Chi Zhou

Digital material fabrication using mask‐image‐projection‐based stereolithography

Purpose – The purpose of this paper is to present a mask‐image‐projection‐based stereolithography (MIP‐SL) process that can combine two base materials with various concentrations and structures to produce a solid object with desired material characteristics. Stereolithography is an additive manufacturing process in which liquid photopolymer resin is cross‐linked and converted to solid. The fabrication of digital material requires frequent resin changes during the building process. The process presented in this paper attempts to address the related challenges in achieving such fabrication capability.Design/methodology/approach – A two‐channel system design is presented for the multi‐material MIP‐SL process. In such a design, a coated thick film and linear motions in two axes are used to reduce the separation force of a cured layer. The material cleaning approach to thoroughly remove resin residue on built surfaces is presented for the developed process. Based on a developed testbed, experimental studies we...

3D Printing of Graphene Aerogels

3D printing of a graphene aerogel with true 3D overhang structures is highlighted. The aerogel is fabricated by combining drop-on-demand 3D printing and freeze casting. The water-based GO ink is ejected and freeze-cast into designed 3D structures. The lightweight (<10 mg cm(-3) ) 3D printed graphene aerogel presents superelastic and high electrical conduction.

A Fast Mask Projection Stereolithography Process for Fabricating Digital Models in Minutes

The purpose of this paper is to present a direct digital manufacturing (DDM) process that is an order of magnitude faster than other DDM processes that are currently available. The developed process is based on a mask-image-projection-based stereolithography (MIP-SL) process, in which a digital micromirror device (DMD) controls projection light to selectively cure liquid photopolymer resin. In order to achieve high-speed fabrication, we investigate the bottom-up projection system in the MIP-SL process. A two-way linear motion approach has been developed for the quick spreading of liquid resin into uniform thin layers. The system design and related settings for achieving a fabrication speed of a few seconds per layer are presented. Additionally, the hardware, software, and material setups for fabricating three-dimensional (3D) digital models are presented. Experimental studies using the developed testbed have been performed to verify the effectiveness and efficiency of the presented fast MIP-SL process. The test results illustrate that the newly developed process can build a moderately sized part within minutes instead of hours that are typically required.

A layerless additive manufacturing process based on CNC accumulation

Purpose – Most current additive manufacturing (AM) processes are layer based. By converting a three‐dimensional model into two‐dimensional layers, the process planning can be dramatically simplified. However, there are also drawbacks associated with such an approach such as inconsistent material properties and difficulty in embedding existing components. The purpose of this paper is to present a novel AM process that is non‐layer based and demonstrate its unique capability.Design/methodology/approach – An AM process named computer numerically controlled (CNC) accumulation has been developed. In such a layerless AM process, a fiber optic‐cable connected with an ultraviolet (UV) LED and related lens is served as an accumulation tool. The cable is then merged inside a tank that is filled with UV‐curable liquid resin. By controlling the on/off state of the UV‐LED and the multi‐axis motion of the cable, a physical model can be built by selectively curing liquid resin into solid.Findings – It is found that the ...

3D stereolithography printing of graphene oxide reinforced complex architectures.

Properties of polymer based nanocomposites reply on distribution, concentration, geometry and property of nanofillers in polymer matrix. Increasing the concentration of carbon based nanomaterials, such as CNTs, in polymer matrix often results in stronger but more brittle material. Here, we demonstrated the first three-dimensional (3D) printed graphene oxide complex structures by stereolithography with good combination of strength and ductility. With only 0.2% GOs, the tensile strength is increased by 62.2% and elongation increased by 12.8%. Transmission electron microscope results show that the GOs were randomly aligned in the cross section of polymer. We investigated the strengthening mechanism of the 3D printed structure in terms of tensile strength and Youngs modulus. It is found that an increase in ductility of the 3D printed nanocomposites is related to increase in crystallinity of GOs reinforced polymer. Compression test of 3D GOs structure reveals the metal-like failure model of GOs nanocomposites.

Credit scoring model based on neural network with particle swarm optimization

Credit scoring has gained more and more attentions both in academic world and the business community today. Many modeling techniques have been developed to tackle the credit scoring tasks. This paper presents a Structure-tuning Particle Swarm Optimization (SPSO) approach for training feed-forward neural networks (NNs). The algorithm is successfully applied to a real credit problem. By simultaneously tuning the structure and connection weights of NNs, the proposed algorithm generates optimized NNs with problem-matched information processing capacity and it also eliminates some ill effects introduced by redundant input features and the corresponding redundant structure. Compared with BP and GA, SPSO can improve the pattern classification accuracy of NNs while speeding up the convergence of training process.

