• Journal of Information Processing Systems
• /
• v.18 no.2
• /
• pp.268-281
• /
• 2022

The development of information technology is bringing many changes to everyday life, and machine learning can be used as a technique to solve a wide range of real-world problems. Analysis and utilization of data are essential processes in applying machine learning to real-world problems. As a method of processing data in machine learning, we propose an approach based on applying multiple linear regression models by interlacing data to the task of classifying similar software. Linear regression is widely used in estimation problems to model the relationship between input and output data. In our approach, multiple linear regression models are generated by training on interlaced feature data. A combination of these multiple models is then used as the prediction model for classifying similar software. Experiments are performed to evaluate the proposed approach as compared to conventional linear regression, and the experimental results show that the proposed method classifies similar software more accurately than the conventional model. We anticipate the proposed approach to be applied to various kinds of classification problems to improve the accuracy of conventional linear regression.

• Advances in nano research
• /
• v.12 no.1
• /
• pp.37-47
• /
• 2022

In this paper, the non-linear static analysis of Timoshenko nanobeams consisting of bi-directional functionally graded material (BFGM) with immovable ends is investigated. The scratching in the FG nanobeam mid-plane, is the source of nonlinearity of the bending problems. The nonlocal theory is used to investigate the non-linear static deflection of nanobeam. In order to simplify the formulation, the problem formulas is derived according to the physical middle surface. The Hamilton principle is employed to determine governing partial differential equations as well as boundary conditions. Moreover, the differential quadrature method (DQM) and direct iterative method are applied to solve governing equations. Present results for non-linear static deflection were compared with previously published results in order to validate the present formulation. The impacts of the nonlocal factors, beam length and material property gradient on the non-linear static deflection of BFG nanobeams are investigated. It is observed that these parameters are vital in the value of the non-linear static deflection of the BFG nanobeam.

• Biomaterials and Biomechanics in Bioengineering
• /
• v.2 no.4
• /
• pp.207-223
• /
• 2015

In this paper a method for simultaneous swift non-linear analysis and optimal design/posture of mechanical/biomechanical systems is presented. The method is developed to get advantages of iterations in non-linear analysis and/or generations in genetic algorithm (GA) for the purpose of efficient analysis within the optimal design/posture. The method is applicable for both size and geometry optimizations wherein material and geometry non-linearity are present. In addition to established mechanical systems, the method can solve biomechanical models of human musculoskeletal system. Optimization-based procedures are popular methods for resolving the redundancy at joints wherein the number of unknown muscle forces is far more than the number of equilibrium equations. These procedures involve optimization of a cost function(s) which is assumed to be consistent with the central nervous system's strategy when activating muscles to assure equilibrium. However, because of the complexity of biomechanical problems (i.e., due to non-linear biomaterial, large deformation, redundancy of the problem and so on) efficient analysis are required within optimization procedures as suggested in this paper.

• Smart Structures and Systems
• /
• v.26 no.1
• /
• pp.23-33
• /
• 2020

Majority of the damage in engineering structures is nonlinear. Damage sensitive features (DSFs) extracted by traditional methods from linear time series models cannot effectively handle nonlinearity induced by structural damage. A new DSF is proposed based on vector space cosine similarity (VSCS), which combines K-means cluster analysis and Bayesian discrimination to detect nonlinear structural damage. A reference autoregressive moving average (ARMA) model is built based on measured acceleration data. This study first considers an existing DSF, residual standard deviation (RSD). The DSF is further advanced using the VSCS, and then the advanced VSCS is classified using K-means cluster analysis and Bayes discriminant analysis, respectively. The performance of the proposed approach is then verified using experimental data from a three-story shear building structure, and compared with the results of existing RSD. It is demonstrated that combining the linear ARMA model and the advanced VSCS, with cluster analysis and Bayes discriminant analysis, respectively, is an effective approach for detection of nonlinear damage. This approach improves the reliability and accuracy of the nonlinear damage detection using the linear model and significantly reduces the computational cost. The results indicate that the proposed approach is potential to be a promising damage detection technique.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.3 no.1
• /
• pp.24-33
• /
• 1999

For performance prediction and diagnostics of gas turbine engines, linear and non-linear gas path analysis are applied. In order to find optimal instrument parameters to detect the physical faults such as (outing, erosion and corrosion, non-linear gas path analysis is used. A typical industrial gas turbine engine, TB5000, is used to study the effect of physical faults on engine performance. Through comparison of RMS error between linear and non-linear gas path analysis, the optimal instrument parameters can be defined. As a result, it is found that the linear GPA has the level of error introduced by the assumption of the linear mode: can be of the same order of magnitude as the fault being soughtwhile the non-linear GPA can be solved the non-linear relationships between dependent and independent parameters using an iterative method such as the Newton-Raphson method with sufficient accuracy.

