摘要
干化学分析仪器的常规故障快速检测方法以故障具体表现的特征值作为判断依据,存在一定局限性,导致故障分类准确度较低,因此,针对全血谷丙转氨酶干化学分析检测仪在医院应用过程中的故障快速检测方法进行研究。首先建立全血谷丙转氨酶干化学分析检测仪的故障树,熟悉检测仪内部构造、功能、原理等,选取顶事件、边界条件,计算底事件关键重要度,建立故障树模型;为实现故障准确分类,引入瞬态小信号采样技术,将信号数据行加窗分段得到短时相关函数,完成噪声去除后可以得到背景均值以及协方差,用数据门限检验数据段,确定瞬态小信号位置,完成故障类型的快速检测和分类判断。仿真实验结果表明,相同实验条件下,设计方法的平均准确率比常规方法高出58.2%,验证了设计方法的有效性。
The conventional rapid fault detection method of dry chemical analysis instrument takes the characteristic value of the specific performance of the fault as the basis of judgment, which has some limitations and leads to low fault classification accuracy.Therefore, the rapid fault detection method of whole blood alanine aminotransferase dry chemical analysis instrument in the process of application in hospital was studied.Firstly, the fault tree of the whole blood alanine aminotransferase dry chemical analysis detector was established, and the internal structure, function and principle of the detector were familiar.The top events and boundary conditions were selected to calculate the critical importance of the bottom events, and the fault tree model was established.In order to achieve accurate fault classification, the transient small signal sampling technology is introduced, and the signal data rows are segmented with Windows to obtain short-term correlation functions.After noise removal, the background mean value and covariance can be obtained.The data segments are tested with data thresholds to determine the location of transient small signals, and the rapid detection and classification of fault types are completed.Simulation results show that under the same experimental conditions, the average accuracy of the design method is 58.2% higher than that of the conventional method, which verifies the effectiveness of the design method.
作者
完燕华
WAN Yanhua(The 980 Hospital People’s Liberation Army of China,Hebei Shijiazhuang 050051,China)
出处
《自动化与仪器仪表》
2021年第5期53-56,共4页
Automation & Instrumentation
基金
青年科学基金项目(No.21405091)。
关键词
全血谷丙转氨酶
干化学分析
故障快速检测
whole blood alanine aminotransferase
dry chemical analysis
fast fault detection
