基于STL文件的柱状支撑结构自动生成算法

熔融挤压三维打印以热塑性材料为原料逐层打印完成,由于熔丝只能沉积在已存在的物体上层,模型悬空部位需要添加支撑结构。为解决该问题,提出一种基于STL(Stereo Lithography)文件的稀疏柱状支撑结构自动生成算法。算法通过对比三角面片倾斜角度与模型临界角的大小,获取独立的待支撑区域,然后基于边长自适应法和射线与面片相交法得到待支撑点集,自动生成支撑结构。此外为了保证模型表面质量和支撑结构的稳定性,支撑结构不在模型表面生成。调整支撑结构形状,进一步提高支撑结构的稳定性。通过实验证明本文算法比Cura软件节约15%的材料消耗,支撑结构更容易剥离,模型表面质量更好。     

基于体素模型的3D打印支撑算法

3D 打印技术是通过对材料的逐层堆积来构建物体,但对模型悬空的区域需要添加 支撑结构。支撑结构不仅会造成打印材料的浪费,而且会延长打印时间并对模型外表有所损伤。 为此,提出一种基于体素模型的支撑算法,针对体素化后的模型,分析体素之间的相互支撑作用, 并引入体素支撑能量函数概念和计算方法,计算出需要添加支撑的体素,从而得到需要添加支撑 结构的区域,并由该区域生成支撑结构,之后通过实验对算法进行验证。实验结果显示该算法能 够更加准确地对模型生成支撑,同时,基于体素模型的支撑算法对于模型内部支撑计算,也具有 很好的适用性。     

面向耗材节省的三维打印路径规划算法研究

为降低三维打印（three-dimensional printing,3D）耗材费用并进一步提高打印效率,给出一种面向熔融沉积制造的三维打印路径规划算法。该方法综合考虑打印耗材、打印效率以及打印表面质量等因素,通过网格模型及其支撑的相邻层片轮廓关系求得可稀疏打印区域;基于多边形扫描线算法以及多边形单调链关系,得到能够连续打印的路径区域;最终通过区域路径稀疏化得到改进的打印路径。通过复杂网格模型的三维打印路径规划实例,验证了算法的有效性。该算法能够降低打印耗材数量,并进一步提高打印效率。     

FDM快速成型支撑结构自动生成算法的研究

FDM系统中零件原形的加工是靠热喷头挤出熔丝逐层堆积而成的。因此,支撑结构的添加对零件原形的制作起着至关重要的作用。针对支撑结构的需求,采用了一种新的支撑自动生成算法。该算法基于STL模型首先需提取待支撑区域,然后添加相应的支撑结构,最后确定待支撑区域的起点和终点。实际应用表明,生成支撑的效果良好。     

基于深度剥离的三维打印模型朝向优化算法

为了提高三维打印产品表面质量且有效节省材料,提出一种面向熔融挤压式三维打印机的模型朝向优化算法.该算法基于深度剥离技术,以打印朝向为自变量构建模型悬空面积及其支撑结构体积的双目标优化函数;利用模式搜索方法设计一个两步走的优化方案来确定最优的打印朝向.实验结果表明,利用深度剥离技术能够较为准确、高效地检测出打印模型悬空区域,并计算出相应的悬空面积和支撑体积;设计的优化方案也能够根据双目标的权重比确定一个较优的打印朝向,不但减少了支撑结构与模型表面的接触范围,还能降低支撑结构的体积.文中算法在提高模型表面质量以及减少支撑结构的材料消耗两方面具有优势.     

NURBS体参数化模型的无支撑打印算法

处理大角度悬垂结构的打印问题是无支撑打印的主要挑战。针对NURBS（non-uniform rational Bspline）体参数化模型的多自由度无支撑三维打印技术,提出一种锥形切片算法。通过对模型的三维文件进行几何映射,利用水平切片算法对其进行切片,得到模型变形后的连续打印路径G-code;然后对得到的G-code进行逆映射,生成适用于六轴机械臂的多自由度连续打印路径。利用锥形切片算法规划打印路径,可以在无须支撑的情况下,依靠自支撑结构实现对悬空结构部分的打印。最后,通过仿真实验和对比实验,验证了锥形切片算法的准确性、可行性和有效性。锥形切片算法的引入为NURBS体参数化模型的无支撑三维打印提供了一种更加高效、经济和环保的解决方案。     

熔丝制造的三维连续编织填充图案

为提高熔丝制造(FFF)工件的机械强度,降低各向异性,提出一种3D连续编织路径规划方法.采用连续纤维增强丝材作为打印材料,设计了1个8层循环结构,利用3D打印机喷嘴挤出丝材生成经纬纱线,控制FFF平台在z方向的运动,生成类似3D编织的连续沉积路径,不同层的纤维相互交错和嵌入以实现相邻切片平面之间的互锁,以提高层内与层间...     

