Enhanced Hybrid Model to Predict the Surface Roughness of Honed cylinder Bore

2021 ◽  
pp. 1-25
Author(s):  
Burhan Afzal ◽  
Xueping Zhang ◽  
Anil Srivastava
Keyword(s):  
Surface Roughness ◽  
Hybrid Model ◽  
Surface Texture ◽  
Prediction Accuracy ◽  
Analytical Models ◽  
Superior Performance ◽  
Functional Surface ◽  
Model Framework ◽  
Multiple Variables ◽  
Cylinder Bore

Abstract Cylinder bore honing is a finishing process that generates a crosshatch pattern with alternate valleys and plateaus responsible for enhancing lubrication and preventing gas and oil leakage in the engine cylinder bore. The required functional surface in the cylinder bore is generated by a sequential honing process and is characterized by Rk roughness parameters (Rk, Rvk, Rpk, Mr1, Mr2). Predicting the desired surface roughness relies primarily on two techniques: (i) analytical models (AM) and (ii) machine learning (ML) models. Both of these techniques offer certain advantages and limitations. AM's are interpretable as they indicate distinct mapping relation between input variables and honed surface texture. However, AM's are usually based on simplified assumptions to ensure the traceability of multiple variables. Consequently, their prediction accuracy is adversely impacted when these assumptions are not satisfied. However, ML models accurately predict the surface texture but their prediction mechanism is challenging to interpret. Furthermore, the ML models' performance relies heavily on the representativeness of data employed in developing them. Thus, either prediction accuracy or model interpretability suffers when AM and ML models are implemented independently. This study proposes a hybrid model framework to incorporate the benefits of AM and ML simultaneously. In the hybrid model, an Artificial neural network (ANN) compensates the AM by correcting its error. This retains the physical understanding built into the model while simultaneously enhancing the prediction accuracy. The proposed approach resulted in a hybrid model that significantly improved the prediction accuracy of the AM and additionally provided superior performance compared to independent ANN.

Honed surface multi-scale Prediction based on multi-stage honing process of engine cylinder bore

2019 ◽  
pp. 1-32
Author(s):  
Zaoyang Zhou ◽  
Xueping Zhang ◽  
Kunlun Lv ◽  
Jun Wu ◽  
Zhenqiang Yao ◽  
...  
Keyword(s):  
Surface Roughness ◽  
Surface Texture ◽  
Time Discretization ◽  
Single Step ◽  
Analytical Methodology ◽  
Duration Time ◽  
Engine Cylinder ◽  
Two Stages ◽  
Cylinder Bore ◽  
Abrasive Size

Abstract Sequential honing process is usually implemented in engine cylinder bore processing to obtain the cross-hatched surface texture with excellent function to balance lubricant storage capacities and supporting performance. Many researches have devoted to correlating honed surface quality of cylinder bore with honing process parameters by means of experiments or simulations. Quite a few efforts have addressed the effect of sequential multiple steps on the surface texture in the honing of engine cylinder bore. However, these researches cannot provide an explicit and analytical methodology to predict honed surface texture efficiently and accurately. This paper presents an analytical and explicit methodology to incorporate a proposed microscale abrasive model into the analytical simulation process of sequential honing. The proposed abrasive model synthetically considers the shape, size, posture, and position of abrasives randomly distributed in honing stone, which is incorporated into honing head motions in terms of rotation, oscillation and feeding. The kinematics of honing head is calculated by space-time discretization to capture the interaction between honing stones and cylinder bore surface. The above procedure acts as each single step for the sequential honing processes. This study investigates the sequential honing of two stages including semi-finish honing and plateau honing at different feeding speeds by applying the abrasive model with different abrasive sizes. The formation of cross-hatched surface texture was successfully achieved sequentially by semi-finish honing and plateaus honing. Then the Abbott-Firestone Curve of the honed surface can be obtained to analyze the influences of abrasive size and honing time of two stages on the surface roughness. Correctness of surface roughness predicted by the model is verified by comparing with a group of experiment measurements in terms of Abbott-Firestone Curve. Most errors of all the predicted Rk roughness family roughness parameters in the two honing stages are less than 15%. Based on the model, simulations are done to analyze the influences of abrasive size and honing duration time of two stages on the surface roughness. The result shows that the larger abrasive used in finish honing leads to the decrease of the material portions Mr1, Mr2 and the increase of the reduced valley depth Rvk. The longer plateau honing duration time is preferred to produce the larger Mr1, Mr2 and the smaller Rvk.

