An Investigation Of Carbon Emission Prediction And Optimization Of Fused Deposition Modeling Process Using Hybrid Data-Driven And Physical Modelling Method

"Carbon peak and carbon neutrality" is a critical objective in China’s green and low-carbon economic development.The 14 th Five-Year Plan is regarded as a crucial phase for China to achieve its carbon peak goal.The State Council and the Ministry of Industry and Information Technology have issued critical directives urging for the reduction of industrial carbon emissions.As an essential technology in intelligent manufacturing,additive manufacturing has significant potential for energy conservation and carbon reduction due to its unique forming process.However,the energy consumption and carbon emission prediction mechanism of the additive manufacturing process is still unsatisfactory.Its model accuracy is low,and the lack of collaborative optimization research on carbon emissions and part quality somewhat limits its application in the green and low-carbon manufacturing domain.To address these limitations,this paper selects the widely utilized melting deposition forming technique in additive manufacturing technology as the research object.Mechanism-level analysis of the energy flow and carbon footprint of the melting deposition forming process is conducted.High-quality and low-carbon manufacturing processes of melting deposition forming were further studied with the data-mechanism mixed-driven method.The following specifics are covered:Firstly,the article presents the analysis of the numerical control instruction G-code and printing process,and a fusion between the process mechanism modeling and convolutional neural network modeling.A prediction model for the process time and equipment energy consumption component power is established to enhance the accuracy and modeling efficiency of carbon emission prediction models.The proposed melting deposition forming electric energy consumption prediction model based on data-mechanism mixed-driven underwent printing experiments,and its efficacy compared with traditional mechanism models and specific energy consumption models.The feasibility and high precision of the established model were verified.Secondly,to achieve the "high quality and low-carbon" manufacturing objective,the article introduces the carbon emission factor and the coefficient of energy content per unit weight to quantify the carbon emissions produced by the material consumption during the process.Combining the process energy consumption prediction model,the melting deposition forming process carbon emissions prediction model was established.A quality regression function model of the-formed parts was established and verified through the response surface experiments and analysis of the influence of nozzle temperature,printing speed,and layer thickness on the accuracy error of the size of the formed parts in the width and height direction.The article proposes a cooperative optimization model of "high quality and low carbon," and introduces improved multi-objective grey wolf algorithm,and entropy-weighted TOPSIS-grey correlation analysis method to rank the optimization model and resolve it.Multiple sets of the process parameter combinations were used to test the solution,and it was concluded that its comprehensive performance is superior.Finally,on the basis of theoretical research,the "Melting Deposition Forming Process Carbon Emission and Parameter Optimization System" is designed and developed.The system incorporates data collection,information management,energy consumption,and carbon emission prediction model construction and measurement,s high-quality,and low-carbon target collaborative optimization,and scheme decision-making modules.The study follows the "green and sustainable development’’ concept of high-quality and low-carbon and provides new ideas and data support for research on carbon emission prediction and parameter optimization of additive manufacturing.