在线光谱水质检测仪器是现代水资源环境监测技术的重要发展方向之一, 具有多参数监测和准确度高、 重复性好的技术优势, 然在线被测水样的光谱信号处理是其关键核心技术, 为此, 基于连续光谱分析, 建立了在线被测水样光谱测量信号的数学模型, 提出了基于双波长光强比值不变性的光谱测量信号系统误差处理方法, 并结合小波多分辨率滤波噪声处理技术, 系统研究了基于在线被测水质参数光谱特征的背景干扰处理方法。 以上信号处理方法应用于自主研制的多参数光谱水质监测仪器, 在线检测标准环境水样及实际环境水样中的化学需氧量、 六价铬和阴离子表面活性剂等水质参数, 并与国家标准分析方法展开了现场对比测试, 仪器的关键参数重复性(相对标准偏差RSD≤10%)与准确度(实际水样比对试验相对误差A≤10%)均达到并优于国家环境保护技术标准要求, 表明该信号处理方法能够有效消除在线水质检测光谱测量信号的系统误差及噪声与背景干扰, 对于提升光谱水质监测仪器的技术性能具有重要的作用。

Based on continuous spectrum analysis, the mathematical model for spectrum signal was established. And the spectrum signal’s systematic error processing method based on the invariance of the ratio of the light intensities at any two wavelengths in the range of continuous spectrum was put forward. Combined with wavelet multi-resolution filtering noise processing techniques, the background interference processing method was established based on the spectral characteristics of the measured water quality parameter. These signal processing methods were applied to our independently developed multi-parameter water quality monitoring instrument to on-line measure COD (chemical oxygen demand), six valence chromium and anionic surfactant in the normative and actual environmental water samples, and the monitoring instrument had good repeatability (10%) and high accuracy (±10%) to meet the technical requirements of national environmental protection standards, which was verified by the contrast experiment with China national standard analysis method for determination of the three water quality parameter. The results showed that the researched signal processing methods were able to effectively reduce the spectrum signal’s systematic error and the interference from noise and background, which was very important to improve the water quality monitoring instrument’s technical function.