Abstract The deterioration of transformer oil caused by particle contamination, especially copper particle, seriously affects the safe operation of oil-immersed transformers. Therefore, it is very important for the detection of metal particles in transformer oil. A new method based on laser-induced breakdown spectroscopy (LIBS) was first applied in the transformer oil diagnosis field. By screening out effective spots, the correlation between Cu particle concentration and LIBS signal was obtained after testing a set of oil samples with concentration range from 0.54 to 12.4 μg/g; the calibration curves of Cu I 324.75 nm /CN 358.6 nm and Cu I 327.39 nm /CN 358.6 nm were therefore established. The results shown high linearity, and the determination coefficient R2 of the two calibration curves in range from 0.54 to 3.1 μg/g were superior to 0.99. The limit of detection (LOD) inferred from Cu I 324.75 nm /CN 358.6 nm fitting curve was 0.77 μg/g and met the critical value specified by industry standards. Due to its flexibility, it can be used as a new-type of particle concentration detection method for transformer oil.

中文翻译：

---

激光诱导击穿光谱法检测油浸式变压器铜颗粒污染

摘要 颗粒污染，尤其是铜颗粒引起变压器油变质，严重影响油浸式变压器的安全运行。因此，对变压器油中金属颗粒的检测非常重要。一种基于激光诱导击穿光谱（LIBS）的新方法首次应用于变压器油诊断领域。通过筛选有效斑点，对一组浓度范围为0.54～12.4 μg/g的油样进行检测，得到Cu颗粒浓度与LIBS信号的相关性；因此建立了Cu I 324.75 nm /CN 358.6 nm和Cu I 327.39 nm /CN 358.6 nm的校准曲线。结果显示出良好的线性，两条校准曲线在0.54～3.1 μg/g范围内的测定系数R2均优于0.99。从 Cu I 324.75 nm /CN 358.6 nm 拟合曲线推断的检测限 (LOD) 为 0.77 μg/g，符合行业标准规定的临界值。由于其灵活性，可作为一种新型的变压器油颗粒浓度检测方法。