内部结构自支撑的三维打印浮体平衡优化算法

为了使浮体能够在熔融挤压式三维打印机中一次成型,提出一种内部结构自支撑的浮体平衡优化算法.首先在给定漂浮姿势的输入模型内部,利用递归细分方法生成自支撑的格子结构,并把输入模型与格子结构组合为初始浮体模型.然后以初始浮体模型中格子单元体积为变量构建平衡优化方程,根据方程解算结果调整初始浮体模型中每个格子单元的体积,从而生成最终的浮体模型.实验结果表明,文中算法生成的浮体模型可以使用熔融挤压式三维打印机直接打印,打印的实体不但能够按照预设的姿势漂浮在液体中,而且具有较好的抗压能力;算法中的递归细分策略能够在模型内部产生较多的格子单元,具有节省打印材料的优势.     

面向3D打印的自支撑连通性填充结构设计

针对大多数3D打印轻质化填充结构因不能够自支撑而导致模型无法打印成型,额外添加支撑结构又会造成模型内部残留打印材料无法排出的问题,提出一种面向3D打印的自支撑连通性填充结构设计方法.首先,基于每层打印切片构建填充结构;其次,通过调整参数来控制不同层正三角形连续地周期性变化,实现填充结构的自支撑性和连通性;最后,根据每层的几何信息生成可以直接用于打印的G-code文件.通过进行打印实物、打孔灌水和力学测试实验,实验结果表明,该方法可以打印具有不同自支撑角度且包含2个连通区域的填充结构,与5种典型填充结构相比,该填充结构力学性能良好.     

装配体模型优化的免装配打印方法EI北大核心CSCD

针对包含机构语义的装配体模型实施高质量的免装配打印(涉及支撑结构的清除、打印材料的消耗以及机构语义关联表面的打印质量等)不仅困难且仍缺乏自动方法的问题,提出一种装配体模型优化的免装配打印方法.首先基于包围盒剖分和体素关联关系对装配体模型进行体素化;然后采用改进的Bresenham算法快速生成体素路径,并据此快速分析支撑结构的可清除性;提出一种增量式的体素重关联方法,可在零件之间的相对布局改变后避免耗时的模型重体素化;最后以同时满足支撑结构可清除、支撑结构消耗材料量尽量少,以及运动接口面的打印质量尽量好为目标,通过改进粒子群算法确定和调整装配体模型的打印方向及其零件之间的相对布局.在带有典型机构语义的若干装配体模型上进行实验的结果表明,所提方法计算的方向和零件之间的相对布局打印模型的支撑结构均可完全清除,支撑结构所消耗的材料以及机构语义关联表面的粗糙度均大幅度减少(小);与基于八叉树分析支撑结构可清除性的方法以及基于零件布局调整开展免装配机构打印的最新方法相比,该方法对于包含机构语义的装配体模型开展高质量免装配打印具有突出的适用性和优势.     

Clever Support: Efficient Support Structure Generation for Digital Fabrication

We introduce an optimization framework for the reduction of support structures required by 3D printers based on Fused Deposition Modeling (FDM) technology. The printers need to connect overhangs with the lower parts of the object or the ground in order to print them. Since the support material needs to be printed first and discarded later, optimizing its volume can lead to material and printing time savings. We present a novel, geometry‐based approach that minimizes the support material while providing sufficient support. Using our approach, the input 3D model is first oriented into a position with minimal area that requires support. Then the points in this area that require support are detected. For these points the supporting structure is progressively built while attempting to minimize the overall length of the support structure. The resulting structure has a tree‐like shape that effectively supports the overhangs. We have tested our algorithm on the MakerBot® Replicator? 2 printer and we compared our solution to the embedded software solution in this printer and to Autodesk® Meshmixer? software. Our solution reduced printing time by an average of 29.4% (ranging from 13.9% to 49.5%) and the amount of material by 40.5% (ranging from 24.5% to 68.1%).     

一种基于分布式数据库的全局频繁项挖掘算法

针对FDM（Fast Distributed Association Rules Mining）可能会出现全局频繁项集丢失的问题,引入双射空间的思想和标识位,提出一种DS-CFDM（Double space-Fast Distributed association rules Mining based on Center）算法,算法实现通过一中心站点来协调信息的传输以此来减少网络通信的负担。     

A texture image denoising approach based on fractional developmental mathematics

The traditional integer-order computation-based denoising approaches often blur the edges and textural details of an image. To solve this problem, from the viewpoint of system evolution, and based on the features of fractional calculus, we propose to implement a texture image denoising approach based on fractional developmental mathematics (FDM) which applies a novel mathematical method, fractional calculus, to image denoising. First, we synopsize the necessary theoretical background of fractional calculus. Second, we derive the necessary mathematical models for implementation of a texture image denoising approach based on FDM. We derive fractional Green’s formula, fractional Gauss’ formula, and fractional Stokes’ formula, and fractional Euler–Lagrange equation. Then, a texture image denoising approach based on FDM is proposed. Third, we implement comparative experiments. We firstly derive the numerical implementation of FDM. Then, we study the capability of preserving the edges and textural details of FDM by comparative experiments. The comparative experimental results show that the capability of preserving the edges and textural details of the FDM-based denoising algorithm is obviously superior to that of traditional integer-order computation-based algorithms, especially for texture detail rich images.     

