽略刀具姿态控制所带来的旋转轴速度和加速度不连续等问题,提出一种基于刀具姿态控制的五轴数控加工指令点插补算法。该算法根据给定的数控加工程序确定刀尖点位置集合和刀轴矢量集合,分别采用样条曲线对离散刀尖点和刀轴矢量进行拟合,得到刀尖点曲线和刀具姿态曲线,在此基础上进行指令点插补。实验结果表明,该算法能够将离散刀轴矢量拟合成位于单位球面的二阶连续曲线,保证旋转轴速度和加速度的连续性。     

数控加工仿真中自由曲面的碰撞干涉检验

碰撞干涉检验是数控加工仿真的主要目的之一.文章解决了数控加工仿真中的自由曲面碰撞干涉的问题:把碰撞干涉分为四种情况,对其分别提出不同的检验方法,并阐述了具体的解决方案.这些方法对3-5轴数控机床均适用.其中提出的刀具与曲面全局干涉的检查算法解决了求曲面到刀具最小距离的问题,即根据曲面到刀轴最小距离判断刀具是否与曲面发生干涉.     

模具零件的最小包围盒生成算法

提出一种实用的最小包围盒算法,将任意位置的复杂三维模型投影在3个主平面上,通过分析投影外轮廓的最小包围矩形来最终确定三维模型的最小包围盒。算法不再采用传统算法中旋转模型的方法,而根据旋转投影的外轮廓来确定最小包围盒,提高了算法的效率。同时算法通过分析识别模型的基本形状,将其分为箱体类、回转类和异形类,以快速确定一个主投影方向,将三维问题转化为求二维平面上的最小包围矩形,进一步提高了计算效率。算法可以嵌入NX等商用CAD系统,经过大量测试对比,相对传统算法更高效,并且稳定可靠,可广泛应用于任意位置的复杂模具零件的生成和干涉检查等。     

五轴联动刀轴矢量插补优化算法

目的研究解决五轴联动机床旋转轴角度采用线性插补方式生成的加工轨迹,导致刀具姿态偏离所设计的理想平面,引发刀具姿态误差的问题,减少非线性误差,提高零件表面质量。方法首先对旋转轴角度线性插补方式引发刀具姿态误差的原理进行了分析,提出了一种刀轴矢量插补优化算法。然后在线性插补的基础上,根据提出的刀轴矢量插补优化算法保证首末点间的刀轴插补矢量始终位于首末刀轴矢量所构成的平面内,实现刀具姿态优化,并在MATLAB中对线性插补和矢量插补优化两种方式进行仿真分析,观测出对应方式下刀轴插补矢量的空间位置。最后利用叶片试件在AB型转台摆头类型机床上进行仿真和加工验证,对比两种刀轴矢量插补方式仿真数据。结果在VERICUT同等条件下仿真,刀轴矢量采用线性插补时,叶片进出汽边误差值分别为-0.218 66 mm和-0.312 58 mm;刀轴矢量插补优化后,叶片进出汽边误差值分别为-0.095 46 mm和-0.099 05 mm。刀具姿态经过插补优化算法后,叶片进出汽边的过切值明显降低。结论刀具姿态经过插补优化算法后,叶片过切值的大小和数目明显减少,使得非线性误差明显降低,从而提高了零件表面质量。     

The shape characteristic detection of tool breakage in milling operations

Detection of tool failure is very important in automated manufacturing. All previously developed tool breakage detection approaches in milling operations have adopted the strategy of parameter detection in which the detection of tool breakage was carried out according to values of specific parameters selected to reflect tool state (with or without tool breakage). In this paper the new concept of shape characteristic detection of tool breakage in milling operations is proposed. The detection of tool breakage is conducted according to the shape characteristics of discrete dyadic wavelet decomposition of cutting force. By means of the proposed method, the influence caused by the variation of cutting parameters and transients is eliminated. The proposed method is conducted in two steps. In the first step, cutting force signals are decomposed by discrete dyadic wavelet, with the shape characteristic vectors then being generated by the proposed shape characteristic vector-generating algorithm. In the second step, the shape characteristic vectors are fast classified by the ART2 neural networks. The accuracy and effectiveness of the proposed method are verified by numerous experiments.     