Simulating the Sequential Honing Process of Engine Cylinder Bore by Modeling Abrasives in Honing Stone

2019 ◽  
Author(s):  
Zaoyang Zhou ◽  
Xueping Zhang ◽  
Kunlun Lv ◽  
Jun Wu ◽  
Zhenqiang Yao ◽  
...  
Keyword(s):  
Surface Roughness ◽  
Surface Texture ◽  
Time Discretization ◽  
Single Step ◽  
Analytical Methodology ◽  
Duration Time ◽  
Engine Cylinder ◽  
Two Stages ◽  
Cylinder Bore ◽  
Abrasive Size

Abstract Sequential honing process is usually implemented in engine cylinder bore processing to obtain the cross-hatched surface texture with excellent function to balance lubricant storage capacities and supporting performance. Many researches have devoted to correlating honed surface quality of cylinder bore with honing process parameters by means of experiments or simulations. Quite a few efforts have addressed the effect of sequential multiple steps on the surface texture in the honing of engine cylinder bore. However, these researches cannot provide an explicit and analytical methodology to predict honed surface texture efficiently and accurately. This paper presents an analytical and explicit methodology to incorporate a proposed microscale abrasive model into the analytical simulation process of sequential honing. The proposed abrasive model synthetically considers the shape, size, posture, and position of abrasives randomly distributed in honing stone, which is incorporated into honing head motions in terms of rotation, oscillation and feeding. The kinematics of honing head is calculated by space-time discretization to capture the interaction between honing stones and cylinder bore surface. The above procedure acts as each single step for the sequential honing processes. This study investigates the sequential honing of two stages including semi-finish honing and plateau honing at different feeding speeds by applying the abrasive model with different abrasive sizes. The formation of cross-hatched surface texture was successfully achieved sequentially by semi-finish honing and plateaus honing. Then the Abbott-Firestone Curve of the honed surface can be obtained to analyze the influences of abrasive size and honing time of two stages on the surface roughness. Correctness of surface roughness predicted by the model is verified by comparing with a group of experiment measurements in terms of Abbott-Firestone Curve. Most errors of all the predicted Rk roughness family roughness parameters in the two honing stages are less than 15%. Based on the model, simulations are done to analyze the influences of abrasive size and honing duration time of two stages on the surface roughness. The result shows that the larger abrasive used in finish honing leads to the decrease of the material portions Mr1, Mr2 and the increase of the reduced valley depth Rvk. The longer plateau honing duration time is preferred to produce the larger Mr1, Mr2 and the smaller Rvk.

scholarly journals Surface roughness estimation for FDM systems

2021 ◽  
Author(s):  
Behnam Nourghassemi
Keyword(s):  
Surface Roughness ◽  
Hybrid Model ◽  
Least Square ◽  
Experimental Results ◽  
Analytical Models ◽  
Support Vector ◽  
Empirical Estimation ◽  
Standard Format ◽  
Svm Model ◽  
Analytical Estimation

By selecting the optimal build angle, the surface roughness of rapid prototyped parts can be minimized. The objective of this study is to develop a model for estimation of surface roughness as a function of build angle and other build parameters for parts built by Fusion Deposition Modeling (FDM) machines. For that purpose, principles of the FDM technique, along with other rapid prototyping techniques, are reviewed and various standards for surface roughness measurements are introduced. Different analytical models for the estimation of surface roughness for FDM systems, which were proposed in the literature, are reviewed and reformulated in a standard format for comparison reasons. A new hybrid model is proposed for analytical estimation of the surface roughness based on experimental results and comparison of the models. In addition, Least Square Support Vector Machine (LS-SVM) is applied for an empirical estimation of the surface roughness. Robustness of the LS-SVM model is studied and its performance is compared to the hybrid model. The experimental results confirm better results for the LS-SVM model.

