Data fusion analysis in the powder-bed fusion AM process monitoring by Dempster-Shafer evidence theory

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Yingjie Zhang ◽  
Wentao Yan ◽  
Geok Soon Hong ◽  
Jerry Fuh Hsi Fuh ◽  
Di Wang ◽  
...  
Keyword(s):  
Data Fusion ◽  
Process Monitoring ◽  
Feature Fusion ◽  
Evidence Theory ◽  
Process Condition ◽  
Monitoring Systems ◽  
Fusion Method ◽  
Powder Bed Fusion ◽  
Content Type ◽  
Powder Bed

Purpose This study aims to develop a data fusion method for powder-bed fusion (PBF) process monitoring based on process image information. The data fusion method can help improve process condition identification performance, which can provide guidance for further PBF process monitoring and control system development. Design/methodology/approach Design of reliable process monitoring systems is an essential approach to solve PBF built quality. A data fusion framework based on support vector machine (SVM), convolutional neural network (CNN) and Dempster-Shafer (D-S) evidence theory are proposed in the study. The process images which include the information of melt pool, plume and spatters were acquired by a high-speed camera. The features were extracted based on an appropriate image processing method. The three feature vectors corresponding to the three objects, respectively, were used as the inputs of SVM classifiers for process condition identification. Moreover, raw images were also used as the input of a CNN classifier for process condition identification. Then, the information fusion of the three SVM classifiers and the CNN classifier by an improved D-S evidence theory was studied. Findings The results demonstrate that the sensitivity of information sources is different for different condition identification. The feature fusion based on D-S evidence theory can improve the classification performance, with feature fusion and classifier fusion, the accuracy of condition identification is improved more than 20%. Originality/value An improved D-S evidence theory is proposed for PBF process data fusion monitoring, which is promising for the development of reliable PBF process monitoring systems.

scholarly journals Thermographic measurements of the commercial laser powder bed fusion process at NIST

2016 ◽  
Vol 22 (5) ◽  
pp. 778-787 ◽  
Author(s):  
Brandon Lane ◽  
Shawn Moylan ◽  
Eric P. Whitenton ◽  
Li Ma
Keyword(s):  
Process Monitoring ◽  
High Speed ◽  
Frame Rate ◽  
High Magnification ◽  
Powder Bed Fusion ◽  
Laser Powder Bed Fusion ◽  
Content Type ◽  
Powder Bed ◽  
Temperature Range ◽  
Melt Pool

Purpose Quantitative understanding of the temperatures, gradients and heating/cooling rates in and around the melt pool in laser powder bed fusion (L-PBF) is essential for simulation, monitoring and controls development. The research presented here aims to detail experiment design and preliminary results of high speed, high magnification, in-situ thermographic monitoring setup on a commercial L-PBF system designed to capture temperatures and dynamic process phenomena. Design/methodology/approach A custom door with angled viewport was designed for a commercial L-PBF system which allows close access of an infrared camera. Preliminary finite element simulations provided size, speed and scale requirements to design camera and optics setup to capture melt pool region temperatures at high magnification and frame rate speed. A custom thermal calibration allowed maximum measurable temperature range of 500°C to 1,025°C. Raw thermographic image data were converted to temperature assuming an emissivity of 0.5. Quantitative temperature results are provided with qualitative observations with discussion regarding the inherent challenges to future thermographic measurements and process monitoring. Findings Isotherms around the melt pool change in size depending on the relative location of the laser spot with respect to the stripe edges. Locations near the edges of a stripe are cooled to lower temperatures than the center of a stripe. Temperature gradients are highly localized because of rough or powdery surface. At a specific location, temperatures rise from below the measurable temperature range to above (<550°C to >1100°C) within two frames (<1.11 m/s). Particle ejection is a notable phenomenon with measured ejection speeds >11.7 m/s. Originality/value Several works are detailed in the Introduction of this paper that detail high-speed visible imaging (not thermal imaging) of custom or commercial LBPF processes, and lower-speed thermographic measurements for defect detection. However, no work could be found that provides calibrated, high-speed temperature data from a melt-pool monitoring configuration on a commercial L-PBF system. In addition, the paper elucidates several sources of measurement uncertainty (e.g. calibration, emissivity and time and spatial resolution), describes inherent measurement challenges based on observations of the thermal images and discusses on the implications to model validation and process monitoring and control.

scholarly journals Electronic Systems Diagnosis Fault in Gasoline Engines Based on Multi-Information Fusion

Sensors ◽  
2018 ◽  
Vol 18 (9) ◽  
pp. 2917 ◽  
Author(s):  
Jie Hu ◽  
Tengfei Huang ◽  
Jiaopeng Zhou ◽  
Jiawei Zeng
Keyword(s):  
Data Fusion ◽  
Information Fusion ◽  
Feature Fusion ◽  
Rapid Development ◽  
Back Propagation ◽  
Evidence Theory ◽  
Decision Fusion ◽  
Support Vector ◽  
Diagnostic Model ◽  
Fusion Method