Multitool and Multi-Axis Computer Numerically Controlled Accumulation for Fabricating Conformal Features on Curved Surfaces

In engineering systems, features such as textures or patterns on curved surfaces are common. In addition, such features, in many cases, are required to have shapes that are conformal to the underlying surfaces. To address the fabrication challenge in building such conformal features on curved surfaces, a newly developed additive manufacturing (AM) process named computer numerically controlled (CNC) accumulation is investigated by integrating multiple tools and multiple axis motions. Based on a fiber optical cable and a light source, a CNC accumulation tool can have multi-axis motion, which is beneficial in building conformal features on curved surfaces. To address high resolution requirement, the use of multiple accumulation tools with different curing sizes, powers, and shapes is explored. The tool path planning methods for given cylindrical and spherical surfaces are discussed. Multiple test cases have been performed based on a developed prototype system. The experimental results illustrate the capability of the newly developed AM process and its potential use in fabricating conformal features on given curved surfaces. [DOI: 10.1115/1.4026898]

Separation force analysis and prediction based on cohesive element model for constrained-surface Stereolithography processes

Constrained-surface based Additive Manufacturing (AM) processes have been widely used in both academia and industry for the past few years. Despite the advantages of constrained-surface based AM processes, it has not been widely used in practice. A main reason for this is that a substantial separation force is required to separate the cured part from the material vat during the pulling-up stage, which may damage the cured part and reduce the reliability of the process. The solutions proposed previously to reduce this separation force recommend using an intermediate coating material (e.g., Teflon and silicone films) between the cured part and the vat. This, however, has only negligible effects in reducing the separation force. In this work, the pulling-up process is modeled within the framework of mechanics-based principles. In particular, the cohesive zone model (CZM) is adopted to characterize the separation mechanism, and finite element (FE) simulation is carried out to investigate the separation process using the commercially available FE software, Abaqus. A new simple optimization scheme is also proposed to estimate the constitutive cohesive stiffness parameters from experimental measurements. These constitutive parameters are very difficult to estimate using the standard mechanical tests. The proposed work based on sound mechanics-based principles can be used for reliable prediction of pulling-up speed, and thus, is likely to be useful in devising an adaptive closed-loop system to control the pulling-up process and achieve a reliable AM approach. We formulate cohesive zone model to characterize the separation process.We establish an optimization model to evaluate the mechanical parameters.The effectiveness of the proposed technique is validated by computer-simulated experiments.Fabrication performance can be significantly improved by the proposed approach.

Pattern Classification and Prediction of Water Quality by Neural Network with Particle Swarm Optimization

Water pollution has posed a severe problem in modern society. Evaluation of water quality is a meaningful topic today. To identify the specific water category and predict the water quality in the future, a particle swarm optimization (PSO) based artificial neural network (ANN) approach is presented. The data investigated from the Yangtze River are chosen as the original cases to construct the ANN model and testify both the classification and prediction ability of this method. Compared with other classical methods, the proposed one can obtain high quality and efficiency without losing computational expense. Experimental results show PSO is a robust training algorithm and could be extended to other real world pattern classification and prediction applications

A Novel Low-Cost Stereolithography Process Based on Vector Scanning and Mask Projection for High-Accuracy, High-Speed, High-Throughput, and Large-Area Fabrication

Photopolymerization based process is one of the most popular additive manufacturing processes. Two primary configurations for this process are laser based vector by vector scanning (0D) and projection based layer by layer exposing (2D). With the highly focused fine laser, the scanning based process can accomplish very high surface finishing and precision, however, due to the serial nature of scanning, this process suffers from the problem of slow speed. In contrast with laser scanning, projection based process can form the whole layer in one exposure, which leads to higher fabrication efficiency. However, due to the limited resolution of projection device and various optical defects, the surface quality will be significantly deteriorated for large area fabrication. To solve this problem, a novel hybrid process by integrating vector scanning and mask projection has been presented. In this process, laser is focused into a fine spot and used to scan the boundary of the layer, whereas the projector is focused onto a large platform surface and used to form the interior area of the layer. An efficient slicing method is proposed for extracting the contour for laser scanning. A slice to image conversion algorithm is also developed to convert the offset contour to grayscale image for mask projection. Experimental results have verified that the proposed hybrid process can significantly improve the fabrication speed without losing the surface quality.Copyright