Machining Experiments Research Articles

Research on Helical Electrode Electrochemical Drilling Assisted by Anode Vibration for Jet Micro-Hole Arrays on Tube Walls

The electrochemical cutting technique, utilizing electrolyte flushing through micro-hole arrays in the radial direction of a tube electrode, offers the potential for cost-effective and high-surface-integrity machining of large-thickness, straight-surface structures of difficult-to-cut materials. However, fabricating the array of jet micro-holes on the tube electrode sidewall remains a significant challenge, limiting the broader application of this technology. To enhance the efficiency and quality of machining these jet micro-holes on the tube sidewall, a helical electrode electrochemical drilling method assisted by anode vibration has been proposed. The influence of parameters, such as the rotational direction and speed of the helical electrode, as well as the vibration amplitude and frequency of the workpiece, on the machining results was investigated using fluid field simulation and machining experiments. It was found that these auxiliary movements could facilitate the renewal of electrolytes within the machining gap, thereby enhancing the efficiency and quality of electrochemical drilling. Using the optimized machining parameters, an array of 10 jet micro-holes with a diameter of 200 μm was machined on the metal tube sidewall. Electrochemical cutting with radial electrolyte flushing tests were then performed through these micro-holes.

Experimental study on the influence of lapping plate materials on the quality of both sides cylindrical rollers machining

The both-sides machining method can obtain high precision cylindrical rollers, but there is a drawback that the lapping plate is easy to wear, which restricts further improvement in the quality of rollers machining. Aiming to solve this problem, a both-sides machining method using hard ceramic lapping plate is proposed. Friction and wear experiments with different lapping plate materials, along with the corresponding comparative machining experiments, demonstrated the superior performance of Al2O3 ceramic lapping plate in terms of roundness (0.23 μm), cylindricity (0.58 μm), straightness (0.489 μm), surface roughness (11.3 nm). By studying the influence of performance indicators, the optimal process parameters for cylindrical rollers machining were obtained. The durability of the Al2O3 ceramic lapping plate was also analyzed. The study fully confirm that the hard ceramic significantly enhances lapping plate wear resistance, machining precision and consistency, which provides a new approach to ultra-precision machining of cylindrical rollers.

Unraveling pathogenesis and potential biomarkers for autism spectrum disorder associated with HIF1A pathway based on machine learning and experiment validation.

Autism spectrum disorder (ASD) is a neurodevelopmental disorder with a high social burden and limited treatments. Hypoxic condition of the brain is considered an important pathological mechanism of ASD. HIF1A is a key participant in brain hypoxia, but its contribution to the pathophysiological landscape of ASD remains unclear. ASD-related datasets were obtained from GEO database, and HIF1A-related genes from GeneCards. Co-expression module analysis identified module genes, which were intersected with HIF1A-related genes to identify common genes. Machine learning identified hub genes from intersection genes and PPI networks were constructed to explore relationships among hub and HIF1A. Single-cell RNA sequencing analyzed hub gene distribution across cell clusters. ASD mouse model was created by inducing maternal immune activation (MIA) with poly(I:C) injections, verified through behavioral tests. Validation of HIF1A pathway and hub genes was confirmed through Western Blot, qPCR, and immunofluorescence in ASD mice and microglia BV-2 cells. Using CEMiTool and GeneCards, 45 genes associated with ASD and HIF1A pathway were identified. Machine learning identified CDKN1A, ETS2, LYN, and SLC16A3 as potential ASD diagnostic markers. Single-cell sequencing pinpointed activated microglia as key immune cells. Behavioral tests showed MIA offspring mice exhibited typical ASD-like behaviors. Immunofluorescence confirmed the activation of microglia and HIF1A pathway in frontal cortex of ASD mice. Additionally, IL-6 contributed to ASD by activating JUN/HIF1A pathway, affecting CDKN1A, LYN, and SLC16A3 expression in microglia. HIF1A-related genes CDKN1A, ETS2, LYN, and SLC16A3 are strong diagnostic markers for ASD and the activation of IL-6/JUN/HIF1A pathway in microglia contributes to the pathogenesis of ASD.

