基于DZZ5气象仪器设备维护及保障措施研究  

Research on Maintenance and Guarantee Measures of Meteorological Instrument Equipment Based on DZZ5

作　　者：顾建兵 姚淑萍 马宁 GU Jianbing;YAO Shuping;MA Ning(Shitanjingmeteorologicalstation,Ningxia Hui Autonomous region,ShiZuiShan,Ningxia 530022,China;Ningxia Meteorological Information Center,Yinchuan,Ningxia 530022,China)

机构地区：[1]宁夏回族自治区石炭井气象站,宁夏石嘴山530022 [2]宁夏气象信息中心,银川530022

出　　处：《自动化与仪器仪表》2024年第7期60-65,共6页Automation & Instrumentation

摘　　要：为对气象仪器进行更加有效的日常维护并提高仪器运行的稳定性,以DZZ5气象仪器为研究对象,提出一种基于改进一类支持向量机结合长短时记忆网络的DZZ5气象仪器状态监测模型。其中,以一类支持向量机作为基础的状态识别分类方法,引入粒子群优化算法并结合长短时记忆网络对支持向量机进行优化,进一步提升状态识别的综合效果。结果表明,与传统的LSTM识别模型相比,构建的基于PSO-OCSVM-LSTM的DZZ5气象仪器异常状态监测模型具有更强的识别性能,能够对仪器的状态进行更加准确的识别和分类,误识别情况较少;将构建的仪器异常状态监测模型应用于实际的工作场景中时,模型表现较好,符合实际的工作需求。综上,构建的DZZ5气象仪器异常状态监测模型性能优良,能够进行对气象仪器的日常运行状态进行实时监控,同时能够对仪器故障进行准确识别,能够应用于实际的气象仪器状态监测,帮助管理人员进行更加便利的仪器日常维护,保障仪器的稳定运行。In order to perform more effective daily maintenance on meteorological instruments and improve their operational stability,a DZZ5 meteorological instrument status monitoring model based on an improved class of support vector machines combined with long short-term memory networks is proposed,taking the DZZ5 meteorological instrument as the research object.Among them,a state recognition classification method based on a type of support vector machine is introduced,and particle swarm optimization algorithm is combined with long short-term memory network to optimize the support vector machine,further improving the comprehensive effect of state recognition.The results show that compared with traditional LSTM recognition models,the DZZ5 meteorological instrument abnormal state monitoring model based on PSO-OCSVM-LSTM has stronger recognition performance and can more accurately identify and classify the instrument's state,with fewer misidentification cases;When the constructed instrument anomaly monitoring model is applied to practical work scenarios,the model performs well and meets practical work requirements.In summary,the DZZ5 meteorological instrument abnormal state monitoring model constructed has excellent performance,which can monitor the daily operation status of meteorological instruments in real time and accurately identify instrument faults.It can be applied to actual meteorological instrument state monitoring,helping management personnel to carry out more convenient instrument daily maintenance and ensuring the stable operation of the instrument.

关 键 词：DZZ5气象仪器 状态监测 设备维护 支持向量机 LSTM 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]