储罐底板腐蚀状态的人工神经网络智能评价方法

摘要

根据储罐底板在线检测的声发射信息和外观检查信息，确定与储罐底板腐蚀状态相关的表征因素，应用人工神经网络智能评价方法，分别建立基于外观检查信息、基于声发射信息和基于在线检测信息的储罐底板腐蚀状态评价模型。通过对测试样本的评价，对比声发射检测评价结果，其中基于在线检测信息的储罐底板腐蚀状态评价模型的准确率为94％，该模型能够对储罐底板腐蚀状态进行准确的评价，实现储罐底板声发射在线检测评价的智能化。%According to the acoustic emission information and the appearance inspection information of tank bottom online testing, the external factors associated with tank bottom corrosion status were confirmed. Applying artificial neural network intelligent evaluation method, three tank bottom corrosion status evaluation models based on appearance inspection information, acoustic emission information and online testing information were established. Comparing with the result of acoustic emission online testing through the evaluation of test sample, the accuracy of the evaluation model based on online testing information was 94%. The evaluation model could evaluate tank bottom corrosion accurately and realize acoustic emission online testing intelligent evaluation of tank bottom.