Intelligent Modeling of Complex Manufacturing Processes Using Hierarchical Fuzzy Basis Function Networks

2001 ◽  
Author(s):  
Cheol W. Lee ◽  
Taejoon Choi ◽  
Yung C. Shin
Keyword(s):  
Experimental Data ◽  
Basis Function ◽  
Process Models ◽  
Analytical Models ◽  
Manufacturing Processes ◽  
Suitable Model ◽  
Training Algorithm ◽  
Intelligent Modeling ◽  
The Hierarchical Structure ◽  
Model Structures

Abstract This paper presents a generalized modeling approach to modeling of complex manufacturing processes. Fuzzy basis function networks with a novel training algorithm are used to capture the cause-effect relationships of complex manufacturing processes. The modeling scheme allows for utilization of the existing knowledge in the form of analytical models, experimental data and heuristic rules in developing a suitable model. The method is implemented for the surface grinding processes based on the hierarchical structure of fuzzy basis function networks proposed by Lee and Shin [21]. Process models for surface roughness and residual stress are developed based on the available grinding model structures with a small number of experimental data to demonstrate the concept. The accuracy of developed models is validated through independent sets of grinding experiments.

Modeling of Complex Manufacturing Processes by Hierarchical Fuzzy Basis Function Networks With Application to Grinding Processes

2004 ◽  
Vol 126 (4) ◽  
pp. 880-890 ◽  
Author(s):  
Cheol W. Lee ◽  
Yung C. Shin
Keyword(s):  
Least Squares ◽  
Hierarchical Structure ◽  
Basis Function ◽  
Least Squares Method ◽  
Fuzzy Neural Networks ◽  
Process Models ◽  
Manufacturing Processes ◽  
Training Algorithm ◽  
Modeling Framework ◽  
Grinding Processes

A framework for modeling complex manufacturing processes using fuzzy neural networks is presented with a novel training algorithm. In this study, a hierarchical structure that consists of fuzzy basis function networks (FBFN) is proposed to construct comprehensive models of the complex processes. A new adaptive least-squares (ALS) algorithm, based on the least-squares method and genetic algorithm (GA), is proposed for autonomous learning and construction of FBFNs without any human intervention. Simulation studies are performed to demonstrate advantages of the proposed modeling framework with the training algorithm in modeling complex manufacturing processes. The proposed method is implemented for the surface grinding processes based on the hierarchical structure of FBFNs. Process models for surface roughness and residual stress are developed based on the available grinding model structures with a small number of experimental data to demonstrate the concept. The accuracy of developed models is validated through independent sets of grinding experiments.

Analysis and design of suitable model structures for activated sludge tanks with circulating flow

1999 ◽  
Vol 39 (4) ◽  
pp. 55-60 ◽  
Author(s):  
J. Alex ◽  
R. Tschepetzki ◽  
U. Jumar ◽  
F. Obenaus ◽  
K.-H. Rosenwinkel
Keyword(s):  
Activated Sludge ◽  
Large Scale ◽  
Wastewater Treatment Plants ◽  
Treatment Plant ◽  
High Sensitivity ◽  
Suitable Model ◽  
Model Structure ◽  
Analysis And Design ◽  
Hydraulic Behaviour ◽  
Model Structures

Activated sludge models are widely used for planning and optimisation of wastewater treatment plants and on line applications are under development to support the operation of complex treatment plants. A proper model is crucial for all of these applications. The task of parameter calibration is focused in several papers and applications. An essential precondition for this task is an appropriately defined model structure, which is often given much less attention. Different model structures for a large scale treatment plant with circulation flow are discussed in this paper. A more systematic method to derive a suitable model structure is applied to this case. Results of a numerical hydraulic model are used for this purpose. The importance of these efforts are proven by a high sensitivity of the simulation results with respect to the selection of the model structure and the hydraulic conditions. Finally it is shown, that model calibration was possible only by adjusting to the hydraulic behaviour and without any changes of biological parameters.

Simulation and prediction of optical characterization of casting (Adamantan-Fulgide) thin films using (CTANNs) approach

2019 ◽  
Vol 33 (11) ◽  
pp. 1950093 ◽  
Author(s):  
A. M. A. EL-Barry ◽  
D. M. Habashy
Keyword(s):  
Experimental Data ◽  
Thin Films ◽  
Back Propagation ◽  
Optical Characterization ◽  
Experimental Information ◽  
Training Algorithm ◽  
Energy Gaps ◽  
Resilient Back Propagation ◽  
Good Agreement

For reinforcement, the photochromic field and the cooperation between the theoretical and experimental branches of physics, the computational, theoretical artificial neural networks (CTANNs) and the resilient back propagation (R[Formula: see text]) training algorithm were used to model optical characterizations of casting (Admantan-Fulgide) thin films with different concentrations. The simulated values of ANN are in good agreement with the experimental data. The model was also used to predict values, which were not included in the training. The high precision of the model has been constructed. Moreover, the concentration dependence of both the energy gaps and Urbach’s tail were, also tested. The capability of the technique to simulate the experimental information with best accuracy and the foretelling of some concentrations which is not involved in the experimental data recommends it to dominate the modeling technique in casting (Admantan-Fulgide) thin films.

