Process Optimization Framework Research Articles

Research on Building Process Optimization Framework Based on BIM-DT

Research on Building Process Optimization Framework Based on BIM-DT

Assessment of data‐driven modeling approaches for chromatographic separation processes

AbstractChromatographic separation processes are described by nonlinear partial differential and algebraic equations, which may result in high computational cost, hindering further online applications. To decrease the computational burden, different data‐driven modeling approaches can be implemented. In this work, we investigate different strategies of data‐driven modeling for chromatographic processes, using artificial neural networks to predict pseudo‐dynamic elution profiles, without the use of explicit temporal information. We assess the performance of the surrogates trained on different dataset sizes, achieving good predictions with a minimum of 3400 data points. Different activation functions are used and evaluated against the original high‐fidelity model, using accuracy, interpolation, and simulation time as performance metrics. Based on these metrics, the best performing data‐driven models are implemented in a process optimization framework. The results indicate that data‐driven models can capture the nonlinear profile of the process and that can be considered as reliable surrogates used to aid process development.

Active interlocking metasurfaces enabled by shape memory alloys

Interlocking metasurfaces (ILMs) are a newly developed joining technology that relies on arrays of interlocking features that transmit force and constrain motion between adjoining bodies in one or more directions. This study explores harnessing the shape memory effect (SME) in Nickel-Titanium shape memory alloys (NiTi SMAs) in structures fabricated using additive manufacturing (AM) to advance the development of active ILMs by creating unit cells that open or close at specific temperatures. The study encompasses designing and fabricating two distinct interlocking array configurations using near-equiatomic NiTi powder and the laser powder bed fusion (L-PBF) AM technique, following a previously developed AM process optimization framework to manufacture defect-free parts. To guide the design process, finite element analysis (FEA) was employed to predict strain values during engage-disengage cycles. The martensitic transformation characteristics of the ILMs were characterized. Thermomechanical testing revealed that the ILMs demonstrate high locking force once engaged, coupled with complete shape recovery and good cyclic stability. Digital image correlation (DIC) was also employed to validate the FEA predictions during the engage-disengage cycles. The results indicate that NiTi SMA-based ILMs can be designed and fabricated into complex shapes using L-PBF. By leveraging the SME, the functionality of an ILM can be improved upon. The combination of computational modeling, additive manufacturing, and thermomechanical and physical property characterization provides a framework for designing future ILMs out of active materials.

Comparative Study of Multi‐objective Bayesian Optimization and NSGA‐III based Approaches for Injection Molding Process

AbstractInjection molding is a prevalent method for producing plastic components, yet determining the ideal process parameters has predominantly relied on heuristic approaches. In this research, a data‐driven injection molding process optimization framework is developed to simultaneously minimize warpage, cycle time, and clamping force. Employing multi‐objective Bayesian optimization (MBO), the framework is applied to a fan blade model for verification. Incorporating those objectives enables the selection of an injection molding machine with the proper maximum clamping force within acceptable tolerance and production time. Furthermore, non‐moldable regimes are considered in design space, which allows less prior knowledge for setting design space. The optimized results are compared with those obtained using NSGA‐III, a well‐established genetic algorithm‐based optimization technique, as a benchmark. The optimization frameworks produce a Pareto front for the three‐dimensional outputs, revealing distinct trade‐off relationships between cycle time and warpage, as well as clamping force and warpage. The MBO framework demonstrates a superior Pareto front compared to NSGA‐III when utilizing a limited data set, underscoring its benefits in scenarios where costly simulations or experiments are necessary. These findings are anticipated to contribute to the optimization of manufacturing processes, ultimately enhancing productivity in real‐world industries.

A machine learning framework for process optimization in aerosol jet 3D printing

As a burgeoning three-dimensional (3D) printing technology, aerosol jet printing (AJP) technique has the characteristics of direct writing and customizing microelectronic components with flexible substrates. Therefore, it has been widely applied to manufacture different electronic devices. Although AJP has unique advantages over traditional methods, the electrical performance of printed electronic devices is significantly reduced because of the inferior printing qualities, such as high overspray, low level of line thickness and high level of edge roughness. Therefore, producing lines with high-controllability and high-aspect ratio is urgent for AJP technology. In this research, a machine learning scheme is developed for process optimization in AJP. In the proposed scheme, a support vector machine is combined with Latin hyper sampling to determine an optimal operating window of AJP, producing conductive lines with better edge definition and reduced overspray. Then, based on the identified 3D operating window, the conflicting relationship between the deposited line width and thickness was revealed based on the developed Gaussian process regression models. Following that, via a non-dominated sorting genetic algorithm, the conflicting printed line morphology was further optimized under dual conflicting targets for maximizing line thickness and customizing line width, which helps to produce high-controllability and high-aspect ratio lines for AJP. The optimization results demonstrated the validity of the proposed approach, which is beneficial to the systemic optimization of the entire printing process.