Forecast assessment of water regime efficiency and changes in water consumption on drained lands of Western Polissia of Ukraine in a changing climate

The article presents forecast data on the impact of changing weather and climatic conditions on evaporation and water demand on the drained lands of Western Polissia of Ukraine in different periods of their development. These data are the basis for the development of appropriate adaptive measures aimed at reducing the negative impact of changing weather and climate conditions on drained lands and increasing the efficiency of their use. Computer simulation modeling of various climatic scenarios was planned to accomplish the objective. The forecast was made through a machining experiment based on the implementation of a corresponding complex of forecast-simulation models regarding the main regime-technological variable parameters of drainage systems, local climatic conditions, water regime, water management technologies, and productivity of drained lands for schematized natural, agrotechnical, and meliorative conditions. Vegetation values of total evaporation and water demand formation of drained lands were determined for the long-term forecast module of water supply regarding variable climatic and agro-meliorative conditions. The technological efficiency of different irrigation technologies (subsoil irrigation, sprinkler irrigation) of drained lands was evaluated. The research showed that in the Western Polissia of Ukraine, the water supply module for irrigating drained lands varies significantly depending on the type of crops, drained soil (mineral, peat), irrigation technologies, and heat and moisture supply conditions during the vegetation period. It has been experimentally determined that under current climatic conditions, the value of the water supply module is 2.9 times higher than the existing recommended design values, and under the predicted conditions of their change, this excess is 5.2 times. At the same time, the water demand of cultivated crops increases almost 2–3 times and determines the need to increase water consumption for moistening drained lands in conditions of sufficient water shortage. It has been established that the main content of the necessary adaptive measures is to increase the moisture availability of soils with regulated water regime by switching from periodic to regular moistening of them through the use of appropriate resource-saving technologies for regulating the water regime.

Grey Relationship Analysis of Cutting Forces and Surface Roughness in Turning using Cryo-treated and Untreated Cutting Tools

The aim of this experimental study is to establish the optimum heat treatment procedure and machining parameters to obtain values of cutting force and surface roughness in machining AISI 1050 workpiece by utilizing grey relations analysis based on Taguchi method. In this regard, machining operations were carried out at three different cutting speeds and three different feed rates at a constant depth of cut. Taguchi mixed-level design L18 (2^2 3^2) was chosen for machining experiments. The optimum heat treatment and machining parameters for determination of cutting force and surface roughness values were identified according to the fact that higher the Signal/Noise ratio is better based on grey relations analysis of Taguchi method. The results showed that the optimum parameters for surface roughness and cutting forces were obtained as 0.1 mm/rev at feed rate and 180 m/min at cutting speed with cryo-treated cutting tool. According to ANOVA table and Signal/Noise ratios, it was specified that the most effective parameters on cutting forces and surface roughness were feed rate, cutting speed and heat treatment conditions, respectively.

Alumina-enriched sunflower bio-oil in machining of Hastelloy C-276: a fuzzy Mamdani model-aided sustainable manufacturing paradigm

