节点文献
基于神经网络-遗传算法优化柴油机油台架试验
Optimization for Engine Test of Engine Oil Based on Genetic Algorithm-neural Network
【摘要】 根据15W/40 CD柴油机油的模拟实验与台架试验的基础数据,用人工神经网络(ANN)的反向传播算法建立了模拟实验与台架试验神经网络预测模型,该神经网络模型合适的拓扑结构为5-7-1,学习速率为0.2,动量因子为0.9。探讨了用模拟实验数据预测台架试验结果的可能性,检验证明用人工神经网络方法建立的模型能准确预报15W/40CD柴油机油的台架试验结果。该神经网络预测模型用遗传算法优化,得到了15W/40 CD级柴油机油能通过台架试验的最优模拟实验结果。
【Abstract】 Based on the data of bench test and engine test of 15W/40 CD grade engine oil,an artificial neural network(ANN) model was developed for predicting the relationship between engine test and bench test using back-propagation algorithm.The appropriate topology of ANN was 5-7-1.The learning rate of ANN was 0.2,and the momentum factor of ANN was 0.3.It is shown that the ANN model can correlate and predict results of engine test of 15W/40 CD grade engine oil with much accuracy.The ANN model was optimized by incorporating genetic algorithm.Optimal bench test of 15W/40 CD grade engine oil was obtained using the ANN model developed.
【Key words】 diesel engine oil; bench test; engine test; artificial neural network; genetic algorithm;
- 【文献出处】 润滑与密封 ,Lubrication Engineering , 编辑部邮箱 ,2006年04期
- 【分类号】TK427
- 【被引频次】3
- 【下载频次】174