摘要
粗差的存在会对大坝安全监测的结果造成较大的影响,而现今传统的粗差识别方法还存在着操作繁琐、应用数据类型有限、易少判误判等缺陷。因此,为了弥补传统粗差识别方法的不足,提高粗差识别的效率,采取了借鉴人工免疫的工作模式对监测数据进行分层打分的方法,将其与现有粗差检验方法结合,提出并设计出基于人工免疫模式的多层粗差检验法。以实际工程大坝水平位移监测数据的粗差识别为例,对工程的水平位移监测数据进行粗差识别。算例验证结果表明:这种新型的多层粗差检验法对大容量数据的粗差识别操作更为简便、效率更高。与传统粗差识别相比,其数据识别种类变广,而且其多层打分的检验模式使得识别结果更为严谨,可以有效减少粗差错判少判的情况。
Gross error has great impact on the result of dam safety monitoring, but the conventional method of gross error discrimination still has the defects of complicated operation, limited application data types, prone to less-judgement and misjudgement, etc. In order to make up for the shortcomings of the conventional method of gross error discrimination and improve the efficiency of gross error discrimination, the monitoring data are scored in layers herein with the reference from the working mode of artificial immunity, which is combined with those current gross error detection methods, and then an artificial immune mode-based multi-layer gross error detection method is proposed and designed. By taking the gross error discrimination of the horizontal displacement monitoring data of an actual dam project as a case, the gross error of the horizontal displacement of the project is discriminated. The calculation case successfully verified that this new kind of multi-layer gross error detection method is more simple and convenient for the gross error discrimination of large volume data with more higher efficiency. Compared with the conventional method of gross error discrimination, the types of data discrimination made by this method become wider, while its detection mode of multi-layer scoring makes the discrimination result more precise, and then can effectively reduce less-judgement and misjudgement.
作者
陶园
刘晓青
TAO Yuan;LIU Xiaoqing(College of Water Conservancy and Hydropower,Hohai University,Nanjing210098,Jiangsu,China)
出处
《水利水电技术》
北大核心
2019年第3期66-71,共6页
Water Resources and Hydropower Engineering
基金
国家重点研发计划"水库大坝安全诊断与智慧管理关键技术与应用--大型复杂水工结构性能演化测试装备与智能诊断技术"(2018YFC0407102)
关键词
粗差识别
人工免疫模式
大坝安全监测
gross error discrimination
artificial immune mode
dam safety monitoring