The rapid development of electronic techniques in automobile has led to an increase of potential safety hazards, thus, a strong on-board diagnostic (OBD) system is desperately needed. To solve the problem of OBD insensitivity to manufacture errors or aging faults, the paper proposes a novel multi information fusion method. The diagnostic model is composed of a data fusion layer, feature fusion layer, and decision fusion layer. They are based on the back propagation (BP) neural network, support vector machine (SVM), and evidence theory, respectively. Algorithms are mainly focused on the reliability allocation of diagnostic results, which come from the data fusion layer and feature fusion layer. A fault simulator system was developed to simulate bias and drift faults of the intake pressure sensor. The real vehicle experiment was carried out to acquire data that are used to verify the availability of the method. Diagnostic results show that the multi-information fusion method improves diagnostic accuracy and reliability effectively. The study will be a promising approach for the diagnosis bias and drift fault of sensors in electronic control systems.

A printability assessment framework for fabricating low variability nickel-niobium parts using laser powder bed fusion additive manufacturing

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Bing Zhang ◽  
Raiyan Seede ◽  
Austin Whitt ◽  
David Shoukr ◽  
Xueqin Huang ◽  
...  
Keyword(s):  
Mechanical Properties ◽  
Additive Manufacturing ◽  
Uncertainty Quantification ◽  
Process Optimization ◽  
Powder Bed Fusion ◽  
Assessment Framework ◽  
Laser Powder Bed Fusion ◽  
Content Type ◽  
Powder Bed ◽  
Analytical Thermal Model

Purpose There is recent emphasis on designing new materials and alloys specifically for metal additive manufacturing (AM) processes, in contrast to AM of existing alloys that were developed for other traditional manufacturing methods involving considerably different physics. Process optimization to determine processing recipes for newly developed materials is expensive and time-consuming. The purpose of the current work is to use a systematic printability assessment framework developed by the co-authors to determine windows of processing parameters to print defect-free parts from a binary nickel-niobium alloy (NiNb5) using laser powder bed fusion (LPBF) metal AM. Design/methodology/approach The printability assessment framework integrates analytical thermal modeling, uncertainty quantification and experimental characterization to determine processing windows for NiNb5 in an accelerated fashion. Test coupons and mechanical test samples were fabricated on a ProX 200 commercial LPBF system. A series of density, microstructure and mechanical property characterization was conducted to validate the proposed framework. Findings Near fully-dense parts with more than 99% density were successfully printed using the proposed framework. Furthermore, the mechanical properties of as-printed parts showed low variability, good tensile strength of up to 662 MPa and tensile ductility 51% higher than what has been reported in the literature. Originality/value Although many literature studies investigate process optimization for metal AM, there is a lack of a systematic printability assessment framework to determine manufacturing process parameters for newly designed AM materials in an accelerated fashion. Moreover, the majority of existing process optimization approaches involve either time- and cost-intensive experimental campaigns or require the use of proprietary computational materials codes. Through the use of a readily accessible analytical thermal model coupled with statistical calibration and uncertainty quantification techniques, the proposed framework achieves both efficiency and accessibility to the user. Furthermore, this study demonstrates that following this framework results in printed parts with low degrees of variability in their mechanical properties.

Modeling and analysis of the on-demand spare parts supply using additive manufacturing

2019 ◽  
Vol 25 (3) ◽  
pp. 473-487 ◽  
Author(s):  
Yuan Zhang ◽  
Stefan Jedeck ◽  
Li Yang ◽  
Lihui Bai
Keyword(s):  
Supply Chain ◽  
Additive Manufacturing ◽  
Spare Parts ◽  
Production Optimization ◽  
Arrival Process ◽  
Powder Bed Fusion ◽  
Content Type ◽  
Modeling And Analysis ◽  
Powder Bed ◽  
On Demand

PurposeDespite the widespread expectation that additive manufacturing (AM) will become a disruptive technology to transform the spare parts supply chain, very limited research has been devoted to the quantitative modeling and analysis on how AM could fulfill the on-demand spare parts supply. On the other hand, the choice of using AM as a spare parts supply strategy over traditional inventory is a rising decision faced by manufacturers and requires quantitative analysis for their AM-or-stock decisions. The purpose of this paper is to develop a quantitative performance model for a generic powder bed fusion AM system in a spare parts supply chain, thus providing insights into this less-explored area in the literature.Design/methodology/approachIn this study, analysis based on a discrete event simulation was carried out for the use of AM in replacement of traditional warehouse inventory for an on-demand spare parts supply system. Generic powder bed fusion AM system was used in the model, and the same modeling approach could be applied to other types of AM processes. Using this model, the impact of both spare parts demand characteristics (e.g. part size attributes, demand rates) and the AM operations characteristics (e.g. machine size and postpone strategy) on the performance of using AM to supply spare parts was studied.FindingsThe simulation results show that in many cases the AM operation is not as cost competitive compared to the traditional warehouse-based spare parts supply operation, and that the spare parts size characteristics could significantly affect the overall performance of the AM operations. For some scenarios of the arrival process of spare parts demand, the use of the batched AM production could potentially result in significant delay in parts delivery, which necessitates further investigations of production optimization strategies.Originality/valueThe findings demonstrate that the proposed simulation tool can not only provide insights on the performance characteristics of using AM in the spare parts supply chain, especially in comparison to the traditional warehousing system, but also can be used toward decision making for both the AM manufacturers and the spare parts service providers.

