【摘要】 建立了一种人工神经网络-X射线荧光光谱法测定钢中酸溶铝的方法,用X射线荧光光谱法测定低合金钢中总铝值,应用所建立的ANN-BP网络模型,输入总铝含量直接预测出酸溶铝含量。同时使用改进的BP算法,避免了神经网络学习中可能产生的麻痹现象。该方法用于钢中酸溶铝的测定,结果满意。更多还原

【Abstract】 Total aluminum content in low alloy steel was determined by XRFS, and by applying the ANN-BP network model, the content of acid-soluble aluminum was calculated from the data of total aluminum content input. The application of the improved BP algorithm, paralysis which may happen in the training procedure of ANN, was effectively avoided. In its practical application, 17 standard samples were used as training collection, 6 standard samples were used as a collection of prediction, and 7 low alloy steel samples were taken to be analyzed by the proposed method. Results obtained are in conformity with those obtained by the chemical method.更多还原