Advancements in machine learning for material design and process optimization in the field of additive manufacturing

Additive manufacturing technology is highly regarded due to its advantages,such as high precision and the ability to address complex geometric challenges.However,the development of additive manufacturing process is constrained by issues like unclear fundamental principles,complex experimental cycles,and high costs.Machine learning,as a novel artificial intelligence technology,has the potential to deeply engage in the development of additive manufacturing process,assisting engineers in learning and developing new techniques.This paper provides a comprehensive overview of the research and applications of machine learning in the field of additive manufacturing,particularly in model design and process development.Firstly,it introduces the background and significance of machine learning-assisted design in additive manufacturing process.It then further delves into the application of machine learning in additive manufacturing,focusing on model design and process guidance.Finally,it concludes by summarizing and forecasting the development trends of machine learning technology in the field of additive manufacturing.

Cookie Baking Process Optimization and Quality Analysis Based on Food 3D Printing

In order to obtain better quality cookies, food 3D printing technology was employed to prepare cookies. The texture, color, deformation, moisture content, and temperature of the cookie as evaluation indicators, the influences of baking process parameters, such as baking time, surface heating temperature and bottom heating temperature, on the quality of the cookie were studied to optimize the baking process parameters. The results showed that the baking process parameters had obvious effects on the texture, color, deformation, moisture content, and temperature of the cookie. All of the roasting surface heating temperature, bottom heating temperature and baking time had positive influences on the hardness, crunchiness, crispiness, and the total color difference(ΔE) of the cookie. When the heating temperatures of the surfac and bottom increased, the diameter and thickness deformation rate of the cookie increased. However,with the extension of baking time, the diameter and thickness deformation rate of the cookie first increased and then decreased. With the surface heating temperature of 180 ℃, the bottom heating temperature of 150 ℃, and baking time of 15 min, the cookie was crisp and moderate with moderate deformation and uniform color. There was no burnt phenomenon with the desired quality. Research results provided a theoretical basis for cookie manufactory based on food 3D printing technology.

Recovery of iron from copper tailings via low-temperature direct reduction and magnetic separation: process optimization and mineralogical study

Currently, the majority of copper tailings are not effectively developed. Worldwide, large amounts of copper tailings generated from copper production are continuously dumped, posing a potential environmental threat. Herein, the recovery of iron from copper tailings via low-temperature direct reduction and magnetic separation was conducted; process optimization was carried out, and the corresponding mineralogy was investigated. The reduction time, reduction temperature, reducing agent (coal), calcium chloride additive, grinding time, and magnetic field intensity were examined for process optimization. Mineralogical analyses of the sample, reduced pellets, and magnetic concentrate under various conditions were performed by X-ray diffraction, optical microscopy, and scanning electron microscopy-energy-dispersive X-ray spectrometry to elucidate the iron reduction and growth mechanisms. The results indicated that the optimum parameters of iron recovery include a reduction temperature of 1150A degrees C, a reduction time of 120 min, a coal dosage of 25%, a calcium chloride dosage of 2.5%, a magnetic field intensity of 100 mT, and a grinding time of 1 min. Under these conditions, the iron grade in the magnetic concentrate was greater than 90%, with an iron recovery ratio greater than 95%.

Process Optimization for AZ91 Mg-alloy Low-pressure EPC Process

The influence of a key process variable on the mold filling characteristics of AZ91 Mg-alloy was studied in the low pressure EPC process.The applied flow quantity of insert gas from 1 to 5 m~3/h associated with the pressurizing rate in the low pressure EPC casting process was considered for rectangle and L-shape plate casting. The experimental results show that there is an optimal flow quantity of insert gas for good mold filling characteristics in AZ91 Mg-alloy low-pressure EPC process. The optimal flow quantity of insert gas for the specimens is 3 to 4 m~3/h. Either less or higher than the optimal flow quantity of insert gas would lead to misrun defects or folds, blisters and porosity defects. The practice of hub casting confirmed that the low-pressure EPC process with an optimal processing variable exemplified as 4 m~3/h gas flow quantity was capable of producing complicated magnesium castings without misrun defects.

