scholarly journals Modified Cascade Correlation Neural Network and its Applications to Multidisciplinary Analysis Design and Optimization in Ship Design

10.5772/9148 ◽  
2010 ◽  
Author(s):  
Adeline Schmitz ◽  
Frederick Courouble ◽  
Hamid Hefazi ◽  
Eric Besnar
Keyword(s):  
Neural Network ◽  
Ship Design ◽  
Design And Optimization ◽  
Multidisciplinary Analysis ◽  
Cascade Correlation ◽  
Analysis Design

Analysis, Design, and Optimization of Thermal In-Plane Microactuator—Chevron Type

2011 ◽  
Vol 8 (3) ◽  
pp. 102-109
Author(s):  
K.B. Puneeth ◽  
K.N. Seetharamu
Keyword(s):  
Neural Network ◽  
Mechanical Response ◽  
Thermal Response ◽  
Structural Response ◽  
Design Tool ◽  
Thermal Actuator ◽  
Coupled Models ◽  
Element Analysis ◽  
Design And Optimization ◽  
Analysis Design

A predictive model of thermal actuator behavior has been developed and validated that can be used as a design tool to customize the performance of an actuator to a specific application. Modeling thermal actuator behavior requires the use of two sequentially or directly coupled models, the first to predict the temperature increase of the actuator due to the applied voltage and the second to model the mechanical response of the structure due to the increase in temperature. These models have been developed using ANSYS for both thermal response and structural response. Consolidation of FEA (finite element analysis) results has been carried out using an ANN (artificial neural network) in MATLAB. It is seen that an ANN can be successfully employed to interpolate and predict FEA results, thus avoiding necessity of running FEA code for every new case. Furtheroptimization of geometry for maximum actuation length has been carried out using a GA (genetic algorithm) in MATLAB. The results of the GA were verified against the ANN and FEA results.

Gibbs Systems Dynamics: A Simple But Powerful Tool for Process Analysis, Design and Optimization

2002 ◽  
Author(s):  
Pierre Neveu ◽  
Nathalie Mazet
Keyword(s):  
Process Analysis ◽  
Irreversible Processes ◽  
Systems Dynamics ◽  
Vapor Compression ◽  
Technical Data ◽  
Design And Optimization ◽  
The Mean ◽  
Analysis Design ◽  
Phenomenological Coefficients ◽  
Condensing Pressure

Dynamic process modeling by the mean of Equivalent Gibbs systems is described here. It allows to model a large number of processes and only requires standard engineering knowledge. This method is issued from thermodynamics of irreversible processes, initiated by I. Prigogine, but applied here to process engineering. First, an Equivalent Gibbs System (EGS) is defined for each component involved in the process. In such system, mass, energy and entropy are linked through Gibbs equation and entropy production can easily be expressed according to fluxes and their related forces. Assuming linear phenomenological laws, phenomenological coefficients can be calculated from common engineering correlations, or evaluated from technical data if available. As an example, a conventional vapor compression chiller is simulated. Three control modes are analyzed on an exergy basis: on/off control with constant or floating condensing pressure, PID control with variable compressor speed.

Neural Network Studies. 3. Variable Selection in the Cascade-Correlation Learning Architecture

1998 ◽  
Vol 38 (4) ◽  
pp. 651-659 ◽  
Author(s):  
Vasyl V. Kovalishyn ◽  
Igor V. Tetko ◽  
Alexander I. Luik ◽  
Vladyslav V. Kholodovych ◽  
Alessandro E. P. Villa ◽  
...  
Keyword(s):  
Neural Network ◽  
Variable Selection ◽  
Cascade Correlation ◽  
Correlation Learning ◽  
Network Studies
Sign in / Sign up

Export Citation Format

Share Document