scholarly journals Multi-source data fusion method for structural safety assessment of water diversion structures

2021 ◽  
Author(s):  
Sherong Zhang ◽  
Ting Liu ◽  
Chao Wang
Keyword(s):  
Data Fusion ◽  
Safety Assessment ◽  
Evidence Theory ◽  
Weighted Average ◽  
Water Diversion ◽  
Sensor Data ◽  
Structural Safety ◽  
Fusion Method ◽  
Bhattacharyya Distance ◽  
Source Data

Abstract Building safety assessment based on single sensor data has the problems of low reliability and high uncertainty. Therefore, this paper proposes a novel multi-source sensor data fusion method based on Improved Dempster–Shafer (D-S) evidence theory and Back Propagation Neural Network (BPNN). Before data fusion, the improved self-support function is adopted to preprocess the original data. The process of data fusion is divided into three steps: Firstly, the feature of the same kind of sensor data is extracted by the adaptive weighted average method as the input source of BPNN. Then, BPNN is trained and its output is used as the basic probability assignment (BPA) of D-S evidence theory. Finally, Bhattacharyya Distance (BD) is introduced to improve D-S evidence theory from two aspects of evidence distance and conflict factors, and multi-source data fusion is realized by D-S synthesis rules. In practical application, a three-level information fusion framework of the data level, the feature level, and the decision level is proposed, and the safety status of buildings is evaluated by using multi-source sensor data. The results show that compared with the fusion result of the traditional D-S evidence theory, the algorithm improves the accuracy of the overall safety state assessment of the building and reduces the MSE from 0.18 to 0.01%.

3D printing of clay paste enhanced by scrap polymer from powder bed processes

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Mattia Mele ◽  
Michele Ricciarelli ◽  
Giampaolo Campana
Keyword(s):  
Additive Manufacturing ◽  
Volumetric Flow Rate ◽  
The Body ◽  
Fused Deposition Modelling ◽  
Powder Bed Fusion ◽  
Dimensional Error ◽  
Content Type ◽  
Powder Bed ◽  
Fused Deposition ◽  
Polymer Fraction

Purpose Powder bed additive manufacturing processes are widespread due to their many technical and economic advantages. Nevertheless, the disposal of leftover powder poses a problem in terms of process sustainability. The purpose of this paper is to provide an alternative solution to recycle waste PA12 powder from HP multi jet fusion. In particular, the opportunity to use this material as a dispersion in three-dimensional (3D) printed clay is investigated. Design/methodology/approach A commercial fused deposition modelling printer was re-adapted to extrude a viscous paste composed of clay, PA12 and water. Once printed, parts were dried and then put in an oven to melt the polymer fraction. Four compositions with different PA12 concentration were studied. First, the extrudability of the paste was observed by testing different extrusion lengths. Then, the surface porosities were evaluated through microscopical observations of the manufactured parts. Finally, benchmarks with different geometries were digitalised via 3D scanning to analyse the dimensional alterations arising at each stage of the process. Findings Overall, the feasibility of the process is demonstrated. Extrusion tests revealed that the composition of the paste has a minor influence on the volumetric flow rate, exhibiting a better consistency in the case of long extrusions. The percentage of surface cavities was proportional to the polymer fraction contained in the mix. From dimensional analyses, it was possible to conclude that PA12 reduced the degree of shrinkage during the drying phase, while it increased dimensional alterations occurring in the melting phase. The results showed that the dimensional error measured on the z-axis was always higher than that of the XY plane. Practical implications The method proposed in this paper provides an alternative approach to reuse leftover powders from powder bed fusion processes via another additive manufacturing process. This offers an affordable and open-source solution to companies dealing with polymer powder bed fusion, allowing them to reduce their environmental impacts while expanding their production. Originality/value The paper presents an innovative additive manufacturing solution for powder reuse. Unlike the recycling methods in the body of literature, this solution does not require any intermediate transformation process, such as filament fabrication. Also, the cold material deposition enables the adoption of very inexpensive extrusion equipment. This preliminary study demonstrates the feasibility and the benefits of this process, paving the way for numerous future studies.

