【摘要】 以灰色系统理论为基础,研究变压器油中溶解气体体积分数的灰色预测新模型及其应用。建立了新的改进型非等间隔灰色预测模型。该模型首先针对变压器油中溶解气体分析试验在时间上具有非等间隔的特点,把非等间隔序列经一次拟合变换成等间隔序列。为了提高预测精度,采用一次指数平滑运算对强随机性的电力变压器的油中溶解气体数据进行平滑处理,对前面得到的等间隔数据重新生成,构成一个新序列。此外,还从背景值改造方面对预测模型进行了改进。通过2个实例进行了对比试验,表明所提出的非等间隔的模型具有明显的优势,预测效果普遍高于传统的灰色预测模型,验证了所提出方法的正确性和方案的可行性。更多还原

【Abstract】 Based on the gray system theory,the gray prediction model for dissolved gases in power transformer and its application are studied. To obtain higher accuracy,an improved grey prediction model is developed. As the intervals between two analytic experiments of transformer dissolved gas are unfixed,the data sequence sampled with unequal intervals is converted into the data sequences with equal intervals by first -order fitting,which is then smoothed by the first -order exponent flatness operation to improve prediction accuracy because of its strong randomness. The background value of the prediction model is also reconstructed. Two cases show that,compared with the traditional model,the improved model has better accuracy,which verifies its correctness and feasibility.更多还原