• Earthquakes and Structures
• /
• v.14 no.2
• /
• pp.167-178
• /
• 2018

A roller compacted concrete (RCC) dam should be analyzed under seismic ground motions for different conditions such as empty reservoir and full reservoir conditions. This study presents three-dimensional earthquake response and performance of a RCC dam considering materially non-linearity. For this purpose, Cine RCC dam constructed in Aydın, Turkey, is selected in applications. The three-dimensional finite element model of Cine RCC dam is obtained using ANSYS software. The Drucker-Prager material model is considered in the materially nonlinear time history analyses for concrete and foundation rock. Furthermore, hydrodynamic effect was investigated in linear and non-linear dynamic analyses. Researchers observe that how the tensile and compressive stresses change by hydrodynamic pressure effect. The hydrodynamic pressure of the reservoir water is modeled with the fluid finite elements based on the Lagrangian approach. In this study, dam body and foundation are modeled with welded contact. The displacements and principle stress components obtained from the linear and non-linear analyses with and without reservoir water are compared each other. Principle stresses during earthquake were obtained at the most critical point in the upstream face of dam body. Besides, the change of displacements and stresses by crest length were investigated. Moreover demand-capacity ratio criteria were also studied under linear dynamic and nonlinear analysis. Earthquake performance analyses were carried out for different cases and evaluated. According to linear and nonlinear analysis, hydrodynamic water effect is obvious in full reservoir situation. On the other hand, higher tensile stresses were observed in linear analyses and then non-linear analyses were performed and compared with each other.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.367-372
• /
• 2006

When we design a linear motor, the thermal behavior investigation is one of great important considerations with respect to uniform thrust force and thermal deformation of a linear motor. In this study, we conduct the research for the structural design of the linear motor for LMTT(Linear Motor-based Transfer Technology) which is the next generation of container horizontal transfer system in order to automate a container terminal. After the dimensions of main parts for a linear motor were set up, we carried out the thermal-structural analysis of the linear motor considering the thermal analysis of the stator module.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.1256-1261
• /
• 2004

Nonlinear Response Optimization using Equivalent Static Loads (NROESL) method/algorithm is proposed to perform optimization of non-linear response structures. It is more expensive to carry out nonlinear response optimization than linear response optimization. The conventional method spends most of the total design time on nonlinear analysis. Thus, the NROESL algorithm makes the equivalent static load cases for each response and repeatedly performs linear response optimization and uses them as multiple loading conditions. The equivalent static loads are defined as the loads in the linear analysis, which generates the same response field as those in non-linear analysis. The algorithm is validated for the convergence and the optimality. The function satisfies the descent condition at each cycle and the NROESL algorithm converges. It is mathematically validated that the solution of the algorithm satisfies the Karush-Kuhn-Tucker necessary condition of the original nonlinear response optimization problem. The NROESL algorithm is applied to two structural problems. Conventional optimization with sensitivity analysis using the finite difference method is also applied to the same examples. The results of the optimizations are compared. The proposed method is very efficient and derives good solutions.

• Journal of Mechanical Science and Technology
• /
• v.20 no.7
• /
• pp.1089-1097
• /
• 2006

The market demand of display devices is drastically increasing in the information technology age. The research on OLED (Organic Light Emitting Diodes) display with the luminescence in itself is being more paid attention than LCD (Liquid Crystal display) with the light source from the back. The vapor deposition process is most essential in manufacturing OLED display. The temperature distribution of the linear cell in this process is closely related to securing the uniformity of organic materials on the substrate. This work analyzed the temperature distribution depending on the intervals between the crucible and the heating band as well as on the amount of the heat flux from the heating band. Moreover, the roles of the water jacket and the configuration of the cover within the linear cell were examined through the temperature analysis for six configurations of the linear cell. Under the above temperature analysis, the variations in the intervals and the amount of the heat flux were considered to have an effect on building the uniform temperature distribution within the crucible. It is predicted that the water jacket and the adequate configuration of the cover will prevent the blowout and clogging phenomena, respectively. The results can be used as the fundamental data for designing the optimal linear cell.

• Structural Engineering and Mechanics
• /
• v.17 no.6
• /
• pp.751-766
• /
• 2004

This paper extends the use of the hierarchic degenerated shell element to geometric non-linear analysis of composite laminated skew plates by the p-version of the finite element method. For the geometric non-linear analysis, the total Lagrangian formulation is adopted with moderately large displacement and small strain being accounted for in the sense of von Karman hypothesis. The present model is based on equivalent-single layer laminate theory with the first order shear deformation including a shear correction factor of 5/6. The integrals of Legendre polynomials are used for shape functions with p-level varying from 1 to 10. A wide variety of linear and non-linear results obtained by the p-version finite element model are presented for the laminated skew plates as well as laminated square plates. A numerical analysis is made to illustrate the influence of the geometric non-linear effect on the transverse deflections and the stresses with respect to width/depth ratio (a/h), skew angle (${\beta}$), and stacking sequence of layers. The present results are in good agreement with the results in literatures.