基于激光光声光谱的变压器油中溶解气体在线监测预警装置研制

Development of On-line Monitoring and Early Warning Device for Dissolved Gas in Transformer Oil Based on Laser Photoacoustic Spectroscopy

油中溶解气体的相关参量能很好地表征设备的工作状态,因此开展变压器油中溶解气体分析,对于预知变压器潜伏性故障、预防规模化停电事故意义重大。针对现有色谱法监测的弊端,基于激光光声光谱原理,设计了气体监测模块,研制了变压器油中溶解气体在线监测预警装置,并建立了相对应的气体溯源校准方法。实验结果表明:在低、中、高不同浓度范围内,H_(2)、C_(2)H_(2)、CH_(4)、C_(2)H_(4)、C_(2)H_(6)、CO及CO_(2)的最大相对误差分别为-12.51%、-29.25%、-20.04%、-19.31%、19.11%、-26.7%及-20.18%,均在±30%误差范围内。总烃(C1+C2)的最大相对误差为18.87%,处于±20%误差限值内,满足技术规范要求。低、中、高浓度范围内总烃的RSD分别为2.46%、2.87%和4.3%,均不大于5%,表现出良好重复性。该装置可满足电力行业恶劣工况条件的监测需求,可为电力行业油中溶解气体在线监测提供有效的实践经验。

The related parameters of dissolved gas in oil can well characterize the working state of equipment.Therefore,the analysis of dissolved gas in transformer oil is of great significance to predict transformer latent faults and prevent large-scale power failure.Aiming at the disadvantages of existing chromatography detection,based on the principle of laser photoacoustic spectroscopy,a gas detection module is designed,an online monitoring and early warning device for dissolved gas in transformer oil is developed,and the corresponding gas traceability calibration method is established.The experimental results show that the maximum relative errors of H_(2),C_(2)H_(2),CH_(4),C_(2)H_(4),C_(2)H_(6),CO and CO_(2)in the low,medium and high concentration ranges are-12.51%,-29.25%,-20.04%,-19.31%,19.11%,-26.7%and-20.18%,respectively,all within the error range of±30%.The maximum relative error of total hydrocarbon(C1+C2)is 18.87%,which is within the error limit of±20%,which meets the requirements of the technical specifications.The RSD of total hydrocarbons in the low,medium and high concentration ranges are_(2).46%,2.87%and 4.3%,respectively,all of which are not more than 5%,showing good repeatability.The device can meet the monitoring needs of harsh working conditions in the power industry,and can provide effective practical experience for the online monitoring of dissolved gas in oil in the power industry.