Abstract: In water quality testing, the data is processed by the chromatography analysis method. Based on the requirement of the process of chromatography signal in water quality testing, the differential operation is adopted to realize the accurate identification of the starting point and the ending point of the chromatographic peak. ?Under the precondition of chromatographic signal data smoothing processing, this algorithm judges whether the chromatographic signal value of two adjacent points in a group of data complies with the average value of all difference value in this group, so as to determine the staring point of the peak rising channel and a termination point of the peak downward channel. Using the starting point and ending point obtained in the previous step, it can identify the type of peak accurately. Experimental result shows that this algorithm has higher accuracy and faster operation, and it can avoid the instability caused by the threshold set artificially as well as realize?the?automatic?spectrum?peak?recognition.

Key words: water quality testing, differential operation, chromatography peaks&rsquo, identification, type of chromatography peak, starting point of peak rising channel, ending point of peak rising channel

摘要： 水质检测中常用色谱分析法处理数据，针对水质检测中对获取色谱信号数据分析的要求，采用差分运算，实现色谱谱峰起始点和终止点准确识别。该算法在色谱信号数据平滑处理前提下，通过判断一组数据内相邻两点色谱信号数值变化是否符合该组数据所有差值的平均值来确定峰的上升通道起始点和下降通道终止点，对识别出的起始点和终止点进行二次判断，可准确识别峰的类型。实验结果表明，该算法准确率高，运算速度更快，可以避免人工设定阈值时产生的不稳定性，实现了全自动化谱峰识别。

关键词: 水质检测, 差分运算, 谱峰识别, 峰类别, 峰上升通道起始点, 峰下降通道终止点