MICROSTRVCTURE SIMULATION AND PROCESS OPTIMIZATION OF TURBINE BLADE CASTINGS

The probabilistic modeling is applied to calculate microstructural features of the thin complex smprolloy turbine blades cast by the vacuum investment process. The random distribution, orientation and physical mechanism of the nucleation, the growth kinetics of dendrites and the columnar-to-equiaxed transition (CET) are considered.Capitalizing on these simulating schemes, the comprehensive influence of key process variables on the scale and uniformity of grains has been involved quantitatively. The validity of the modeling is confirmed by selection of the optimum process variables.

From statistical analysis to process optimization during cladding using a Nd:YAG laser

Application of statistical methods to optimize the process parameters was achieved by employing full factorial design of experiments,which was accomplished by cladding using stepwise ramped laser power.The correlations between clad geometry and dilution(clad characteristics)and the main process parameters laser power(P_(l)),cladding speed(v_(c)),the powder feed rate(m)were obtained through application of variance analysis technique(ANOVA).The obtained correlations between the main processing parameters and the clad characteristics are discussed and a statistical model was developed.The desirability investigations using the developed statistical model were performed by considering the clad geometry,aspect ratio,dilution and hardness.Optimal parameters for cladding Stellite 6 on AISI 420 steel substrate and for cladding Nucalloy 488V on S355 J2 steel substrate were obtained.The optimal processing parameters can be applied to clad other materials with similar chemical compositions.

Process Optimization of RTS Technology for Ultra-Low Sulfur Diesel

The RTS technology can produce ultra-low sulfur diesel at lower costs using available hydrogenation catalyst and device.However,with the increase of the mixing proportion of secondary processed diesel fuel in the feed,the content of nitrogen compounds and polycyclic aromatic hydrocarbons in the feed increased,leading to the acceleration of the deactivation rate of the primary catalyst and the shortening of the service cycle.In order to fully understand the reason of catalyst deactivation,the effect of mixing secondary processed diesel fuel oil on the operating stability of the catalyst in the first reactor was investigated in a medium-sized fixed-bed hydrogenation unit.The results showed that the nitrogen compounds mainly affected the initial activity of the catalyst,but had little effect on the stability of the catalyst.The PAHs had little effect on the initial activity of the catalyst,but could significantly accelerate the deactivation of the catalyst.Combined with the analysis of the reason of catalyst deactivation and the study of RTS technology,the direction of RTS technology process optimization was put forward,and the stability of catalyst was improved obviously after process optimization.

Heat treatment process optimization for face gears based on deformation and residual stress control

In this paper, based on the principle of heat transfer and thermal elastic-plastic theory, the heat treatment process optimization scheme for face gears is proposed according to the structural characteristics of the face gear and material properties of 12Cr2Ni4 steel. To simulate the effect of carburizing and quenching process on tooth deformation and residual stress distribution, a heat treatment analysis model of face gears is established, and the microstructure, stress and deformation of face gear teeth changing with time are analyzed. The simulation results show that face gear tooth hardness increases, tooth surface residual compressive stress increases and tooth deformation decreases after heat treatment process optimization. It is beneficial to improving the fatigue strength and performance of face gears.

Design Process Optimization Based on Design Process Gene Mapping

The idea of genetic engineering is introduced into the area of product design to improve the design efficiency. A method towards design process optimization based on the design process gene is proposed through analyzing the correlation between the design process gene and characteristics of the design process. The concept of the design process gene is analyzed and categorized into five categories that are the task specification gene, the concept design gene, the overall design gene, the detailed design gene and the processing design gene in the light of five design phases. The elements and their interactions involved in each kind of design process gene signprocess gene mapping is drawn with its structure disclosed based on its function that process gene.