Low-Carbon-Driven Product Life-Cycle Process Optimization Framework for Manufacturing Equipment

Because of the increasing concern regarding environmental sustainability, manufacturing equipment faces a challenge to optimize carbon emissions from a life-cycle perspective. Current approaches lack the overall low-carbon optimization flow guidance needed in order to take effective measures for manufacturing equipment. Consequently, this paper proposes a framework of low-carbon optimization for manufacturing equipment. Firstly, a four-layer framework for low-carbon optimization processes is established, consisting of the optimization operation layer, life-cycle layer, optimization method layer, and tools and data layer. Then, the characteristics, functions, and technologies involved in the four layers are elaborated on in detail. Finally, the honing machine considering the reduction of carbon emissions is given as an example. The results indicate that by using the proposed methodology effectively, the carbon emissions of the lower column, a key part of the honing machine, can be reduced by 4.38% without sacrificing structure performance. The framework can provide specific guidance on achieving a low-carbon optimization process and contribute to the sustainable development of energy-intensive manufacturing industries.

A reinforcement learning/ad-hoc planning and scheduling mechanism for flexible and sustainable manufacturing systems

The process scheduling is still considered a crucial subject for manufacturing industry, due to the ever-changing circumstances dictated by the nowadays product demand and customer trends. These conditions are often associated with increasing costs and energy consumption, considerably affecting the long-term sustainability of manufacturing plants. To mitigate that effect, one should create an effective strategy tailoring integrated operations and processes to the customer demand and trends faced by the nowadays industry. A well-known approach to this matter is the technologies introduced by manufacturing paradigms, e.g., Industry 4.0 and smart manufacturing. As suggested in literature, these technologies are capable of helping decision-makers by continuously gathering significant information about the state of machinery and manufactured goods. This information is thereafter utilized to identify weaknesses and strengths demonstrated within manufacturing plants. To this end, the present paper presents a process optimization framework implemented in a three-stage production line prone to systematic degradation faults. Aiming at strengthening profitability, the framework engages reinforcement learning with ad-hoc manufacturing/maintenance control in decision-making carried out in implemented machines. Simulation experiments showed improved process planning and inventory management enabling cost-effective green and sustainable manufacturing in manufacturing plants.

A lightweight Fe–Mn–Al–C austenitic steel with ultra-high strength and ductility fabricated via laser powder bed fusion

Lightweight Fe–Mn–Al–C steels have become a topic of significant interest for the defense and automotive industries. These alloys can maintain high strength and ductility while also reducing weight in structural applications. Conventionally processed Fe–Mn–Al–C austenitic steels with high Al content (∼9 wt%) demonstrate greater than 1.5 GPa strength with 35% elongation. Several recent studies have demonstrated success in fabricating steel parts using laser powder bed fusion (L-PBF) additive manufacturing (AM), which can generate near-net-shape components with complex geometries and is capable of local microstructural control. However, studies on L-PBF processing of Fe–Mn–Al–C alloys have focused on low Al content (<5 wt%) compositional regimes representing alloys that undergo transformation-induced plasticity (TRIP) and twinning-induced plasticity (TWIP). Here, we present the effects of L-PBF processing on the microstructure and mechanical properties of an Fe–30Mn–9Al–1Si-0.5Mo-0.9C austenitic steel. A process optimization framework is employed to determine an ideal L-PBF processing space that will result in >99% density parts. Implementing this framework resulted in near-fully dense specimens fabricated over a broad range of process parameters. Additionally, two bi-directional scan rotation strategies (90° and 67°) were applied to understand their effects on texture and anisotropy in this material. As-printed specimens displayed considerable work-hardening characteristics with average strengths of up to 1.3 GPa and 36% elongation in the build direction. However, solidification microcracks oriented in the build direction resulted in anisotropy in tensile strength and ductility resulting in average strengths of 1.1 GPa and 20% elongation perpendicular to the build direction. The successful L-PBF fabrication of Fe–30Mn–9Al–1Si-0.5Mo-0.9C presented here is expected to open new avenues for weight reduction in structural applications with a high degree of control over part topology.