With the increasing emphasis on sustainable manufacturing practices, eco-friendly lubricants have gained significant attention to moderate the friction coefficient at the tool-work interface. In line with this, the contemporary study aimed to examine the viability of Alumina-enriched sunflower bio-oil as a metalworking fluid. Different volume fractions of Alumina nanoparticles (varying from 0 to 1 vol%) were mixed with sunflower bio-oil, and the physical properties, for instance, contact angle and dynamic viscosity, were analyzed to determine the optimal concentration of Alumina. Subsequently, machining experiments were executed on Hastelloy C-276 under various lubricating conditions, including dry cutting, compressed air, sunflower bio-oil, and 0.6 vol% Alumina-sunflower bio-oil. A comparative analysis among these lubricating mediums demonstrated that sunflower bio-oil with a 0.6 vol% Alumina concentration outperformed others, resulting in a significant reduction of surface roughness, and tool wear by 73.31%, and 82.14% respectively when compared to dry machining. Besides, the utilization of 0.6 vol% Alumina-sunflower bio-oil has demonstrated a reduction of 17.86% in total machining cost, along with reductions of 15.44% in energy consumption and carbon emissions, when compared to dry machining. Finally, a Taguchi-designed experiment consisting of sixteen trials was performed in different lubricating conditions, and a Fuzzy-Mamdani model was employed to achieve a sustainable machining environment. The sustainability assessment results indicated that a cutting speed of 75 m/min, feed of 0.05 mm/tooth, depth of cut of 0.15 mm, and the utilization of the 0.6 vol% Alumina-sunflower bio-oil resulted in the most sustainable machining environment, with the highest Multi-Performance Characteristics Index of 0.75.

A Five-Axis Toolpath Corner-Smoothing Method Based on the Space of Master–Slave Movement

The smoothing of linear toolpaths plays is critical in improving machining quality and efficiency in five-axis CNC machining. Existing corner-smoothing methods often overlook the impact of spline curvature fluctuations, which may lead to acceleration variations, hindering surface quality improvements. The paper presents a five-axis toolpath corner-smoothing method based on the space of master–slave movement (SMM), aiming to minimize curvature fluctuations in five-axis machining and improve surface quality. The concept of movement space in master–slave cooperative motion is introduced, where the tool tip position and tool orientation are decoupled into a main motion trajectory and two master–slave movement space trajectories. By deriving the curvature monotony conditions of a dual Bézier spline, a G2-continuous tool tip corner-smoothing curve with minimal curvature fluctuations is constructed in real-time. Subsequently, using the SMM and the asymmetric dual Bézier spline, a high-order continuous synchronization relationship between the tool tip position and tool orientation is established. Simulation tests and machining experiments show that with our smoothing algorithm, maximum acceleration values for each axis were reduced by 21.05%, while jerk was lowered by 22.31%. These results indicate that trajectory smoothing significantly reduces mechanical vibrations and improves surface quality.

Investigation of the Effect of Al2O3 Nanoparticle-Added MQL Lubricant on Sustainable and Clean Manufacturing

In this study, in order to improve the characteristics of the vegetable-based cutting fluids used in the MQL technique and increase the machining performance of MQL and its positive effects on sustainable manufacturing, the effects of the MQL method with nano-Al2O3 additives on surface roughness (Ra) and cutting temperature (Ctt) were examined through turning experiments carried out by adding nano-Al2O3 to the vegetable-based cutting fluid. For this purpose, machining tests were carried out on hot work tool steel alloyed with Cr-Ni-Mo that has a delivery hardness of 45 HRC. In hard machining experiments, three techniques for cooling and lubricating (dry cutting, MQL, and nano-MQL), three cutting speeds (V) (100, 130, 160 m/min), three feed rates (f) (0.10, 0.125, and 0.15 mm/rev), and two different ceramic cutting tools (uncoated and TiN-coated with PVD methods) were used as control factors. For Ra, the nano-MQL method provided an average of 21.49% improvement compared to other cooling methods. For Ctt, this rate increased to 26.7%. In crater wear areas, the nano-MQL method again exhibited the lowest wear values, decreasing performance by approximately 50%. The results of this research showed that the tests conducted using the cooling of nano-MQL approach produced the best results for all output metrics (Ra, Ctt, and crater wear).