模糊数据挖掘

本文在数据库中知识发现（KDD）和数据挖掘（DM）技术的基础上,提出了模糊数据库中知识发现（KDFD）和模糊数据挖掘（FDM）的概念与技术,并给出FDM的算法,它能有效地挖掘出模糊数据库中潜在的有价值的知识。本文具体讨论了模糊关联规则及模糊数据依赖的挖掘。     

模糊数据挖掘技术(FDM)在交通量预测中的应用

根据影响交通量的因素,利用模糊数据挖掘方法(FDM)对交通量进行预测,这种预测方法不需要精确的数学模型,算法过程易于理解,解决了仅靠复杂的算法和推理并不能完全发现知识的问题,且还可编制一个程序,使用更方便。     

Clean color: Improving multi‐filament 3D prints

Fused Filament Fabrication is an additive manufacturing process by which a 3D object is created from plastic filament. The filament is pushed through a hot nozzle where it melts. The nozzle deposits plastic layer after layer to create the final object. This process has been popularized by the RepRap community. Several printers feature multiple extruders, allowing objects to be formed from multiple materials or colors. The extruders are mounted side by side on the printer carriage. However, the print quality suffers when objects with color patterns are printed – a disappointment for designers interested in 3D printing their colored digital models. The most severe issue is the oozing of plastic from the idle extruders: Plastics of different colors bleed onto each other giving the surface a smudged aspect, excess strings oozing from the extruder deposit on the surface, and holes appear due to this missing plastic. Fixing this issue is difficult: increasing the printing speed reduces oozing but also degrades surface quality – on large prints the required speed level become impractical. Adding a physical mechanism increases cost and print time as extruders travel to a cleaning station. Instead, we rely on software and exploit degrees of freedom of the printing process. We introduce three techniques that complement each other in improving the print quality significantly. We first reduce the impact of oozing plastic by choosing a better azimuth angle for the printed part. We build a disposable rampart in close proximity of the part, giving the extruders the opportunity to wipe oozing strings and refill with hot plastic. We finally introduce a toolpath planner avoiding and hiding most of the defects due to oozing, and seamlessly integrating the rampart. We demonstrate our technique on several challenging multiple color prints, and show that our tool path planner improves the surface finish of single color prints as well.     

Effect of process parameters on micro hardness of Al–Al2O3 composite prepared using an alternative reinforced pattern in fused deposition modelling assisted investment casting

Aluminium matrix composites (AMCs) have numerous applications in manufacturing sector. In the present research work, fused deposition modelling (FDM) assisted investment casting (IC) route has been explored for the development of Al–Al₂O₃ based functionally graded material. Initially, two alternative FDM filaments consisting of nylon-6, Al and Al₂O₃ (in different proportions) were fabricated on single screw extruder. These filaments were further used to prepare reinforced FDM patterns (which were barrel finished) for IC process. Taguchi L18 orthogonal array has been used to study the effect of processing parameters (like: filament proportion, volume of reinforced FDM pattern, density of FDM pattern, barrel finishing time, barrel finishing media weight and numbers of IC slurry layer) on micro hardness (M_H) of Al–Al₂O₃ composites. Analysis of variance has been used to study the percentage contribution of input parameters and their significance effect on M_H. The results of the study highlighted that density of FDM pattern and numbers of IC slurry layer have contributed around 80% for M_H. Micro-structural analysis carried out on castings has justified the M_H data.     

Development of ‘Multi-arm Bioprinter’ for hybrid biofabrication of tissue engineering constructs

This paper highlights the development of ‘Multi-arm Bioprinter (MABP)’ capable of concurrent multi-material deposition with independent motion path and dispensing parameters including deposition speed, material dispensing rate, and nozzle travel velocity for use in tissue engineering. In this research, the system is designed to concurrently print a filament structure and deposit cell spheroids between the filaments to create a hybrid structure to support the cell spheroids in three dimensions (3Ds). This process can be used with multiple cell types and is capable of reducing the fabrication time while using optimized dispensing parameters for each material. A novel method of dispensing the crosslinking solution using a co-axial nozzle was also developed and demonstrated in this paper. Cell-laden structures were fabricated through concurrent deposition of cell-encapsulated filaments and with cell spheroids to validate this concept. Rheology studies were then conducted to determine the effects of crosslink flow on filament width, hydrogel dispensing pressure on filament width, and dispensing time interval on spheroid diameter.