Development of a process optimization framework for fabricating fully dense advanced high strength steels using laser directed energy deposition

Development of a process optimization framework for fabricating fully dense advanced high strength steels using laser directed energy deposition

Smart contract design and process optimization of carbon trading based on blockchain: The case of China's electric power sector

Carbon trading is an important market instrument and policy tool to promote China's practice of low-carbon energy conservation and strive to achieve carbon peaking by 2030 and carbon neutrality by 2060. To improve market flow and respond to the national strategy to support the development of carbon trading in China, a smart carbon trading process optimization framework based on blockchain technology is developed in this article. First, the paper analyzes the current state of carbon trading in China's power sector and identifies the future development trend and barriers to carbon trading in China. Second, a framework for a smart carbon trading system based on the structure and traits of blockchain technology is created to optimize the carbon trading process. Third, the smart contract for the smart carbon trading system is built (including execution logic, transaction matching, contract fulfillment, and other elements). This mechanism measures the reputation of trading users. Finally, policy suggestions are provided to solve the current issues and challenges. According to the research conducted for this paper on “blockchain + carbon trading in China,” carbon trading supported by blockchain technology can guarantee transaction security and efficiency, accurately preserve transaction records, and offer a high level of automated settlement.

A process optimization framework for laser direct energy deposition: Densification, microstructure, and mechanical properties of an Fe[sbnd]Cr alloy

Laser Direct Energy Deposition (DED) is a metal additive manufacturing technique with the ability to fabricate large and complex parts through deposition of metal powders. However, achieving high-density parts and targeted build heights using DED can be challenging due to the large number of highly sensitive process variables. This work proposes a robust fabrication parameter optimization framework to generate process maps for primary parameters in DED, including laser power, scan speed, mass flow rate, hatch spacing, and layer height. Simple single-track experiments were utilized to map out the parameter space, and a combination of geometric criteria for hatch spacing and layer height were proposed to determine parameter sets that achieve both targeted build heights and mitigate porosity formation. Using this framework, specimens with >99 % density and consistent mechanical properties were successfully fabricated over a wide range of process parameters for an Fe-9wt.%Cr (Fe9Cr) alloy, a surrogate for radiation damage-resistant reduced activation ferritic/martensitic (RAFM) steels. Processing these materials using DED is of particular interest in the development of plasma facing components for nuclear fusion applications. The microstructure and mechanical properties of as-printed Fe9Cr were characterized using optical and electron microscopy, X-ray diffraction, and uniaxial tensile tests. As-printed Fe9Cr displayed ~25 % elongation and ultimate tensile strengths of up to 475 MPa which is comparable to similar wrought alloys. The proposed framework will allow for accelerated DED parameter optimization for novel alloy systems, as well as open the possibility for local microstructure control while simultaneously mitigating defect formation.

In-situ optimization of thermoset composite additive manufacturing via deep learning and computer vision

With the advent of extrusion additive manufacturing (AM), fabrication of high-performance thermoset composites without the need of tooling has become a reality. However, finding an optimal set of printing parameters for these thermoset composites during extrusion requires tedious experimentation as composite ink properties can vary significantly with respect to environmental parameters such as temperature and relative humidity. Addressing this challenge, this study presents a novel optimization framework that utilizes computer vision and deep learning (DL) to optimize the calibration and printing processes of thermoset composite AM. Unlike traditional DL models where printing parameters are determined prior to printing, our proposed framework dynamically and autonomously adjusts the printing parameters during extrusion. A novel DL integrated extrusion AM system is developed to determine the optimal printing parameters including print speed, road width, and layer height for a given composite ink. This closed loop system is consisted of a computer communicating with an extrusion AM system, a camera to perform in-situ imaging and several high accuracy convolution neural networks (CNNs) selecting the ideal process parameters for composite AM. The results show that our proposed process optimization framework was able to autonomously determine these parameters for a carbon fiber-composite ink. Consequently, specimens with complex geometries could be fabricated without visible defects and with maximum fiber alignment and thus enhancing the mechanical performance of the specimen’s composite material. Moreover, our proposed framework minimizes a labor-intensive procedure required to additively manufacture thermoset composites by optimizing the extrusion process without any user intervention.

