基于神经网络和遗传算法的镍—钨—碳化硅复合电刷镀研究

Research on Compound Brush-plating of Ni-W-Sic with Neural Networks and Genetic Algorithms

利用神经网络建立了Ni-W-Sic复合电刷镀镀层的工艺数学模型,并用该模型预测的工艺参数对该复合镀层耐磨性能的影响进行了研究,同时利用遗传算法求出最佳的工艺参数组合,试验表明,按优化参数制得的复合镀层具有较好的耐磨性,可用于提高柴油机活塞环的使用寿命,特别是对烧重油,甚至烧渣油的柴油机活塞环更有利。

The mathematical model for technology of compound brush-plating of Ni-W-Sic is built with neural net- works and this paper makes researches on the effect of technology parameters that the model predict on wear resist- ance of composite coatings. The optimum combination of technology parameters are obtained with genetic algo- rithms. Experiment shows that the composite coatings made with these optimized parameters have better wear resist- ance and can be used to raise service life of the piston ring. It is especially favorable for piston ring of diesel engine that consumes heavy or residual fuel.