Due to the stratified construction of additive manufacturing (AM), the building orientation of an object greatly affects the manufacturing process and quality of the products. This article proposes an optimization algorithm for AM that comprehensively covers the three main aspects affected by the building direction: efficiency, surface quality, and internal properties. The goal of optimizing efficiency is to cut the manufacturing and postprocessing times by reducing the support volume. In terms of surface quality optimization, a new and comprehensive mathematical model is established. Considering not only the influence of the stepping effect on the whole model but also the supporting contact area, the optimization to the salient area was innovatively proposed. The salient area was defined by cone curvature, which has a significant influence on appearance. Another novel point is the quantitative optimization of the internal properties, which is based on the anisotropic characteristics of AM. The mechanical properties are taken into consideration in this article, and other anisotropic properties can be added to the optimization algorithm by using the same method. Afterward, a particle swarm optimization algorithm was adopted to synchronously optimize the targets just cited, according to the degree of importance of each factor for specific applications. The algorithm was implemented to optimize several different cases with different characteristics. Further, the results in the experiments manifested that the models printed in the optimized orientation performed better than the initial models, and the corresponding comprehensive evaluation scores were improved as well, with an optimization range of 70–90%. And the rate of optimization for support volume, surface roughness, salient area roughness, and maximum tensile strength, respectively, reached 20.9%, 57.3%, 59.5%, and 293.0%.

中文翻译：

---

基于多目标优化的增材制造建筑方位确定

由于增材制造（AM）的分层构造，物体的构建方向极大地影响了产品的制造过程和质量。本文提出了一种增材制造优化算法，全面涵盖受建筑方向影响的三个主要方面：效率、表面质量和内部特性。优化效率的目标是通过减少支撑体积来缩短制造和后处理时间。在表面质量优化方面，建立了一个新的综合数学模型。不仅考虑了台阶效应对整个模型的影响，还考虑了支撑接触面积，创新性地提出了对显着区域的优化。显着区域由圆锥曲率定义，这对外观有很大的影响。另一个新颖点是内部属性的定量优化，这是基于 AM 的各向异性特性。本文考虑了力学特性，可以使用相同的方法将其他各向异性特性添加到优化算法中。然后，根据每个因素对特定应用的重要程度，采用粒子群优化算法同步优化刚才提到的目标。实施该算法以优化具有不同特征的几种不同情况。此外，实验结果表明，以优化方向打印的模型比初始模型表现更好，相应的综合评价分数也有所提高，优化幅度为70-90%。支撑体积、表面粗糙度、显着区域粗糙度和最大抗拉强度优化率分别达到20.9%、57.3%、59.5%和293.0%。