Procedia CIRP最新文献

Sensor data is increasingly offering better operational visibility. However, the data deluge is also posing cost and complexity challenges on the data analytics pipeline, which comprises of edge computing, power, transmission, and storage for data-driven decision making. To address the data deluge problem, we propose a machine learning assisted approach of collecting less data upfront to solve different sensor data analytics problems. While sampling at Nyquist rates, we do not collect every data point, but rather sample according to the information content in the signal. A comprehensive experimental design is undertaken to show that collecting more than a certain fraction of raw data only leads to infinitesimal performance improvements. The engineering advantages of the proposed near real-time approach are quantified showing a significant reduction in analytics pipeline resources required for industrial digital transformation applications.

Additive manufacturing with fused filament fabrication (FFF) offers rapid prototyping, design flexibility, and cost efficiency, making it a disruptive technology in comparison with traditional subtractive manufacturing. However, the achievable accuracy of current FFF printers is limited, prompting research into factors affecting print quality. This study focuses on investigating the influence of stepper motors and servo motors on 3D printer performance, specifically in a crossed gantry design commonly used in Cartesian printers. Different motors were used in the experimental setup to perform tests such as circularity testing, operational vibration analysis, and repeatable positioning accuracy measurement. The results demonstrate that servo motors outperform stepper motors in terms of vibration reduction and positioning accuracy. Furthermore, it was found that the choice of motor had a limited impact on the achievable surface quality, while vibrations during movement had a more significant influence. This research contributes to understanding the effects of motor selection on 3D printer performance, providing insights for improving additive manufacturing processes.

A common aerospace and defense industry challenge is low volume production of components for legacy aircraft due to compromised casting and forging supply chains. A hybrid manufacturing approach is presented to address this challenge that uses additive friction stir deposition, structured light scanning, and CNC milling. The paper describes a novel slicing and toolpath development strategy for additive friction stir deposition of a relevant aerospace geometry, post deposition measurement of the two-sided preform and identification of the machining work coordinate system, and five-axis CNC machining to obtain the final part geometry while ensuring stable machining behavior.

Machine tool volumetric compliance quantifies the deflection of the tool tip relative to the workpiece under the effect of a mutual force. Its knowledge can help to virtually monitor tool path errors. This study, conducted under static conditions, aims to establish a model for directly predicting static translational volumetric errors caused by compliance from machine tool CNC feed motor torque outputs without the need to calculate the tool tip disturbance force. Static translational volumetric compliance was measured as proposed in ISO 230-1 in the X, Y and Z directions. The friction torques in each direction were modelled as a function of volumetric errors using Dahl's theory with some adjustments. The friction torque variations were detected. The final model links feed motor torques directly to static translational volumetric errors. With the help of an exponential term, the model proposed has a better performance on capturing the friction variations at the motion starting and movement reversal points. The static translational volumetric compliance models had R²_adj values of 0.999, 0.997, and 0.997 in the X, Y and Z directions, respectively. The RMSE with the best starting estimates on the testing dataset for predicting the static translational volumetric errors directly from the feed motor torques along the X-, Y- and Z-axis were 4.9 μm, 6.6 μm and 2.8 μm under maximum absolute volumetric errors values of 200 μm, 200 μm, and 50 μm, respectively.

Hybrid modeled digital twins, using a combination of physics-based and data-driven models, offer great potential for manufacturing process improvement at different levels of the manufacturing system. Commonly digital twin approaches in academic research have a high level of model input availability. However, in industry this availability level is often not met. This leads to significant gaps for the model development and operation, that hinder the transfer of the benefits to actual manufacturing processes. To overcome these hurdles, this paper presents a conceptual framework for identifying, analyzing, and overcoming model input gaps. To exemplify the proposed approach, we present a polishing process use case based on the data of a hybrid model from previous work.

Smart manufacturing (SM) enhances the competitiveness of manufacturing companies by promoting automation and overall equipment effectiveness (OEE), targeting to produce 100% qualified products fully automatically. One of the key challenges to the SM initiatives is the continuous demand fluctuations in the specification and quantity, especially when a new product variant comes to the production line. Reconfigurable manufacturing (RM) system provides cost-effective, rapid response to abrupt market changes. It provides a solution by its flexibility in repurposing tools, adding machines, and modifying software to rapidly respond to changing demands at low unit costs. The ability of SM technologies through self-programming and cloud computation may significantly complements RM initiatives. There is increasing evidence that SM and RM may augment each other through their complementary strengths, leading to the new paradigm of smart reconfigurable manufacturing (SRM). To highlight the complementary strengths, this paper investigates the converging trend of RM and SM based on natural language processing, e.g., topic modeling and semantic embedding. Key characteristics and industrial use cases are subsequently summarized to systematically delineate the new SRM paradigm and illustrate its advantages and feasibility in practice.

The digital representation of physical products by Digital Twins has been increasingly perceived as a relevant enabler for Product-Service Systems (PSS). Digital Twins concentrate product data that can be applied to generate insights and thus support value-added services. However, the literature on the intersection between Digital Twin and PSS has only recently started to receive more attention. There are still research gaps related to the required data and systems integration. In this context, this research aims to propose a Digital Twin data architecture to support Product-Service Systems operation. A Design Science Research (DSR) approach is applied, and the proposed architecture has been implemented and tested. Assessment results indicated that the proposed Digital Twin architecture fulfills the four requirements established from the literature: 1) facilitate and support the service offering; 2) acquire and transmit field operation and customer data; 3) integrate design and manufacturing data; 4) guarantee real-time monitoring, data integration, and data fidelity. The presented results provide an original contribution to the research area and can serve as a reference for applying Digital Twin to support PSS in practice.