Optimization of Hydrocracking Process for Enhanced BTX Production from Coal Tar-Derived Hydrorefined Products

Hydroconversion of coal tar to produce aromatic hydrocarbons(BTX)represents a crucial strategy for the highvalue hierarchical utilization of coal.This study focused on the hydrocracking of hydrorefined products derived from coal tar to enhance the production of benzene,toluene,and xylene(BTX).Various reaction conditions,including reaction temperature,hydrogen pressure,space velocity,and hydrogen-to-oil volume ratio,were systematically explored to optimize BTX yields while also considering the process’s economic feasibility.The results indicate that increasing the reaction temperature from 360℃ to 390℃ significantly favors the production of BTX,with yields increasing from 21.42%to 41.14%.Similarly,an increase in hydrogen pressure from 4 MPa to 6 MPa boosts BTX production,with yields rising from 36.31%to 41.14%.Reducing the space velocity from 2 h^(-1) to 0.5 h^(-1) also favors the BTX production process,with yields increasing from 37.96%to 45.13%.Furthermore,raising the hydrogen-to-oil volume ratio from 750 to 1500 improves BTX yields from 41.61%to 45.44%.Through economic analysis,the optimal conditions for BTX production were identified as a reaction temperature of 390℃,hydrogen pressure of 5-6 MPa,space velocity of 1 h^(-1),and hydrogen-to-oil volume ratio of 1000,achieving a BTX yield of 43.73%.This investigation highlights the importance of a holistic evaluation of hydrocracking conditions to optimize BTX production.Furthermore,the findings offer valuable insights for the design and operation of industrial hydrocracking processes aimed at efficiently converting coal tar-derived hydrorefined feedstock into BTX.

Mechanism of Rosae Rugosae Flos fl avonoids in the treatment of hyperlipidemia and optimization of extraction process based on network pharmacology

This study aims to identify a natural plant chemical with hypolipidemic effects that can be used to treat high cholesterol without adverse reactions.Through network pharmacology screening,it was found that Rosae Rugosae Flos(RF)flavonoids had potential therapeutic effects on hyperlipidemia and its mechanism of action was discussed.TCMSP and GeneCards databases were used to obtain active ingredients and disease targets.Venn diagrams were drawn to illustrate the findings.The interaction network diagram was created using Cytoscape 3.8.0 software.The PPI protein network was constructed using String.GO and KEGG enrichment analysis was performed using Metascape.The results revealed 2 active flavonoid ingredients and 60 potential targets in RF.The key targets,including CCL2,PPARG,and PPARA,were found to play a role in multiple pathways such as the AGE-RAGE signaling pathway,lipid and atherosclerosis,and cancer pathway in diabetic complications.The solvent extraction method was optimized for efficient flavonoid extraction based on network pharmacology prediction results.This was achieved through a single factor and orthogonal test,resulting in an optimum process with a reflux time of 1.5 h,a solid-liquid ratio of 1:13 g/mL,and an ethanol concentration of 50%.

Optimization of the Pretreatment of the Mixture of Cassava Peelings and Pineapple Fibers Using Response Surface Methodology and a Process Simulator for the Bioethanol Production

The increase in oil prices and greenhouse gas emissions has led to the search for substitutes for fossil fuels. In Cameroon, the abundance of lignocellulosic resources is inherent to agricultural activity. Production of bioethanol remains a challenge given the crystallinity of cellulose and the presence of the complex. The pretreatment aimed to solubilize the lignin fraction and to make cellulose more accessible to the hydrolytic enzymes, was done using the organosolv process. A mathematical modeling was performed to point out the effect of the temperature on the kinetics of the release of the reducing sugars during the pretreatment. Two mathematical model was used, SAEMAN’s model and Response surface methodology. The first show that the kinetic parameters of the hydrolysis of the cellulose and reducing sugar are: 0.05089 min-1, 5358.1461 J·mol-1, 1383.03691 min-1, 51577.6100 J·mol-1 respectively. The second model was used. Temperature is the factor having the most positive influence whereas, ethanol concentration is not an essential factor. To release the maximum, an organosolv pre-treatment of this sub-strate should be carried out at 209.08°C for 47.60 min with an ethanol-water ratio of 24.02%. Organosolv pre-treatment is an effective process for delignification of the lignocellulosic structure.