Optimization of multistage femtosecond laser drilling process using machine learning coupled with molecular dynamics

Although femtosecond laser percussion drilling is widely used in many key industrial manufacturing fields, the quality of micro-holes limits its service performance. The traditional trial and error method and physical model are always time and cost consuming for process optimization. To address this problem, a process optimization framework coupled with molecular dynamics simulation, machine learning and a high-throughput optimization algorithm is proposed. The physical information obtained by molecular dynamics enriches the data set used to train machine learning model, and thus reducing the amount of data required. Machine learning can quickly and accurately establish the regression model between laser parameters and target machining performance, and high-throughput optimization algorithm is responsible for determining the optimal process in the process space. For femtosecond laser percussion drilling, a new four-stage drilling process is proposed and optimized using the coupling process optimization framework to solve efficiency and quality problems. Finally, the experimental validation for Ni-based single crystal superalloy verifies the reliability of the optimized process. The framework can be extended to other complex systems to realize process optimization with high efficiency and low cost in laser material processing technology.

Towards the 5th Industrial Revolution: A literature review and a framework for Process Optimization Based on Big Data Analytics and Semantics

1. EUROSTAT, 2019, Employment Specialisation of EU Regions, Available at: https://ec.europa.eu/eurostat/.... Google Scholar

A three-level hierachical framework for additive manufacturing

Metal alloy additive manufacturing (AM) has gained wide industrial interest in the past decade due to its capability to efficiently deliver complicated mechanical parts with high quality. However, due to a lack of understanding of the fundamental correlation between the operating conditions and build quality, the exploration of the optimal operating policy of the AM process is costly and difficult. In this work, a data-driven process optimization framework has been proposed for the additive manufacturing process, integrating machine learning, finite-element method (FEM) modeling, and cloud-edge data storage/transfer optimization. A three-level hierarchy of local machines, factory clouds, and a research center is introduced with each level responsible for its dedicated tasks. In addition, to ensure the efficiency of data transfer and storage, an edge-cloud data transfer scheme is constructed, which serves as a guideline for the data flow in the AM framework. Moreover, an overview of the connections between the proposed framework and the Industry 4.0 framework is presented.

A machine learning framework for real-time inverse modeling and multi-objective process optimization of composites for active manufacturing control

For manufacturing of aerospace composites, several parts may be processed simultaneously using convective heating in an autoclave. Due to uncertainties including tool placement, convective Boundary Conditions (BCs) vary in each run. As a result, temperature histories in some of the parts may not conform to process specifications due to under-curing or over-heating. Thermochemical analysis using Finite Element (FE) simulations are typically conducted prior to fabrication based on assumed range of BCs. This, however, introduces unnecessary constraints on the design. To monitor the process, thermocouples (TCs) are placed under tools near critical locations. The TC data may be used to back-calculate BCs using trial-and-error FE analysis. However, since the inverse heat transfer problem is ill-posed, many solutions are obtained for given TC data. In this study, a novel machine learning (ML) framework is presented capable of optimizing air temperature cycle in real-time based on TC data from multiple parts, for active control of manufacturing. The framework consists of two recurrent Neural Networks (NN) for inverse modeling of the ill-posed curing problem at the speed of 300 simulations/second, and a classification NN for multi-objective optimization of the air temperature at the speed of 35,000 simulations/second. A virtual demonstration of the framework for process optimization of three composite parts with data from three TCs is presented.

Controlling martensitic transformation characteristics in defect-free NiTi shape memory alloys fabricated using laser powder bed fusion and a process optimization framework

