论文部分内容阅读
金泥的处理工业上普遍采用火法冶金，炉渣组成对获取最高的金回收，率和最低渣中含金量具有重要影响，本文应用神经网络方法，对苏打－硼砂－玻璃－食盐所组成的渣系同熔炼渣含金量的关系进行研究，建立可用于预测不同炉渣组成分渣含金的神经网络模型，包括网络类型，网络结构及其算法），研究结果表明，所建立的三层反向传播神经网络模型可以用于熔炼体系渣含金的预测，且比传统回归分析法有许多突出的优点。
Gold mud treatment industry commonly used pyrometallurgical, slag composition to obtain the highest gold recovery rate and the lowest content of gold has a significant impact, the application of neural network method, the soda - borax - glass - salt slag system composed of Smelting slag gold content of the relationship between the study to establish a different gold content of the slag can be used to predict the gold content of the neural network model, including the type of network, network structure and its algorithm), the results show that the established three layers of reverse propagation neural network model It can be used to forecast the gold content of smelting system slag, and has many outstanding advantages over the traditional regression analysis.