说明：  在时间序列分段线性表示 PLR 基础上 ,提出用相对点平均误差度量子序列过程状态变化程度 ,改 进过程数据 PLR模型分段算法 ,克服采用单一误差算法的模型失配问题 ,更加准确地反应过程状态的变化。
(:Based on PLR piecewise linear representation of time series,RPAE relative point average error is pro2 posed to measure linear degree of process data segmentation and to develop the algorithm of PLR. The enhanced al2 gorithm overcomes the problem of model mismatch and describes the obvious change of process data precisely. Key words:data mining process data representation PLR model)