Small Horizontal Axis Wind Turbine: Analytical Blade Design and Comparison With RANS-Prediction and First Experimental Data

2013 ◽  
Author(s):  
Tom Gerhard ◽  
Michael Sturm ◽  
Thomas H. Carolus
Keyword(s):  
Experimental Data ◽  
Flow Field ◽  
Wind Turbine ◽  
Horizontal Axis ◽  
Analytical Models ◽  
Power Coefficient ◽  
Analysis Tool ◽  
Computational Domain ◽  
Horizontal Axis Wind Turbine ◽  
Data Set

State-of-the-art wind turbine performance prediction is mainly based on semi-analytical models, incorporating blade element momentum (BEM) analysis and empirical models. Full numerical simulation methods can yield the performance of a wind turbine without empirical assumptions. Inherent difficulties are the large computational domain required to capture all effects of the unbounded ambient flow field and the fact that the boundary layer on the blade may be transitional. A modified turbine design method in terms of the velocity triangles, Euler’s turbine equation and BEM is developed. Lift and drag coefficients are obtained from XFOIL, an open source 2D design and analysis tool for subcritical airfoils. A 3 m diameter horizontal axis wind turbine rotor was designed and manufactured. The flow field is predicted by means of a Reynolds-averaged Navier-Stokes simulation. Two turbulence models were utilized: (i) a standard k-ω-SST model, (ii) a laminar/turbulent transition model. The manufactured turbine is placed on the rooftop of the University of Siegen. Three wind anemometers and wind direction sensors are arranged around the turbine. The torque is derived from electric power and the rotational speed via a calibrated grid-connected generator. The agreement between the analytically and CFD-predicted kinematic quantities up- and downstream of the rotor disc is quite satisfactory. However, the blade section drag to lift ratio and hence the power coefficient vary with the turbulence model chosen. Moreover, the experimentally determined power coefficient is considerably lower as predicted by all methods. However, this conclusion is somewhat preliminary since the existing experimental data set needs to be extended.

scholarly journals Approximate Analytical Models of Shock-Wave Structure at Steady Mach Reflection

Fluids ◽  
2021 ◽  
Vol 6 (9) ◽  
pp. 305
Author(s):  
Mikhail V. Chernyshov ◽  
Karina E. Savelova ◽  
Anna S. Kapralova
Keyword(s):  
Experimental Data ◽  
Shock Wave ◽  
Comparative Analysis ◽  
Analytical Model ◽  
Mach Reflection ◽  
Supersonic Jet ◽  
Wave Structure ◽  
Analytical Models ◽  
Analytical Prediction ◽  
Shock Wave Structure

In this study, we obtain the comparative analysis of methods of quick approximate analytical prediction of Mach shock height in planar steady supersonic flows (for example, in supersonic jet flow and in narrowing channel between two wedges), that are developed since the 1980s and being actively modernized now. A new analytical model based on flow averaging downstream curved Mach shock is proposed, which seems more accurate than preceding models, comparing with numerical and experimental data.

scholarly journals Using Bayesian Model Selection to Advise Neutron Reflectometry Analysis from Langmuir-Blodgett Monolayers

2019 ◽  
Author(s):  
Andrew McCluskey ◽  
Tom Arnold ◽  
Joshaniel F. K. Cooper ◽  
Tim Snow
Keyword(s):  
Experimental Data ◽  
Soft Matter ◽  
General Framework ◽  
Neutron Reflectometry ◽  
Analytical Models ◽  
Colloid Interface ◽  
X Ray ◽  
Langmuir Blodgett ◽  
Generic Framework

The analysis of neutron and X-ray reflectometry data is important for the study of interfacial soft matter structures. However, there is still substantial discussion regarding the analytical models<br>that should be used to rationalise relflectometry data. In this work, we outline a robust and generic framework for the determination of the evidence for a particular model given experimental data, by<br>applying Bayesian logic. We apply this framework to the study of Langmuir-Blodgett monolayers by considering three possible analytical models from a recently published investigation [Campbell et al., J. Colloid Interface Sci, 2018, 531, 98]. From this, we can determine which model has the most evidence given the experimental data, and show the effect that different isotopic contrasts of neutron reflectometry will have on this. We believe that this general framework could become an important component of neutron and X-ray reflectometry data analysis, and hope others more regularly consider the relative evidence for their analytical models.<br>

