Soft Sensor with Adaptive Algorithm for Filter Gain Correction in the Online Monitoring System of a Polluted River

Abstract

This paper proposes the realization of a soft sensor using an adaptive algorithm with proportional correction of the gain coefficient for monitoring river water quality. This algorithm makes it possible to monitor online signals of an object described by nonlinear ordinary differential equations. Simulation studies of a biochemically polluted river, for which the water quality was represented by biochemical oxygen demand (BOD) indices and the dissolved oxygen (DO) deficit, were carried out. The algorithm concept uses only online measurements of the object, and adaptive changes in the gain coefficient are determined based on the adaptation error adopted for this purpose. Simulation results indicated the correct functioning of the soft sensor even for inaccurately identified parameters of the mathematical model and for unknown values and intensity of disturbances affecting the object. The quality of the signals monitored via a soft sensor implemented in this way was determined with the root-mean-squared error (RMSE) and mean percentage error (MPE) indicators and compared with the Kalman filter.

State Analysis of the Water Quality in Rivers in Consideration of Diffusion Phenomenon

“…At temperature 20 Dissolved oxygen is one of the most important indicators of water quality: it influences the chemical and biochemical processes that occur in water. DO is an important indicator for the living organisms that make up a river ecosystem [24][25][26]. Aerobic processes occurring in natural waters are essential for reducing pollution in water.…”

From Fully Physical to Virtual Sensing for Water Quality Assessment: A Comprehensive Review of the Relevant State-of-the-Art

“…The application of virtual (or soft) sensing for surface and groundwater quality assessment is emerging. This is evidenced by the low number of publications (16) and the fact that 63% of the articles [22,29,55,85,[94][95][96][97][98][99] were published in the last three years (2019-2021). Therefore, as seen in Figure 4, it is not surprising that artificial neural network (ANN) related algorithms are the most applied ML techniques.…”

“…Consideration of seasonal effects in WQ assessment is vital since WQ variables are often subjected to high variation in concentration based on the time of the year. Based on the papers reviewed, it is observed that four studies [22,95,97,100] did not report on the data collection interval used in building their predictive models. This points to the need for a more rigorous scientific reporting of data collection in data-driven modeling papers.…”

“…Additionally, a less complex predictive model with fewer features will require minimal resources to train and, therefore, will increase the model interpretability by the end-user [85]. Based on Table 7, two studies [97,100] did not give any detail concerning the data used, and therefore it is not clear whether it was processed or not. Seven studies pre-processed the data from the fourteen that provided some details while the other seven did not.…”

Feasible Waveguide Method for Obtaining Electromagnetic Properties of Biaxial Bianisotropic Materials with Strong Magneto-Electric Coupling

“…For the quantitative effect of noise on the extracted parameters, we can perform some numerical analysis using the mean percentage error (MPE) (Hawro et al, 2019 Here, n is the number of frequency points, j is the frequency point, and j S and j R are the values synthesized and retrieved at the frequency point j , respectively. Accordingly, Table 1 shows the MPE values among the synthesized and retrieved y  values of a sample when there is no noise (Fig.…”