粗糙集和神经网络相融合的钻井安全评价模型

发布时间：2018-06-28 22:59

本文选题：安全评价 + 粗糙集　；参考：《西南石油大学学报(自然科学版)》2017年05期

【摘要】：针对动态性、随机性和不确定性较强的钻井作业现场,开展了安全评价的研究。提出了一种基于粗糙集和BP神经网络对作业现场进行安全评价的方法。首先以粗糙集为基础来构建BP神经网络的前置系统,对采集到的样本数据进行属性约简。其次,根据约简结果以及作业当天的事故情况完成了BP神经网络输入层和输出层的设计,并根据输入层和输出层神经元的个数通过试凑法确定网络隐含层的神经元数量范围,并采用训练样本对不同神经元个数所对应的网络模型进行训练,选择网络误差最低的网络作为所构建的网络模型。最后,选取16 d的测试样本对网络进行验证,将网络的输出同作业现场的实际结果进行比较,有14 d的网络结果与实际结果相符,测试准确率达到了87.5%。
[Abstract]:In view of the dynamic, random and uncertain drilling site, the safety evaluation is studied. A method of job site safety evaluation based on rough set and BP neural network is proposed. Firstly, the presystem of BP neural network is constructed on the basis of rough set, and the collected sample data is reduced by attribute. Secondly, the design of the input and output layers of BP neural network is completed according to the reduction result and the accident situation on the day of operation, and the number of neurons in the hidden layer of the network is determined by trial and error method according to the number of neurons in the input layer and output layer. The training samples are used to train the network model corresponding to different number of neurons, and the network with the lowest network error is selected as the network model. Finally, 16 days of test samples are selected to verify the network, and the network output is compared with the actual results of the job site. There are 14 days of network results consistent with the actual results, and the test accuracy is 87.5%.
【作者单位】：西南石油大学计算机科学学院;
【基金】：国家科技重大专项(2011ZX05021-006) 四川省安全生产监督局项目(sichuan-0009-2016AQ)
【分类号】：TE28;TP18

【相似文献】