Laser Powder Bed Fusion (L-PBF) was utilized to fabricate fully dense, near-equiatomic (Ni50.1Ti49.9) and Ni-rich NiTi (Ni50.8Ti49.2) shape memory alloy (SMA) parts which exhibited tensile ductility up to 16%, shape memory strain of 6%, and tensile superelasticity up to 4%. Annealing heat treatments marginally improved the superelasticity of the as-fabricated Ni-rich NiTi parts due to the formation of a small volume fraction of Ni4Ti3 precipitates. The selection of optimum processing parameters that yielded fully dense parts was guided by a process optimization framework based on a computationally inexpensive analytical model used to predict the melt pool dimensions. The framework also included single-track experiments to validate the model predictions and a criterion for the maximum allowable hatch spacing to prevent the formation of lack of fusion porosity. This framework allowed for constructing L-PBF processing maps for the present NiTi SMAs and revealed that fully dense parts could be printed over a wide range of process parameters. By controlling the L-PBF process parameters, in particular laser power, laser scan speed, and volumetric energy density, in the processing space that result in fully dense parts, it was demonstrated systematically that the composition of the printed parts could be precisely changed by controlling the evaporation of Ni. The flexibility of parameter selection to print defect-free NiTi SMAs and composition control by preferential evaporation of Ni opens the possibility to print functional NiTi SMA parts or devices without post-processing.

A distributed feedback-based online process optimization framework for optimal resource sharing

Distributed real-time optimization (RTO) enables optimal operation of large-scale process systems with common resources shared across several clusters. Typically in distributed RTO, the different subsystems are optimized locally, and a centralized master problem is used to coordinate the different subsystems in order to reach system-wide optimal operation. This is especially beneficial in industrial symbiosis, where only limited information can be shared between the different clusters. However, one of the main challenges with this approach is the need to solve numerical optimization problems online for each subsystem. With the recent surge of interest in feedback optimizing control, where the optimization problem is converted into a feedback control problem, this paper proposes a distributed feedback-based RTO (DFRTO) framework for optimal resource sharing in an industrial symbiotic setting. In this approach, a master coordinator updates the shadow price for the shared resource, and the different subsystems locally optimize their operation using feedback control for the given shadow price. The proposed framework is shown to converge to a stationary point of the system-wide optimization problem, and is demonstrated using an industrial symbiotic offshore oil and gas production system with shared resources.

New process optimization framework for laser assisted tape winding of composite pressure vessels: Controlling the unsteady bonding temperature

This paper presents an effective process optimization methodology for laser assisted tape winding (LATW) of complex part geometries by means of a numerical optical-thermal model. A winding path on the cylindrical and ellipsoidal (dome) part of a pressure vessel is considered with varying tooling curvature. First, the process model output is verified with the literature data based on the laser intensity distribution. Then, the transient laser irradiation and temperature distributions on the tape and substrate are described comprehensively. It is shown that the maximum laser intensity increases approximately by 80% and the process (bonding) temperature changes by 80 °C at the intersection of the cylindrical and dome section of the pressure vessel. In order to keep the transient process temperature constant, a robust optimization scheme is utilized by means of a genetic algorithm. The design variable is determined as the total laser power and temperature constraints are defined. The proposed optimization methodology regulates the temperature within 1.5 °C variation with respect to the desired value. In order to compensate the transient local curvature effects on the process temperature, the total laser power varies approximately between 30% and 175% of the reference (non-optimized) case.

An ultra-high strength martensitic steel fabricated using selective laser melting additive manufacturing: Densification, microstructure, and mechanical properties

Martensitic steels have gained renewed interest recently for their use in automotive, aerospace, and defense applications due to their ultra-high yield strengths and reasonable ductility. A recently discovered low alloy martensitic steel, AF9628, has been shown to exhibit strengths greater than 1.5 GPa with more than 10% tensile ductility, due to the formation of ε-carbide phase. In an effort to produce high strength parts with a high degree of control over geometry, the work herein presents the effects of selective laser melting (SLM) parameters on the microstructure and mechanical properties of this new steel. An optimization framework to determine the process parameters for building porosity-free parts is introduced. This framework utilizes the computationally inexpensive Eagar-Tsai model, calibrated with single track experiments, to predict the melt pool geometry. A geometric criterion for determining maximum allowable hatch spacing is also developed in order to avoid lack of fusion induced porosity in the as-printed parts. Using this framework, fully dense samples were successfully fabricated over a wide range of process parameters, allowing the construction of an SLM processing map for AF9628. The as-printed samples displayed tensile strengths of up to 1.4 GPa, the highest reported to date for any 3D printed alloy, with up to 11% elongation. The demonstrated flexibility in process parameter selection, while maintaining full density, opens up the possibility of local microstructural refinement and parameter optimization for improved mechanical properties in as-printed parts. The process optimization framework introduced here is expected to allow successful printing of new materials in an accelerated fashion.