|
|
|
| A metal milling burr prediction based on combination algorithm |
| YUAN Si-Cong, LI Chao, AN Feng, ZHANG Chen, WANG Rong |
| School of Mechanical and Electrical Engineering， Xi′an University of Architecture and Technology， Xi′an 710055, China |
|
|
|
Abstract With banding together the gray theory， the artificial neural network and genetic algorithm on the basis of optimal weights coefficient， we got a combination algorithm. We built three combination models according to three kinds of mathematical methods， which were the combination model of arithmetic mean， the combination model of quadratic sum average and the combination model of proportion average， meanwhile， the models were applied to metal milling burr forecasting of carbon steel 45. We analyzed and calculated the results of the forecasting models by using the three evaluation indexes of prediction error: sum of squared error mean， absolute error mean， relative error. The consequences show that the combination model of arithmetic mean is more consistent with the experimental data， and it has a higher accuracy and stability， which is pretty valuable for metal milling burr forecasting．
|
|
Published: 28 February 2013
|
基于组合算法的金属铣削毛刺预测
利用最优权值系数，将灰色理论、人工神经网络和遗传算法有机结合，构建组合算法，依据3种数学方法建立3种组合模型：组合算术平均模型、组合平方和平均模型以及组合比例平均模型，并分别将3种组合模型应用于45钢铣削毛刺的预测．利用3个预测误差评价指标，即平方和误差指标、平均绝对误差指标和平均相对误差指标，对各模型的预测结果进行分析计算.结果表明，组合算术平均模型所得结果与实验结果取得了较好的吻合，具有较高的精度和稳定性，对于金属铣削毛刺的预测具有实际的应用价值．
关键词：
毛刺,
灰色理论,
BP神经网络,
遗传算法,
最优权值系数,
组合预测
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
