首页|基于DZZ5气象仪器设备维护及保障措施研究

基于DZZ5气象仪器设备维护及保障措施研究

扫码查看
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.
Research on Maintenance and Guarantee Measures of Meteorological Instrument Equipment Based on DZZ5
In order to perform more effective daily maintenance on meteorological instruments and improve their operational stabil-ity,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 algo-rithm 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 instru-ment abnormal state monitoring model based on PSO-OCSVM-LSTM has stronger recognition performance and can more accurately i-dentify 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 opera-tion status of meteorological instruments in real time and accurately identify instrument faults.It can be applied to actual meteorologi-cal instrument state monitoring,helping management personnel to carry out more convenient instrument daily maintenance and ensu-ring the stable operation of the instrument.

DZZ5 meteorological instrumentstatus monitoringequipment maintenancesupport vector machineLSTM

顾建兵、姚淑萍、马宁

展开 >

宁夏回族自治区石炭井气象站,宁夏石嘴山 530022

宁夏气象信息中心,银川 530022

DZZ5气象仪器 状态监测 设备维护 支持向量机 LSTM

2024

自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
年,卷(期):2024.(7)