Subsurface Deformation in Surface Mechanical Attrition Processes

2015 ◽  
Author(s):  
Zhiyu Wang ◽  
Saurabh Basu ◽  
Christopher Saldana
Keyword(s):  
Strain Hardening ◽  
Model Material ◽  
Process Models ◽  
Image Correlation ◽  
Analytical Models ◽  
Subsurface Deformation ◽  
Multi Stage ◽  
Mechanical Attrition ◽  
Dead Metal Zone ◽  
Plane Strain Deformation

A modified expanding cavity model (M-ECM) is developed to describe subsurface deformation for strain-hardening materials loaded in unit deformation configurations occurring in surface mechanical attrition. The predictive results of this model are validated by comparison with unit deformation experiments in a model material, oxygen free high conductivity copper, using a custom designed plane strain deformation setup. Subsurface displacement and strain fields are characterized using in-situ digital image correlation. It is shown that conventional analytical models used to describe plastic response in strain-hardening metals are not able to predict important characteristics of the morphology of the plastic zone, including evolution of the dead metal zone (DMZ), especially at large plastic depths. The M-ECM developed in the present study provides an accurate prediction of the strain distribution obtained in experiment and is of utility as a component in multi-stage process models of the final surface state in surface mechanical attrition.

Access to Empirical Formulations of Combustion Efficiency for Gas Turbine Combustors with Improved Generalization Ability

2020 ◽  
pp. 1-32
Author(s):  
Zhuang Ma ◽  
Tingwei Ji ◽  
Tao Cui ◽  
Yao Zheng
Keyword(s):  
Experimental Data ◽  
Polynomial Regression ◽  
Burning Velocity ◽  
Velocity Model ◽  
Combustion Efficiency ◽  
Symbolic Regression ◽  
Regression Method ◽  
Generalization Ability ◽  
Operating Variables ◽  
Model Structures

Abstract Correlating combustion performance parameters to the main operating variables of combustors with mathematical expressions contributes to reducing the number of experiments and simplifying the design procedure of gas turbines. The application of empirical formulations meets the requirement with finite precision. The present study aims at adopting symbolic regression method to establish empirical formulations to correlate combustion efficiency with the main operating variables of gas turbine combustors. Differing from ordinary data modeling methods that search model parameters only with model structures fixed, symbolic regression method can search the structures and parameters of mathematical models simultaneously. In this article, attempts to correlate the experimental data of Combustor I using the mechanism model of burning velocity model, neural network, polynomial regression and symbolic regression are shown sequentially. Burning velocity model has not satisfactory accuracy by comparing the predictions with the experimental data which means its lower generalization ability. Comparatively, the predictions of the empirical formulation obtained by the present symbolic regression method are in good agreement with the experimental data, and also excel those of neural network and polynomial regression in generalization ability. Another two formulations are obtained by symbolic regression using the experimental data of Combustor II and III, and the different model structures of the two formulations indicate that there is still room for improvement in the present method.

Investigation of Size Effects on Process Models for Plane Strain Microbending

2006 ◽  
Author(s):  
Richard M. Onyancha ◽  
Brad L. Kinsey
Keyword(s):  
Size Effects ◽  
Strain Distribution ◽  
Past Research ◽  
Peak Force ◽  
Process Models ◽  
Analytical Models ◽  
Linear Strain ◽  
Bending Force ◽  
Bending Process ◽  
Individual Grains

Accurate process models provide vital information in the design of manufacturing processes. To characterize bending operations, analytical models have been developed and shown to predict the peak bending forces fairly accurately for sheets in the macro or mesoscale (i.e. sheets with a large number of grains through the thickness). However, whether these models also accurately predict bending forces for sheets in the microscale (i.e. sheets with approximately ten grains or less through the thickness) has not been evaluated. The present study is aimed at investigating the use of two such models from previous work with microscale bending data. In addition, using these previous models as a foundation, additional bending force models were developed to predict the bending force specifically for microscale bending operations. Data analysis showed that the process models from past research, which provide accurate results for macroscale bending, over predict the peak force required for bending microscale sheets. These process models assume a non-linear strain distribution through the thickness and a curved formed wall. The two models developed in this research provide accurate results for the microscale bending tests, however, they under predict the peak force for the macroscale bending operation. These developed process models assume a linear strain distribution through the thickness and a straight formed wall. The linear strain distribution is more appropriate for the microscale bending process as there are few grains through the thickness and the strain in individual grains varies linearly across the grain. The straight formed wall is more appropriate for the microscale bending process as there is not sufficient distance to warrant a curved formed wall assumption. These differences represent size effects for assumptions in the process models. The material used for these investigations was Brass (CuZn15). The sheets had between 2 and 50 grains through the thickness with grain sizes of between 10 μm and 71 μm.

Sign in / Sign up

Export Citation Format

Share Document