On the manufacturability of Inconel 718 thin-walled honeycomb structures by laser powder bed fusion

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
José M. Zea Pérez ◽  
Jorge Corona-Castuera ◽  
Carlos Poblano-Salas ◽  
John Henao ◽  
Arturo Hernández Hernández
Keyword(s):  
Mechanical Properties ◽  
Inconel 718 ◽  
Dimensional Accuracy ◽  
Honeycomb Lattice ◽  
Lattice Structures ◽  
Powder Bed Fusion ◽  
Laser Powder Bed Fusion ◽  
Content Type ◽  
Powder Bed ◽  
Thin Walled

Purpose The purpose of this paper is to study the effects of printing strategies and processing parameters on wall thickness, microhardness and compression strength of Inconel 718 superalloy thin-walled honeycomb lattice structures manufactured by laser powder bed fusion (L-PBF). Design/methodology/approach Two printing contour strategies were applied for producing thin-walled honeycomb lattice structures in which the laser power, contour path, scanning speed and beam offset were systematically modified. The specimens were analyzed by optical microscopy for dimensional accuracy. Vickers hardness and quasi-static uniaxial compression tests were performed on the specimens with the least difference between the design wall thickness and the as built one to evaluate their mechanical properties and compare them with the counterparts obtained by using standard print strategies. Findings The contour printing strategies and process parameters have a significant influence on reducing the fabrication time of thin-walled honeycomb lattice structures (up to 50%) and can lead to improve the manufacturability and dimensional accuracy. Also, an increase in the young modulus up to 0.8 times and improvement in the energy absorption up to 48% with respect to those produced by following a standard strategy was observed. Originality/value This study showed that printing contour strategies can be used for faster fabrication of thin-walled lattice honeycomb structures with similar mechanical properties than those obtained by using a default printing strategy.

Process monitoring of laser powder bed fusion

2021 ◽  
pp. 301-326
Author(s):  
Marco Grasso ◽  
Bianca Maria Colosimo ◽  
Kevin Slattery ◽  
Eric MacDonald
Keyword(s):  
Process Monitoring ◽  
Powder Bed Fusion ◽  
Laser Powder Bed Fusion ◽  
Powder Bed

Camera signal dependencies within coaxial melt pool monitoring in laser powder bed fusion

2020 ◽  
Vol 26 (1) ◽  
pp. 100-106 ◽  
Author(s):  
Tobias Kolb ◽  
Reza Elahi ◽  
Jan Seeger ◽  
Mathews Soris ◽  
Christian Scheitler ◽  
...  
Keyword(s):  
Surface Roughness ◽  
Layer Thickness ◽  
Powder Layer ◽  
Powder Bed Fusion ◽  
Laser Powder Bed Fusion ◽  
Content Type ◽  
Powder Bed ◽  
Sensory Signals ◽  
Melt Pool ◽  
Research Activities

Purpose The purpose of this paper is to analyse the signal dependency of the camera-based coaxial monitoring system QMMeltpool 3D (Concept Laser GmbH, Lichtenfels, Germany) for laser powder bed fusion (LPBF) under the variation of process parameters, position, direction and layer thickness to determine the capability of the system. Because such and similar monitoring systems are designed and presented for quality assurance in series production, it is important to present the dominant signal influences and limitations. Design/methodology/approach Hardware of the commercially available coaxial monitoring QMMeltpool 3D is used to investigate the thermal emission of the interaction zone during LPBF. The raw images of the camera are analysed by means of image processing to bypass the software of QMMeltpool 3D and to gain a high level of signal understanding. Laser power, scan speed, laser spot diameter and powder layer thickness were varied for single-melt tracks to determine the influence of a parameter variation on the measured sensory signals. The effects of the scan direction and position were also analysed in detail. The influence of surface roughness on the detected sensory signals was simulated by a machined substrate plate. Findings Parameter variations are confirmed to be detectable. Because of strong directional and positional dependencies of the melt-pool monitoring signal a calibration algorithm is necessary. A decreasing signal is detected for increasing layer thickness. Surface roughness is identified as a dominating factor with major influence on the melt-pool monitoring signal exceeding other process flaws. Research limitations/implications This work was performed with the hardware of a commercially available QMMeltpool 3D system of an LPBF machine M2 of the company Concept Laser GmbH. The results are relevant for all melt-pool monitoring research activities connected to LPBF, as well as for end users and serial production. Originality/value Surface roughness has not yet been revealed as being one of the most important origins for signal deviations in coaxial melt-pool monitoring. To the best of the authors’ knowledge, the direct comparison of influences because of parameters and environment has not been published to this extent. The detection, evaluation and remelting of surface roughness constitute a plausible workflow for closed-loop control in LPBF.

Sign in / Sign up

Export Citation Format

Share Document