An Interpolator for a Non-Uniform Rational B-Spline Curve with a High Efficiency and Accuracy Using Polynomial Representations

This paper introduces an efficient and accurate interpolator for NURBS (non-uniform rational B-spline) curves, addressing the challenge of regulating feedrate under machining accuracy and dynamic constraints, particularly at sensitive corners. A recursive matrix representation and polynomial conversion are utilized to enhance the computation of NURBS curve intermediate points and derivatives. An improved adaptive planning method is presented to adjust the feedrate at sensitive corners, ensuring that chord error and dynamic constraints are met. The method integrates linear acceleration and deceleration stages to mitigate abrupt changes in acceleration and jerk. Additionally, a prediction–correction scheme-based interpolator is developed, employing an asynchronous mechanism to improve computational efficiency. The proposed method’s effectiveness and correctness are validated through simulation tests and machining experiments.

Dynamic splitting tensile properties of high-strength ultrahigh-toughness cementitious composites (HS-UHTCCs)

In this study, the splitting tests were conducted to examine the effect of strain rate on the splitting tensile performance of high-strength ultrahigh-toughness cementitious composites (HS-UHTCCs). Hydraulic machine and split Hopkinson pressure bar (SHPB) were used to investigate the tensile properties of HS-UHTCCs. The strain rates ranged from 10−6 to 10−3 s−1 for hydraulic machine experiments and from 5 to 15 s−1 for SHPB tests, respectively. The splitting tensile strength, tensile stress-strain curves, dynamic increase factor (DIF) and energy absorption ability of HS-UHTCCs were analyzed. The new DIF models proposed for HS-UHTCCs fit well with the experimental data. In addition, the failure modes of specimens and fibers were investigated. The splitting tensile strain was obtained from digital image correlation (DIC) tests. The results indicate that the strain rate has a significant effect on the splitting tensile strength and energy absorption ability of HS-UHTCCs. On the contrary, the average splitting tensile strain showed a decreasing trend with the increase of strain rate. Besides, the DIF model of HS-UHTCCs was proposed with experimental data. The failure modes of specimens changed at high strain rates, transitioning from remaining intact to completely splitting into half. Two failure patterns of polyethylene (PE) fibers included pull-out failure and rupture were observed, while steel fibers experienced pull-out failure.

Machining-Induced Burr Suppression in Edge Trimming of Carbon Fibre-Reinforced Polymer (CFRP) Composites by Tool Tilting

Several challenges arise during edge trimming of carbon fibre-reinforced polymer (CFRP) composites, such as the formation of machining-induced burrs and delamination. In a recent development, appropriate-quality geometric features in CFRPs can be machined using special cutting tools and optimised machining parameters. However, these suitable technologies quickly become inappropriate due to the accelerated tool wear. Therefore, the main aim of our research was to find a novel solution for maintaining the machined edge quality even if the tool condition changed significantly. We developed a novel mechanical edge-trimming technology inspired by wobble milling, i.e., the composite plate compression is governed by the proper tool tilting. The effectiveness of the novel technology was tested through mechanical machining experiments and compared with that of conventional edge-trimming technology. Furthermore, the influences of the tool tilting angle and the permanent chamfer size on the burr characteristics were also investigated. A one-fluted solid carbide end mill with a helix angle of 0° was applied for the experiments. The machined edges were examined trough stereomicroscopy and scanning electron microscopy. The images were evaluated through digital image processing. Our results show that multi-axis edge-trimming technology produces less extensive machining-induced burrs than conventional edge trimming by an average of 50%. Furthermore, we found that the tool tilting angle has a significant impact on burr size, while permanent chamfer does not influence it. These findings suggest that multi-axis edge trimming offers a strong alternative to conventional methods, especially when using end-of-life cutting tools, and highlight the importance of selecting the optimal tool tilting angle to minimize machining-induced burrs.

Segmented remote plasma source output current tracking based on LQR improved single-neuron PID

Abstract The current plasma concentration control algorithm of plasma power supply has some problems, such as being easily disturbed by the outside world, low tracking control speed, and difficulty in adjusting the appropriate control parameters. In this paper, an improved single-neuron fuzzy PID control algorithm based on LQR is proposed. The learning process of neurons is optimized by introducing a quadratic performance index, and the scale coefficient K is adjusted by fuzzy control. The feasibility of the control algorithm and its advantages compared with other PID controllers are verified by MATLAB discrete simulation and real machine experiments. The results show that the PID control algorithm designed in this paper can reduce the steady-state deviation and overshoot, shorten the adjustment time, and eliminate the adjustment shock, thus significantly improving the processing effect.