scholarly journals Surface roughness estimation for FDM systems

2021 ◽  
Author(s):  
Behnam Nourghassemi
Keyword(s):  
Surface Roughness ◽  
Hybrid Model ◽  
Least Square ◽  
Experimental Results ◽  
Analytical Models ◽  
Support Vector ◽  
Empirical Estimation ◽  
Standard Format ◽  
Svm Model ◽  
Analytical Estimation

By selecting the optimal build angle, the surface roughness of rapid prototyped parts can be minimized. The objective of this study is to develop a model for estimation of surface roughness as a function of build angle and other build parameters for parts built by Fusion Deposition Modeling (FDM) machines. For that purpose, principles of the FDM technique, along with other rapid prototyping techniques, are reviewed and various standards for surface roughness measurements are introduced. Different analytical models for the estimation of surface roughness for FDM systems, which were proposed in the literature, are reviewed and reformulated in a standard format for comparison reasons. A new hybrid model is proposed for analytical estimation of the surface roughness based on experimental results and comparison of the models. In addition, Least Square Support Vector Machine (LS-SVM) is applied for an empirical estimation of the surface roughness. Robustness of the LS-SVM model is studied and its performance is compared to the hybrid model. The experimental results confirm better results for the LS-SVM model.

A Hybrid Physical Model Based Artificial Neural Network Approach to Predict Surface Texture in the Honing of Engine Cylinder Bore

2020 ◽  
Author(s):  
Burhan Afzel ◽  
Xueping Zhang ◽  
Anil K. Srivastava
Keyword(s):  
Surface Roughness ◽  
Physical Model ◽  
Hybrid Model ◽  
Combustion Engine ◽  
Learning Approaches ◽  
Neural Network Approach ◽  
Influence Of Process Parameters ◽  
Technology Improvement ◽  
Artificial Neural ◽  
Cylinder Bore

Abstract This study proposes a hybrid model that utilizes a physical model and Artificial Neural Networks (ANN) approach to predict surface roughness during cylinder bore honing with an improved prediction efficiency of 90% compared to the standalone physical model. As a critical component of internal combustion engine technology, improvement in the surface roughness of cylinder bore can significantly reduce friction, wear and oil consumption, resulting in improved engine performance. Desired surface roughness in cylinder bore is imparted by honing, which serves as the terminal process in cylinder bore manufacturing. The cylinder bore honing process consists of rough honing, fine honing, and plateau honing stage. Each stage further involves variables such as honing stone geometry, grain size, grain concentration, honing speed, pressure, feed, over travel, number of strokes, etc. In literature, different approaches have been proposed to determine the influence of process parameters on the surface roughness of the honed cylinder bore. However, these approaches have their limitations. Experimental based studies are limited by the number of parameters that can be considered, analytical analysis methods involve extensive calculations resulting in reduced computational efficiency and accuracy, while machine learning approaches require a large amount of data. To overcome these limitations, this study employs a hybrid model to investigate the evolution of roughness at the rough and fine stage of the honing process. A two-phase approach is employed; first, a physical model is used to determine the surface roughness using various parameters. Secondly, these results are applied to train the ANN that can predict surface roughness for new parameters with a difference of less than 10% from the physical model.

scholarly journals Influence of Relative Displacement on Surface Roughness in Longitudinal Turning of X37CrMoV5-1 Steel

