神经网络技术在储运工程中的应用综述

Summary on Application of Neural Network Technology to Transportation and Storage Engineering

人工神经网络是近年来得到迅速发展的一种算法 ,在各个领域已经得到了广泛的应用。针对储运工程 ,介绍了多层感知器网络、径向基网络和小波网络在管道智能检测数据分析识别中应用 ,前向神经网络在管道综合可靠度决策评价中应用 ,基于自适应反馈预测模型和自适应函数连接预测模型的天然气消费预测以及Hopfield神经网络在管线优化设计中的应用。可以看出 ,神经网络必将成为解决储运工程问题的有力工具 ,对突破现有技术的瓶颈 ,更深入地探索非线性等复杂现象将起到重大作用。

Artificial Neural Network is a rapidly developing algorithm in recent years and has been used in many domains.Described are application of multi-layer perceptron network, radial basis functional network and wavelet basis functional network to data analysis and identification in intelligent inspection of pipelines, application of feed forward neural network to decision and evaluation of integrated pipeline fiduciary level, application of natural gas consumption forecast based on adaptive feed forward forecast mode and adaptive functional link forecast mode, and application of Hopfield neural network to pipeline optimization design. It can be seen that Neural Network will certainly become an important technique for solving engineering problems in transportation and storage.