机器人仿真系统碰撞检测技术研究

为了保证人机安全,需要对机器人作业轨迹进行避碰检查。针对机器人仿真系统中的碰撞检测问题,研究一种基于八叉树结构的层级式三角形面片碰撞检测模型,以包围盒干涉检查作为粗检测,三角形面片相交检测作为精检测。为提高三角形相交检测效率,提出一种基于截距式方程的三角形干涉算法。将三维空间中的三角形映射到二维平面,避免了传统三角形相交检测中标量判别法与矢量判别法的复杂计算。最后在机器人仿真系统中验证了该碰撞检测算法的有效性。     

自由曲面参数线加工算法

从工程实际应用出发,提出了一种适合裁剪NURBS曲面加工基于参数线的刀轨生成算法。在曲面的参数线上规划刀位规划,将曲面用三角片离散逼近,利用刀具曲面与三角的几何相关性进行了干涉检查及刀位修正。实践表明,算法速度快、稳定性好。     

自由曲面数控加工智能选刀方法的研究与开发

由于自由曲面的几何复杂性,其数控加工的刀具干涉判别和刀具选取往往依靠数控机床编程工程师的经脸,致使加工的安全性、精度、效率及可靠性得不到有效保证.采用改进的遗传算法获取自由曲面的点云数据,在自主开发的系统中重构自由曲面.研究自由曲面加工刀位面生成的几何机理,根据刀具类型和几何尺寸,自动生成自由曲面数控加工的刀位面,可视化判别是否发生刀具干涉.将改进的快速排序算法应用到智能选刀模块,通过对常用类型和尺寸的刀具进行干涉排查,在避免干涉的前提下,智能选取最优化的加工刀具.对三种加工方案进行了测试,验证了智能选刀方法的正确性.     

The analysis of correlative error in principal axis method for five-axis machining of sculptured surfaces

Correlative error is a kind of error between the cutter locations. This text presents a detailed analysis about correlative error in the principal axis method (PAM) that is a strategy of tool positioning for five-axis machining. It points out all the strategies based on linear projection cannot avoid this interference and cannot be disregarded especially in the strategies with large stepover. Accurate calculating is made individually on linear and curve feed and formulae are developed. The author also gave the available application range of PAM and the algorithm for estimating correlative error.     

Sensorless tool failure monitoring system for drilling machines

It is well known that on-line tool condition monitoring has great significance in modern manufacturing processes. In order to prevent possible damages to the workpiece or the machine tool, reliable techniques are required providing an on-line response to an unexpected tool failure. Drilling is one of the most fundamental machining operations and two of the most crucial issues related to it are tool wear and fracture. During the spindle process, the motor driver current is related to the drill condition: power consumption is higher for a worn drill in comparison to a sharp drill for the same process. This difference in power consumption can be self-correlated to obtain the resulting waveform variance to provide a merit figure for tool condition. This paper describes a driver current signal analysis to estimate the tool condition by using the discrete Wavelet Transform in order to extract the information from the original cutting force, and through an autocorrelation algorithm evaluate the tool wear in the form of an asymmetry weighting function. The current is monitored from the motor driver to give a sensorless approach. Experimental results are presented to show the algorithm performance, a complete sensorless tool failure system which allows the detection of tool failure as a function of spindle current in real time.     

Discrete wavelet transform for tool breakage monitoring

Detection of tool breakage is of vital importance in automated manufacturing. Various methods have been attempted, and it is considered that the use of discrete wavelet transform (DWT), which is much more efficient and just as accurate wavelet analysis, may provide a realistic solution to the detection of tool breakage in operation. The DWT uses an analyzing wavelet function which is localized in both time and frequency to detect a small change in the input signals. In addition, it requires less computation than Fast Fourier Transformation (FFT). This paper discusses a tool breakage monitoring system based on DWT of an acoustic emission (AE) and an electric feed current signal using an effective algorithm. The experiment results show overall 98.5% reliability and the good real-time monitoring capability of the proposed methodology for detecting tool breakage during drilling.