Materials ◽  
2021 ◽  
Vol 14 (5) ◽  
pp. 1317
Author(s):  
Michal Skrzyniarz ◽  
Lukasz Nowakowski ◽  
Edward Miko ◽  
Krzysztof Borkowski
Keyword(s):  
Surface Roughness ◽  
Standard Deviation ◽  
Surface Texture ◽  
Roughness Parameter ◽  
Relative Displacement ◽  
Edge Radius ◽  
Shaping Process ◽  
Feed Value ◽  
Longitudinal Turning ◽  
Final Effect

The shaping process of surface texture is complicated and depends on many factors and phenomena accompanying them. This article presents the author’s test stand for the measurement of relative displacements in a tool–workpiece system during longitudinal turning. The aim of this study was to determine the influence of edge radius on the relative displacement between the tool and workpiece. The cutting process was carried out with inserts with different edge radii for X37CrMoV5-1 steel. As a result of the research, vibration charts of the tool–workpiece system were obtained. In the range of feed 0.03–0.18 mm/rev, the values of the standard deviation of relative displacements in the x-axis were obtained in the range of 0.36–0.78 μm for the insert with an edge radius of rn = 48.8 μm. As a result of the work, it was determined that for the feed value of 0.12 mm/rev for all inserts, the relative displacements are the smallest. As the final effect, the formula for forecasting the Ra roughness parameter was presented.

Genetic Programming based Drag Model with Improved Prediction Accuracy for Fluidization Systems

2017 ◽  
Vol 15 (2) ◽  
Author(s):  
R. R. Sonolikar ◽  
M. P. Patil ◽  
R. B. Mankar ◽  
S. S. Tambe ◽  
B. D. Kulkarni
Keyword(s):  
Pressure Drop ◽  
Genetic Programming ◽  
Drag Coefficient ◽  
Prediction Accuracy ◽  
Volume Fraction ◽  
Solid Volume Fraction ◽  
Superior Performance ◽  
Momentum Exchange ◽  
Drag Model ◽  
Parameter Ranges

Abstract The drag coefficient plays a vital role in the modeling of gas-solid flows. Its knowledge is essential for understanding the momentum exchange between the gas and solid phases of a fluidization system, and correctly predicting the related hydrodynamics. There exists a number of models for predicting the magnitude of the drag coefficient. However, their major limitation is that they predict widely differing drag coefficient values over same parameter ranges. The parameter ranges over which models possess a good drag prediction accuracy are also not specified explicitly. Accordingly, the present investigation employs Geldart’s group B particles fluidization data from various studies covering wide ranges of Re and εs to propose a new unified drag coefficient model. A novel artificial intelligence based formalism namely genetic programming (GP) has been used to obtain this model. It is developed using the pressure drop approach, and its performance has been assessed rigorously for predicting the bed height, pressure drop, and solid volume fraction at different magnitudes of Reynolds number, by simulating a 3D bubbling fluidized bed. The new drag model has been found to possess better prediction accuracy and applicability over a much wider range of Re and εs than a number of existing models. Owing to the superior performance of the new drag model, it has a potential to gainfully replace the existing drag models in predicting the hydrodynamic behavior of fluidized beds.

scholarly journals Comparison of Analytical Approaches Predicting the Compressive Strength of Fibre Reinforced Polymers

Materials ◽  
2018 ◽  
Vol 11 (12) ◽  
pp. 2517 ◽  
Author(s):  
Christian Leopold ◽  
Sergej Harder ◽  
Timo Philipkowski ◽  
Wilfried Liebig ◽  
Bodo Fiedler
Keyword(s):  
Compressive Strength ◽  
Prediction Accuracy ◽  
Volume Fraction ◽  
Fibre Volume Fraction ◽  
Experimental Results ◽  
Analytical Models ◽  
Reinforced Polymers ◽  
Reinforced Polymer ◽  
Fibre Reinforced Polymer ◽  
Analytical Approaches

