The main contents of this paper are to verify the environmental factors affecting the power generation of floating photovoltaic systems and to present the power generation prediction model considering environmental factors by using regression analysis and neural networks studied during the last decade. This study focused on a comparative analysis of which model is best suited for the power generation prediction of the floating photovoltaic (PV) system. To compare the power generation characteristics of a floating and a land-based PV system, two identical 2.5 kW PV systems were installed—one on the water surface in the Boryeong Dam, Korea, and the other nearby on dry land—and their performances were compared. The solar irradiance of the floating PV system was 1.1% lower than that of the land-based PV. Nevertheless, the floating PV module temperature was 4.9% lower than that of the land-based PV, generating approximately 3% more power. Using the correlation analysis of data mining techniques, environmental factors affecting the efficiency of the floating PV system were investigated. The correlation coefficient between the module temperature and water temperature was $r=0.6317$ which proves that the high efficiency and low module temperature characteristics of the floating PV system, when compared with that of the land-based PV, are due to the water evaporation effect. Considering environmental factors, power-generation prediction models based on regression analysis and neural networks are presented, and their accuracies are compared. This comparison confirms that the accuracy of the power generation prediction model using neural networks was approximately 2.59% higher than that of the regression analysis method. As a result of adjusting the hidden nodes in the neural network algorithm, it was confirmed that a neural network algorithm with ten hidden nodes was most suitable for calculating the amount of power generation.

In South Korea, a water supply enhancement project is being carried out to preemptively respond to drought and water loss by reducing pipeline leakages and supplying stable tap water through the maintenance of an aging water supply network. In order to reduce water leakage, a District Metered Area (DMA) was established to monitor and predict the minimum night flow based on flow data collected from IoT sensors. In this study, a model based on Multi-Layer Perceptron (MLP) and Long Short-Term Memory (LSTM) was constructed to predict the MNF (minimum night flow) of County Y. The prediction of MNF results was compared with the MLP networks and the LSTM model. The outcome showed that the LSTM-MNF model proposed in this study performed better than the MLP-MNF model. Therefore, the research methods of this study can contribute to technical support for leakage reductions by preemptively responding to the expected increase in leakage through the prediction of the minimum flow at night.

Ultrasonic concentration meters have widely been used at water purification, sewage treatment and waste water treatment plants to sort and transfer high concentration sludges and to control the amount of chemical dosage. When an unusual substance is contained in the sludge, however, the attenuation of ultrasonic waves could be increased or not be transmitted to the receiver. In this case, the value measured by a concentration meter is higher than the actual density value or vibration. As well, it is difficult to automate the residuals treatment process according to the various problems such as sludge attachment or sensor failure. An ultrasonic multi-beam concentration sensor was considered to solve these problems, but an abnormal concentration value of a specific ultrasonic beam degrades the accuracy of the entire measurement in case of using a conventional arithmetic mean for all measurement values, so this paper proposes a method to improve the accuracy of the sludge concentration determination by choosing reliable sensor values and applying a neuro-fuzzy learning algorithm. The newly developed meter is proven to render useful results from a variety of experiments on a real water treatment plant.

In the water treatment process, determining the appropriate dosing of coagulants is crucial for both ensuring water quality and optimizing operational costs. Traditional methods predominantly rely on heuristic approaches or manual calibrations. However, the actual reactions between various water qualities and chemicals are intricate and nonlinear, often making it challenging to stably secure the targeted water quality. In particular, due to industrialization-induced water pollution and rapid climate change, conventional automated water treatment systems alone are insufficient to adequately respond. This research introduces an innovative approach for clarifying and optimizing the dosing of coagulants in water treatment plants using Explainable Artificial Intelligence (XAI), specifically the SHAP and ICE techniques. These methods allow for a nuanced understanding of feature contributions and individual predictions. Our findings suggest that the XAI-driven approach not only improves the accuracy of coagulant dosing but also offers valuable insights to operators, promoting more informed decision-making. The integration of machine learning and XAI in the water treatment field promises a future of more efficient, transparent, and responsible operations.