An Investigation of the Effects of Cutting Edge Geometry and Cooling/Lubrication on Surface Integrity in Machining of Ti-6Al-4V Alloy

This investigation sought to characterize the combined influence of cutting-edge microgeometry and cooling/lubricating strategies on process thermo-mechanics and the resultant surface integrity in orthogonal machining of the Ti-6Al-4V alloy. Reverse waterfall cutting inserts were prepared with varying cutting-edge geometries, and machining experiments were conducted under cryogenic cooling with liquid nitrogen (LN2), minimum quantity lubrication (MQL), and dry machining conditions, using constant machining parameters. The induced surface integrity was characterized in terms of the developed cutting forces and through-thickness microhardness, grain morphology, dislocation generation, and residual stress formation. The experimental results revealed that the governing process physics are strongly influenced by variation in the implemented machining parameters. As a greater proportion of the cutting edge is distributed on the flank face, competing mechanical ploughing and thermal-based frictional effects both become more pronounced. Utilization of advanced cooling strategies to control cutting interface thermal gradients thus provides a processing route to generate tailored microstructures and surface integrity during the machining of this alloy.

Identification of diagnostic markers pyrodeath-related genes in non-alcoholic fatty liver disease based on machine learning and experiment validation

Non-alcoholic fatty liver disease (NAFLD) poses a global health challenge. While pyroptosis is implicated in various diseases, its specific involvement in NAFLD remains unclear. Thus, our study aims to elucidate the role and mechanisms of pyroptosis in NAFLD. Utilizing data from the Gene Expression Omnibus (GEO) database, we analyzed the expression levels of pyroptosis-related genes (PRGs) in NAFLD and normal tissues using the R data package. We investigated protein interactions, correlations, and functional enrichment of these genes. Key genes were identified employing multiple machine learning techniques. Immunoinfiltration analyses were conducted to discern differences in immune cell populations between NAFLD patients and controls. Key gene expression was validated using a cell model. Analysis of GEO datasets, comprising 206 NAFLD samples and 10 controls, revealed two key PRGs (TIRAP, and GSDMD). Combining these genes yielded an area under the curve (AUC) of 0.996 for diagnosing NAFLD. In an external dataset, the AUC for the two key genes was 0.825. Nomogram, decision curve, and calibration curve analyses further validated their diagnostic efficacy. These genes were implicated in multiple pathways associated with NAFLD progression. Immunoinfiltration analysis showed significantly lower numbers of various immune cell types in NAFLD patient samples compared to controls. Single sample gene set enrichment analysis (ssGSEA) was employed to assess the immune microenvironment. Finally, the expression of the two key genes was validated in cell NAFLD model using qRT-PCR. We developed a prognostic model for NAFLD based on two PRGs, demonstrating robust predictive efficacy. Our findings enhance the understanding of pyroptosis in NAFLD and suggest potential avenues for therapeutic exploration.

Influence of TiAlN coating thickness in dry machining of AISI 1045 steel using Finite Element simulation and experiments

