In rural areas, there is limited monitoring of drinking water quality. Reliable water quality monitoring stations are expensive and require high costs for maintenance and calibration process. In this paper, the development of a sustainable water quality monitoring system is proposed. The green analytics principles were considered for developing the proposed system to reduce the measurement’s time consumption and labor cost. Five water quality parameters [pH, oxidation reduction potential (ORP), dissolved oxygen (DO), electrical conductivity (EC), and temperature] have been measured using the developed system. The overall drinking water quality is measured by the proposed partial least squares regression (PLSR) model. The developed system’s performance is determined by mean average percentage error (MAPE), root-mean-square error (RMSE), and R². The traceability of water quality sensors is defined with required uncertainty in water quality parameters. The measured uncertainty is 0.002, 0.892, 0.015, 0.029, and 0.017 for pH, EC, DO, ORP, and temperature, respectively. The relation between estimated and predicted water quality parameters (R² > 0.93) shows that the developed system can be a suitable replacement for traditional water quality monitoring techniques.

在农村地区，对饮用水质量的监测很有限。可靠的水质监测站价格昂贵，维护和校准过程需要高成本。本文提出了一种可持续的水质监测系统的开发。在开发提议的系统时考虑了绿色分析原则，以减少测量的时间消耗和劳动力成本。已使用开发的系统测量了五个水质参数 [pH、氧化还原电位 (ORP)、溶解氧 (DO)、电导率 (EC) 和温度]。总体饮用水质量由建议的偏最小二乘回归 (PLSR) 模型测量。开发系统的性能由平均百分比误差 (MAPE)、均方根误差 (RMSE) 和R ²。水质传感器的可追溯性定义为水质参数所需的不确定性。pH、EC、DO、ORP 和温度的测量不确定度分别为 0.002、0.892、0.015、0.029 和 0.017。估计和预测的水质参数之间的关系 ( R ²  > 0.93) 表明，开发的系统可以很好地替代传统的水质监测技术。