Common analytical models to predict the unidirectional compressive strength of fibre reinforced polymers are analysed in terms of their accuracy. Several tests were performed to determine parameters for the models and the compressive strength of carbon fibre reinforced polymer (CFRP) and glass fibre reinforced polymer (GFRP). The analytical models are validated for composites with glass and carbon fibres by using the same epoxy matrix system in order to examine whether different fibre types are taken into account. The variation in fibre diameter is smaller for CFRP. The experimental results show that CFRP has about 50% higher compressive strength than GFRP. The models exhibit significantly different results. In general, the analytical models are more precise for CFRP. Only one fibre kinking model’s prediction is in good agreement with the experimental results. This is in contrast to previous findings, where a combined modes model achieves the best prediction accuracy. However, in the original form, the combined modes model is not able to predict the compressive strength for GFRP and was adapted to address this issue. The fibre volume fraction is found to determine the dominating failure mechanisms under compression and thus has a high influence on the prediction accuracy of the various models.

Intelligent Monitoring and Prediction of Surface Roughness in Ball-End Milling Process

2011 ◽  
Vol 121-126 ◽  
pp. 2059-2063 ◽  
Author(s):  
Somkiat Tangjitsitcharoen ◽  
Angsumalin Senjuntichai
Keyword(s):  
Surface Roughness ◽  
Cutting Force ◽  
Prediction Accuracy ◽  
End Milling ◽  
Milling Process ◽  
Ball End Milling ◽  
Roughness Model ◽  
Force Ratio ◽  
Tool Diameter ◽  
End Milling Process

In order to realize the intelligent machines, the practical model is proposed to predict the in-process surface roughness during the ball-end milling process by utilizing the cutting force ratio. The ratio of cutting force is proposed to be generalized and non-scaled to estimate the surface roughness regardless of the cutting conditions. The proposed in-process surface roughness model is developed based on the experimentally obtained data by employing the exponential function with five factors of the spindle speed, the feed rate, the tool diameter, the depth of cut, and the cutting force ratio. The prediction accuracy and the prediction interval of the in-process surface roughness model at 95% confident level are calculated and proposed to predict the distribution of individually predicted points in which the in-process predicted surface roughness will fall. All those parameters have their own characteristics to the arithmetic surface roughness and the surface roughness. It is proved by the cutting tests that the proposed and developed in-process surface roughness model can be used to predict the in-process surface roughness by utilizing the cutting force ratio with the highly acceptable prediction accuracy.

Experimental Study on the Effect of Surface Integrity on Fatigue Performance of Aero-Engine Blades

2021 ◽  
Author(s):  
Yun Huang ◽  
Shaochuan Li ◽  
Guijian Xiao ◽  
Benqiang Chen ◽  
Yi He ◽  
...  
Keyword(s):  
Residual Stress ◽  
Surface Roughness ◽  
Fatigue Life ◽  
Fatigue Strength ◽  
Fatigue Test ◽  
Surface Texture ◽  
Surface Characteristics ◽  
Service Performance ◽  
Aero Engine ◽  
Vibration Fatigue

Abstract As the core component of aero-engine, the service performance of aero-engine blade has an important influence on the engine’s reliability and safety performance. Existing studies have shown that machined surface characteristics affect the fatigue strength of components. However, current studies are all based on regular fatigue samples. The structure of blades different from fatigue samples, and the influence mechanism of structural differences on the service performance of blades is still unclear. In addition, the conventional fatigue test conditions are not representative for the blades’ actual service conditions, so it is difficult to realize the processing process for the service performance optimization. In this study, the aero-engine blades processed by abrasive belt grinding and the vibration fatigue test bench were used to explore the influence of surface roughness, surface texture, and surface residual stress on the fatigue performance of aero-engine blades under actual working conditions. The aero-engine blades were ground with different process parameters to obtain different single-factor surface characteristics. By comparing the vibration fatigue life of blades with different surface features, the influence degree of each surface feature on the fatigue life was explored. Results showed that surface roughness has the greatest influence on fatigue strength, followed by residual stress, and surface texture has the least influence on fatigue strength.

Sign in / Sign up

Export Citation Format

Share Document