Design and optimization of purification process of sinomenine hydrochloride

Sinomenine hydrochloride is generally produced from Caulis Sinomenii. At present, the purification process in industrial production suffers from large amount of solid waste, high solvent toxicity, and low sinomenine hydrochloride yield. In this study, a new purification process for sinomenine hydrochloride was proposed by using the extract obtained from acid extraction of Caulis Sinomenii as the starting material.The process included the following steps: alkalization, extraction, water washing, acid–water stripping,drying, and crystallization. 1-Heptanol was used as the extractant. The distribution coefficients of sinomenine and sinomenine hydrochloride in 1-heptanol–water system were 27.4 and 0.0167, respectively.The dissociation constants of sinomenine hydrochloride were 8.27 and 11.24, respectively. Process parameters of the new purification process were optimized with experimental design. The extractant1-heptanol and sinomenine hydrochloride in the crystallization mother solution can be recycled in the new process. The purity of the obtained sinomenine hydrochloride crystals exceeded 85%, and the yield was about 70%. Compared with current industrial processes, safer extractant, less solid waste, and higher sinomenine hydrochloride yield can be achieved using the new purification process of sinomenine hydrochloride provided in this study.

Stage identification and process optimization for fast drilling EDM of film cooling holes using KBSI method

Fast drilling electrical discharge machining(EDM)is widely used in the manufacture of film cooling holes of turbine blades.However,due to the various hole orientations and severe electrode wear,it is relatively intricate to accurately and timely identify the critical moments such as breakout,hole completion in the drilling process,and adjust the machining strategy properly.Existing breakout detection and hole completion determination methods are not suitable for the high-efficiency and fully automatic production of film cooling holes,for they almost all depend on preset thresholds or training data and become less appropriate when machining condition changes.As the breakout and hole completion detection problems can be abstracted to an online stage identification problem,in this paper,a kurtosis-based stage identification(KBSI)method,which uses a novel normalized kurtosis to denote the recent changing trends of gap voltage signals,is developed for online stage identification.The identification accuracy and generalization ability of the KBSI method have been verified in various machining conditions.To improve the overall machining efficiency,the influence of servo control parameters on machining efficiency of each machining stage was analyzed experimentally,and a new stage-wise adaptive control strategy was then proposed to dynamically adjust the servo control parameters according to the online identification results.The performance of the new strategy is evaluated by drilling film cooling holes at different hole orientations.Experimental results show that with the new control strategy,machining efficiency and the machining quality can be significantly improved.

A Review of Process Optimization for Additive Manufacturing Based on Machine Learning

Additive manufacturing(AM),also known as 3D printing,has emerged as a groundbreaking technology that has transformed the manufacturing industry.Its ability to produce intricate and customized parts with remarkable speed and reduced material waste has revolutionized traditional manufacturing approaches.However,the AM process itself is a complex and multifaceted undertaking,with various parameters that can significantly influence the quality and efficiency of the printed parts.To address this challenge,researchers have explored the integration of machine learning(ML)techniques to optimize the AM process.This paper presents a comprehensive review of process optimization for additive manufacturing based on machine learning,highlighting the recent advancements,methodologies,and challenges in this field.

Optimization of Finishing Process and Energy Savings in Denim Textile Facility

Denim is widely accepted among exported textile products due to its aesthetics, appearance, and fashion. Practitioners employed several physical or chemical treatments to improve denim qualities in denim finishing operations. So, several treatment processes, including enzymatic, bleaching, singeing, heat set, and ozone finish, are used, which made this processing more energy consumption and time-consuming. Therefore, it is significant to investigate how changing the chemicals and raw ingredients could improve the finishing process, which is environmentally and economically beneficial for sustainable production practices in the denim finishing process. This study’s research design comprises an experimental investigation in a denim plant in Bangladesh. Two different fabrics were chosen to analyze, determining the potential savings of finishing on the denim fabrics’ performance characteristics. By deducting singeing and heat-set processes, the researchers ran an experimental process by maintaining the same length of fabric. Then, the impacts of finishing process optimization on the mechanical, thermal, and comfort parameters of drape, stiffness, and tear strength were examined. The study’s findings demonstrated that this experiment increased productivity and reduced the finishing unit’s energy consumption without compromising the denim fabrics’ quality. This study significantly impacts environmental sustainability by preserving limited energy resources and manufacturing denim finishing processes.

