The loss changing with time was studied by using the cubic exponential smoothing time series pre-diction method, some issues such as the selection of the smoothing coefficientαand the prediction accuracy problem of single exponential smoothing, double exponential smoothing and cubic exponential smoothing were discussed in this paper.The results showed that the mean absolute error of prediction was minimum if smoothing coefficientαwas 0.6,and the result of cubic exponential smoothing prediction coincided prefer-ably to measure value.%利用三次指数平滑时间序列预测方法，研究了损耗随时间变化的变化情况，分析了三次指数平滑系数α的选择，探讨了一次平滑、二次平滑和三次平滑的精度问题。结果表明，α=0.6时，三次平滑预测平均绝对误差最小，预测结果与实测值吻合度较好。
展开▼
