论文部分内容阅读
本人通过对人工神经网络BP算法的研究,提出了两种基于BP改进算法,并结合泥沙学科中的宽级配非均匀沙最大起动粒径dcmax的推求。采用C#语言,分别实现BP经典算法以及自适应调整学习效率和模拟退火两种改进算法,建立了dcmax预测模型并对预测结果进行验证与分析。表明改进算法的优势所在,体现出改进算法的应用价值和发展前景。
Based on the research of artificial neural network BP algorithm, two kinds of improved algorithm based on BP are proposed, and combined with the wide-range non-uniform sand maximum starting particle size dcmax in sediment research. Using C # language, two improved algorithms of BP classical algorithm, adaptive adjustment learning efficiency and simulated annealing are respectively established. The dcmax prediction model is established and the prediction results are verified and analyzed. The advantage of the improved algorithm is demonstrated, which shows the application value and development prospect of the improved algorithm.