High performance cutting can be accomplished using coated tools with optimal coating thickness. Therefore, Finite Element (FE) simulation and experimental investigation are employed to study the influence of thickness of TiAlN coating in the range of 1 to 4 μm during face milling of AISI 1045 steel under dry condition. FE simulation is successful in effectively investigating the influence of coating thickness in machining performance in terms of cutting forces, Von Mises stress, tool tip temperature and crater wear of uncoated as well as coated milling inserts. Simulation results have predicted cutting forces and temperature with maximum errors of 9.21 % and 12.20 %, respectively. FE simulation further demonstrates 2.100 and 1.602 μm as the depths of crater for 1 and 3 μm thick TiAlN coated inserts, respectively, which are also validated by the experiments. The experimental results demonstrate that 3 μm thick coated tool is superior in mitigating cutting forces, particularly under higher cutting speed (150 m/min), bringing down tool tip temperature (6.67 %) and machined surface roughness (18.18 %) as compared to 1 μm thick TiAlN coated milling insert. The results obtained from the current research work clearly indicate that the effect of coating thickness can be effectively investigated using the FE simulation before finalising the optimal coating thickness for the coating depositions and machining experiments.

Automated pretreatment of environmental water samples and non-targeted intelligent screening of organic compounds based on machine experiments

The complexity of environmental pollutants poses significant challenges for monitoring and analysis, especially with the emergence of numerous emerging contaminants. Traditional analysis methods rely mainly on laboratory analysis, which involves labor-intensive and time-consuming sample preparation procedures and non-target data analysis, greatly limiting the rapid detection of water organic pollutants. In this study, we designed a robot experimenter combined with GC × GC-TOFMS. By configuring self-developed automated analysis software, we established a fully automated process from sample collection to data characterization, for the analysis of organic pollutants. We validated the method with 111 organic standards compounds. The robot performed 2577 actions covering the entire workflow, from water sample collection to sample pre-treatment. The integration of mass spectrometry and related software enabled the automatic analysis of emerging hazardous contaminants, from sampling to the output of detection results. The results showed the automated process could qualitatively identify all compounds and demonstrated good linearity, low detection limits, and excellent quantitative ability within the range of 0.04–0.4 mg/L. The average recoveries of 82.89 % of the samples ranged from 70 % to 120 % (relative standard deviation (RSD) <15 %) at different spiked concentrations. This indicated that the established method could be used for non-targeted analysis of emerging contaminants in environmental water samples. We applied the method to samples from wastewater treatment plants and river sections, identifying 1,902 compounds across 26 categories, including 6 known hazardous contaminants found in all samples. The relative content of these characteristic compounds will inform whether treated wastewater meets discharge standards and aid in tracing the sources of pollutants. Therefore, the development of this fully automated machine experimental method enables real-time and online automatic analysis of organic pollutants in environmental water. The establishment of characteristic fingerprints can provide technical support for early warning and traceability of water quality.

Towards cleaner machining: Experimental investigation of collagen-in-water as a novel type of metalworking fluids – A technical feasibility study

Driven by environmental challenges, health concerns and increasing legislative restrictions, a high demand for research regarding alternative metalworking fluids (MWFs) to substitute the conventional mineral oil-containing fluids emerges. In recent years, bio-based fluids have continuously gained attention, as their usage could mitigate the disadvantages of conventionally used fluids, which usually contain mineral oil. Against this background, this research aims to contributing towards biological MWFs by investigating the technical applicability of collagen-in-water fluids through evaluations and comparisons with conventional MWFs. The analyzed bio-based MWF is based on collagen-hydrolysate powder dissolved in water, with collagen acting as a viscosity-increasing additive. The prepared fluids were investigated regarding their viscosity and tribological behavior in cylinder-on-ring experiments as well as tapping-torque tests. Afterwards, first machining experiments were performed for honing and grinding of hardened steel. Force-led cylinder-on-ring experiments show a significant wear reduction, e.g. a value of 13.4 mm2 at 3 wt.% collagen compared to 33.0 mm2 with pure distilled water. Wear further decreased with increasing collagen content. At 3 wt.% and again compared to pure water, tapping torque was reduced by 29 % to 181.7 Ncm, no further torque reduction was observed with increasing collagen concentration. In honing, roughness was reduced using collagen-in-water, however, an 8 wt.% collagen fluid resulted in an 18 % reduction in material removal rate compared to honing oil. In internal grinding, an 8 wt.% collagen fluid achieved an Ra of approximately 0.8 μm, compared to around 0.55 μm with emulsion. The application results demonstrate a high potential for collagen-in-water fluids, however, adjustments are necessary for obtaining competitive fluids.

