Application of computational intelligence on fitting of micro-drill's main lips

A new approach based on BP neural network integrated with genetic algorithm for fitting of micro-drill's main lips is presented. The network structure is designed according to fitting equation, coordinates of micro-drill sampled points and constant 1 are taken as 3 inputs of network, 1 output i...

Full description

Saved in:
Bibliographic Details
Main Authors: Ge Dong-Yuan, Yao Xi-Fan, Jiang Shi-ming
Format: Conference Proceeding
Language:English
Subjects:
Artificial neural networks
BP Neural Network
Chips
Computational Intelligence
Equations
Fitting
Genetic Algorithm
Lips
Main Lips
Micro-Drill
Performance analysis
Pixel
Signal processing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A new approach based on BP neural network integrated with genetic algorithm for fitting of micro-drill's main lips is presented. The network structure is designed according to fitting equation, coordinates of micro-drill sampled points and constant 1 are taken as 3 inputs of network, 1 output is obtained, and the square of errors between the output and constant 0 is taken as performance index. Weights between input neurons and output neuron are tuned in the light of gradient descent mean, and stable weight values are obtained until the desired performance index is reached. In order to obtain global optimal solution, genetic algorithm is integrated in the linear BP NN, and expression coefficients of main lips line can be solved according to the weigh. Thus chips depth of micro-drills main lips can be measured easily. The approach has advantages of algorithm simple and higher precision over conventional approaches such as least square means and so on.
DOI:10.1109/ICSPS.2010.5555249