SOFT SENSOR DEVELOPMENT BASED ON EXTREME LEARNING MACHINE FOR WATER QUALITY MONITORING

摘要

A major problem in water treatment plants is the continuous difficulty faced in online measurement by means of dedicated measuring hardware and laboratory analysis of certain variables especially related to the composition. Actually, for several reasons, such as the high cost of some sensors, their number, the dedicated time to check out the sensors, the cleaning operation, the calibration routines and the sensor replacement, make their proper operation hard to ensure. Furthermore, in water quality monitoring, there is a huge number of heterogeneous sensors which may be time-consuming in the measurement and processing stages. Nevertheless, soft sensor approach can provide an effective and economic way to solve this problem and any cases of sensor failure. In this paper, we attempt to use Extreme Learning Machine (ELM) in order to develop a soft sensor which is used frequently in water quality monitoring. A comparative study with Support Vector Machine (SVM) in term of learning time and other parameters for regression and classification, is presented. The main objective is to set up a system architecture based on a soft sensor in order to make an adapted decision to the control and monitoring of water quality issues. Discussion of the results should lead to a decisive choice of the most suitable method. An application is provided to focus on the interest of using a chlorine soft sensor as it is accurate, efficient and less cost-effective tool.