An intelligent system for reflow oven temperature settings based on hybrid physics-machine learning model

Lai, Yangyang; Pan, Ke; Cen, Yuqiao; Yang, Junbo; Cai, Chongyang; Yin, Pengcheng; Park, Seungbae

doi:10.1108/ssmt-10-2021-0063

Cited by 26 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the trend of heterogeneous integration technology, advanced package features varied designs such as redistribution layers, interposers and multiple chiplets (Pan et al , 2022). It is not feasible to use detailed models because of their variety and complexity (Lai et al , 2022b). Instead, ultimate compact modeling was used to make the model universal, which means package-structure independence.…”

Section: Field Measurement and Computational Fluid Dynamics Modelingmentioning

confidence: 99%

“…The use of ML models in this study enables the rapid generation of temperature profiles in the same format as the measurement data, representing a significant improvement over the authors' previous approaches where only peak temperature and TAL were available as outputs (Lai et al, 2023(Lai et al, , 2022b. The algorithms play the roles of heat transfer of the forced convection and transient conduction, which are the two dominant heat transfer mechanisms of the reflow soldering process.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reflow profiling with the aid of machine learning models

Lai

Park

2023

SSMT

Self Cite

View full text Add to dashboard Cite

Purpose This paper aims to propose a method to quickly set the heating zone temperatures and conveyor speed of the reflow oven. This novel approach intensely eases the trial and error in reflow profiling and is especially helpful when reflowing thick printed circuit boards (PCBs) with bulky components. Machine learning (ML) models can reduce the time required for profiling from at least half a day of trial and error to just 1 h. Design/methodology/approach A highly compact computational fluid dynamics (CFD) model was used to simulate the reflow process, exhibiting an error rate of less than 1.5%. Validated models were used to generate data for training regression models. By leveraging a set of experiment results, the unknown input factors (i.e. the heat capacities of the bulkiest component and PCB) can be determined inversely. The trained Gaussian process regression models are then used to perform virtual reflow optimization while allowing a 4°C tolerance for peak temperatures. Upon ensuring that the profiles are inside the safe zone, the corresponding reflow recipes can be implemented to set up the reflow oven. Findings ML algorithms can be used to interpolate sparse data and provide speedy responses to simulate the reflow profile. This proposed approach can effectively address optimization problems involving multiple factors. Practical implications The methodology used in this study can considerably reduce labor costs and time consumption associated with reflow profiling, which presently relies heavily on individual experience and skill. With the user interface and regression models used in this approach, reflow profiles can be swiftly simulated, facilitating iterative experiments and numerical modeling with great effectiveness. Smart reflow profiling has the potential to enhance quality control and increase throughput. Originality/value In this study, the employment of the ultimate compact CFD model eliminates the constraint of components’ configuration, as effective heat capacities are able to determine the temperature profiles of the component and PCB. The temperature profiles generated by the regression models are time-sequenced and in the same format as the CFD results. This approach considerably reduces the cost associated with training data, which is often a major challenge in the development of ML models.

show abstract

Section: Field Measurement and Computational Fluid Dynamics Modelingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Reflow profiling with the aid of machine learning models

Lai

Park

2023

SSMT

Self Cite

View full text Add to dashboard Cite

show abstract

“…Serial architectures typically have the data-driven models set in sequence with the partial physics model or used to tune the partial physics model parameters, while parallel architectures usually contain additive or multiplicative ensembles of partial physics and data-driven ML models [23]. Several such hybrid PIML architectures have been reported in the literature in the past few years [24]- [36], spanning over a wide range of applications such as in modeling dynamic systems, cyber-physical systems, robotic systems, flow systems and materials behavior, among others. The OPTMA model [37], is a physics-infused machine learning model which combines an artificial neural network with a partial physics model in order to make predictions.…”

Section: Introductionmentioning

confidence: 99%

Physics Infused Machine Learning Based Prediction of VTOL Aerodynamics with Sparse Datasets

Oddiraju

Amin²,

Piedmonte³

et al. 2023

AIAA AVIATION 2023 Forum

View full text Add to dashboard Cite

Complex optimal design and control processes often require repeated evaluations of expensive objective functions and consist of large design spaces. Data-driven surrogate models such as neural networks and Gaussian processes provide an attractive alternative to expensive simulations and are utilized frequently to represent these objective functions in optimization. However, pure data-driven models, due to a lack of adherence to basic physics laws and constraints, are often poor at generalizing and extrapolating. This is particularly the case, when training occurs over sparse high-fidelity datasets. A class of Physicsinfused machine learning (PIML) models integrate ML models with low-fidelity partial physics models to improve generalization performance while retaining computational efficiency. This paper presents two potential approaches for Physics infused modelling of aircraft aerodynamics which incorporate Artificial Neural Networks with a low-fidelity Vortex Lattice Method model with blown wing effects (BLOFI) to improve prediction performance while also keeping the computational cost tractable. This paper also develops an end-to-end auto differentiable open-source framework that enables efficient training of such hybrid models. These two PIML modelling approaches are then used to predict the aerodynamic coefficients of a 6 rotor eVTOL aircraft given its control parameters and flight conditions. The models are trained on a sparse high-fidelity dataset generated using a CHARM model. The trained models are then compared against the vanilla low-fidelity model and a standard pure data-driven ANN. Our results show that one of the proposed architecture outperforms all the other models and at a nominal increase in execution time. These results are promising and pave way to PIML frameworks which are able generalize over different aircraft and configurations thereby significantly reducing costs of design and control.

show abstract

“…Ganggang Yan et al simulated a resistance furnace control system based on Fuzzy PID algorithm [2]; Bao Qingfeng et al modeled the temperature field of rolling steel heating furnace in metallurgical industrial process based on RHFTF [3]. Yangyang Lai et al provided a solution to determine the optimal pre-set temperature optimization model for reflow ovens [4]; Jing Shilong et al analyzed and optimized the oven temperature profile prediction model for high product heat capacity and poor wet ability in reflow soldering process [5].…”

Section: Introductionmentioning

confidence: 99%

Optimization And Analysis of Furnace Temperature Profile for The Processing Technology of Electronic Components

Li¹

2023

HSET

View full text Add to dashboard Cite

The soldering furnace is an important piece of equipment for processing and soldering electronic components in the integrated circuit design process. After the printed circuit boards are mounted with various electronic components, they are placed in the soldering furnace for heating so that the electronic components are automatically soldered to the circuit boards. In this process, the furnace temperature profile, which records temperature data of the electronic components, is an important reference indicator for measuring the effectiveness of electronic component processing. In the context of electronic component soldering, the transfer speed of the furnace and the setting of the temperature in each temperature zone are decisive in ensuring the quality of the component soldering. Based on the Newton's law of cooling and heat equation, this paper establishes several mathematical models to study the effect of the temperature setting of each temperature zone and the transfer speed of the soldering furnace on the furnace temperature profile with the help of multi-objective planning, circular search algorithm and other mathematical methods. Finally, the corresponding conclusions and optimization methods of furnace temperature profile are drawn, which have certain significance for the processing of electronic components.

show abstract

An intelligent system for reflow oven temperature settings based on hybrid physics-machine learning model

Cited by 26 publications

References 27 publications

Reflow profiling with the aid of machine learning models

Reflow profiling with the aid of machine learning models

Physics Infused Machine Learning Based Prediction of VTOL Aerodynamics with Sparse Datasets

Optimization And Analysis of Furnace Temperature Profile for The Processing Technology of Electronic Components

Contact Info

Product

Resources

About