Multi-objective optimization of process parameters for ultra-narrow gap welding based on Universal Kriging and NSGA Ⅱ

The successful confinement of the arc by the flux band depends on the welding process parameters for achieving single-pass,multi-layer, and ultra-narrow gap welding. The sidewall fusion depth, the width of the heat-affected zone, and the line energy are utilized as comprehensive indications of the quality of the welded joint. In order to achieve well fusion and reduce the heat input to the base metal.Three welding process characteristics were chosen as the primary determinants, including welding voltage, welding speed, and wire feeding speed. The metamodel of the welding quality index was built by the orthogonal experiments. The metamodel and NSGA-Ⅱ(Non-dominated sorting genetic algorithm Ⅱ) were combined to develop a multi-objective optimization model of ultra-narrow gap welding process parameters. The results showed that the optimized welding process parameters can increase the sidewall fusion depth, reduce the width of the heataffected zone and the line energy, and to some extent improve the overall quality of the ultra-narrow gap welding process.

Modeling of an Automatic Optimization System of Cyanide Concentration in Carbon in Leach for Optimal Ore Processing in a Mining Company

The optimization system, which was the subject of our study, is an autonomous chain for the automatic management of cyanide consumption. It is in the phase of industrial automation which made it possible to use the machines in order to reduce the workload of the worker while keeping a high productivity and a quality in great demand. Furthermore, the use of cyanide in leaching tanks is a necessity in the gold recovery process. This consumption of cyanide must be optimal in these tanks in order to have a good recovery while controlling the concentration of cyanide. Cyanide is one of the most expensive products for mining companies. On a completely different note, we see huge variations during the addition of cyanide. Following a recommendation from the metallurgical and operations teams, the control team carried out an analysis of the problem while proposing a solution to reduce the variability around plus or minus 10% of the addition setpoint through automation. It should be noted that this automatic optimization by monitoring the concentration of cyanide, made use of industrial automation which is a technique which ensures the operation of the ore processing chain without human intervention. In other words, it made it possible to substitute a machine for man. So, this leads us to conduct a study on concentration levels in the real world. The results show that the analysis of the modeling of the cyanide consumption optimization system is an appropriate solution to eradicate failures in the mineral processing chain. The trend curves demonstrate this resolution perfectly.

Optimization of the Formulation Process of Glucosamine Chondroitin Sulfate Tablets

[Objectives]This study was conducted to optimize the Formulation Process of glucosamine chondroitin sulfate tablets. [Methods] The orthogonal design with three levels was carried out with microcrystalline cellulose, calcium hydrophosphate and cross-linked polyvinylpyrrolidone as three factors to optimize the preparation process. [Results] When microcrystalline cellulose 200 mg/tablet, calcium hydrophosphate 150 mg/tablet, and cross-linked polyvinylpyrrolidone 80 mg/tablet were added, the angle of repose could meet the requirements of tablet pressing, and the dissolution could reach more than 95% in 30 min. The results of the orthogonal test showed that the dissolution effect of self-made tablets was faster than that of commercial products. [Conclusions] The glucosamine hydrochloride chondroitin sulfate tablets prepared by this prescription have better quality.

Process Optimization and Reaction Mechanism of Removing Copper From an Fe-Rich Pyrite Cinder Using Chlorination Roasting

The aim is to remove copper from a pyrite cinder by optimizing the chlorination roasting process using re-sponse surface methodology （RSM） and the reaction mechanism of chlorination roasting based on thermodynamic calculation was discussed. A quadratic model was suggested by RSM to correlate the key parameters, namely, dos-age of chlorinating agent, roasting temperature and roasting time to the copper volatilization ratio. The results indi- cate that the model is well consistent with the experimental data at a correlation coefficient （R2） of 0.95, and the dosage of chlorinating agent and roasting temperature both have significant effects on the copper volatilization ratio. However, a roasting temperature exceeding 1170 ~C decreases the volatilization ratio. The optimum conditions for removing copper from the cinder were identified as chlorinating agent dosage at 5%, roasting temperature at i155.10 ℃ and roasting time of 10 min; under Such a conditiom a copper volatilization ratid of 95.16% Was a- chieved from the cinder. Thermodynamic calculation shows that SiO2 in the pellet plays a key role in the chlorine re-lease from calcium chloride, and the chlorine release reactions cannot occur without it.