Decoding Multilingual Moral Preferences: Unveiling LLM's Biases through the Moral Machine Experiment

Large language models (LLMs) increasingly find their way into the most diverse areas of our everyday lives. They indirectly influence people's decisions or opinions through their daily use. Therefore, understanding how and which moral judgements these LLMs make is crucial. However, morality is not universal and depends on the cultural background. This raises the question of whether these cultural preferences are also reflected in LLMs when prompted in different languages or whether moral decision-making is consistent across different languages. So far, most research has focused on investigating the inherent values of LLMs in English. While a few works conduct multilingual analyses of moral bias in LLMs in a multilingual setting, these analyses do not go beyond atomic actions. To the best of our knowledge, a multilingual analysis of moral bias in dilemmas has not yet been conducted. To address this, our paper builds on the moral machine experiment (MME) to investigate the moral preferences of five LLMs, Falcon, Gemini, Llama, GPT, and MPT, in a multilingual setting and compares them with the preferences collected from humans belonging to different cultures. To accomplish this, we generate 6500 scenarios of the MME and prompt the models in ten languages on which action to take. Our analysis reveals that all LLMs inhibit different moral biases to some degree and that they not only differ from the human preferences but also across multiple languages within the models themselves. Moreover, we find that almost all models, particularly Llama 3, divert greatly from human values and, for instance, prefer saving fewer people over saving more.

Finding middle grounds for incoherent horn expressions: the moral machine case

Smart devices that operate in a shared environment with people need to be aligned with their values and requirements. We study the problem of multiple stakeholders informing the same device on what the right thing to do is. Specifically, we focus on how to reach a middle ground among the stakeholders inevitably incoherent judgments on what the rules of conduct for the device should be. We formally define a notion of middle ground and discuss the main properties of this notion. Then, we identify three sufficient conditions on the class of Horn expressions for which middle grounds are guaranteed to exist. We provide a polynomial time algorithm that computes middle grounds, under these conditions. We also show that if any of the three conditions is removed then middle grounds for the resulting (larger) class may not exist. Finally, we implement our algorithm and perform experiments using data from the Moral Machine Experiment. We present conflicting rules for different countries and how the algorithm finds the middle ground in this case.

Exploring the CO2 conversion activated by the dielectric barrier discharge plasma assisted with photocatalyst via machine learning

DBD plasma assisted with halide perovskite photocatalysts is very attractive for the conversion of CO2 into high value-added products under mild conditions. However, it is still challenging to further effectively improve the corresponding CO2 conversion efficiency due to the consumption of time, materials, equipments via traditional experimental approach. In this paper, we simulate the CO2 conversion process under the DBD plasma assisted with different photocatalysts by machine learning. The process conditions and material parameters were selected as the feature variables of the machine learning model, and all the features were screened by combining Pearson's correlation coefficient and MIR ranking. The regression algorithms were evaluated using K-fold cross-validation combined with the coefficient of determination (R2), while SVR and BPANN with the best prediction performance for CO2 conversion ratio and energy efficiency were selected to establish the prediction models. In order to increase the accuracy of predictions, the hyperparameters of these two base models were improved using genetic algorithm, particle swarm optimization and Bayesian optimization. The R2 of the GA-SVR-CO2 and Bayesian-BPANN-EE reached 0.8960 and 0.9679 for the testing sets, respectively, and their practical applications were successfully verified. In addition, the effects of feature parameters on the conversion efficiency were revealed by SHAP. SEI scatter plots and Spearman correlation coefficient have been used for correlation analysis to better explore the correlation between machine learning models and experiments. This work brings new insights into the contributions of multiple parameters in the plasma system assisted with photocatalyst for efficient conversion of CO2.