An advanced deep learning model for predicting water quality index

Ecological Indicators, Год журнала: 2024, Номер 160, С. 111806 - 111806

Опубликована: Фев. 29, 2024

Predicting a water quality index (WQI) is important because it serves as an metric for assessing the overall health and safety of bodies. Our paper develops new hybrid model predicting WQI. The study uses combination convolutional neural network (CNN), clockwork recurrent (Clockwork RNN), M5 Tree (CNN-CRNN-M5T) to predict M5T lacks advanced operators extracting meaningful data from parameters, so enhances its ability analyze intricate patterns. general linear analysis variance (GLM-ANOVA) improved version ANOVA. GLM-ANOVA determine significant inputs. As all input variables had p < 0.050, they were defined variables. Results showed that NH-NL PH highest lowest impact, respectively. used CNN-CRNN-M5T, CNN-CRNN, CRNN-M5T, CNN-M5T, CRNN, CNN, models WQI large basin in Malaysia. CNN-CRNN decreased testing mean absolute error (MAE) by 2.1 %, 12 15 CNN-CRNN-M5T increased Nash–Sutcliffe efficiency coefficient other 4–20 % 2.1–19 was reliable tool spatial temporal predictions

Язык: Английский

---

Assessing the effectiveness of long short-term memory and artificial neural network in predicting daily ozone concentrations in Liaocheng City

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Фев. 25, 2025

Ozone pollution affects food production, human health, and the lives of individuals. Due to rapid industrialization urbanization, Liaocheng has experienced increasing ozone concentration over several years. Therefore, become a major environmental problem in City. Long short-term memory (LSTM) artificial neural network (ANN) models are established predict concentrations City from 2014 2023. The results show general improvement accuracy LSTM model compared ANN model. Compared ANN, an increase determination coefficient (R2), value 0.6779 0.6939, decrease root mean square error (RMSE) 27.9895 μg/m3 27.2140 absolute (MAE) 21.6919 20.8825 μg/m3. prediction is superior terms R, RMSE, MAE. In summary, promising technique for predicting concentrations. Moreover, by leveraging historical data enables accurate predictions future on global scale. This will open up new avenues controlling mitigating pollution.

Язык: Английский

---

Comparative Analysis of Convolutional Neural Network-Long Short-Term Memory, Sparrow Search Algorithm-Backpropagation Neural Network, and Particle Swarm Optimization-Extreme Learning Machine Models for the Water Discharge of the Buzău River, Romania

Water, Год журнала: 2024, Номер 16(2), С. 289 - 289

Опубликована: Янв. 15, 2024

Modeling and forecasting the river flow is essential for management of water resources. In this study, we conduct a comprehensive comparative analysis different models built monthly discharge Buzău River (Romania), measured in upper part river’s basin from January 1955 to December 2010. They employ convolutional neural networks (CNNs) coupled with long short-term memory (LSTM) networks, named CNN-LSTM, sparrow search algorithm backpropagation (SSA-BP), particle swarm optimization extreme learning machines (PSO-ELM). These are evaluated based on various criteria, including computational efficiency, predictive accuracy, adaptability training sets. The obtained applying CNN-LSTM stand out as top performers, demonstrating superior efficiency high especially when set containing data series 1984 (putting Siriu Dam operation) September 2006 (Model type S2). This research provides valuable guidance selecting assessing prediction models, offering practical insights scientific community real-world applications. findings suggest that Model S2 preferred choice forecast predictions due its speed accuracy. S (considering recorded 2006) recommended secondary option. S1 (with period 1955–December 1983) suitable other unavailable. study advances field by presenting precise these their respective strengths

Язык: Английский

---

Accurate modeling and prediction of ozone mass transfer in a rotating packed bed based on multilayer perceptron

Chemical Engineering and Processing - Process Intensification, Год журнала: 2025, Номер unknown, С. 110163 - 110163

Опубликована: Янв. 1, 2025

Язык: Английский

---

Spatial Distribution of Particulate Matter in Iran from Internal Factors to the Role of Western Adjacent Countries from Political Governance to Environmental Governance

Earth Systems and Environment, Год журнала: 2024, Номер 8(1), С. 135 - 164

Опубликована: Янв. 1, 2024

Язык: Английский

---

Modeling abrupt changes in mine water inflow trends: A CEEMDAN-based multi-model prediction approach

Journal of Cleaner Production, Год журнала: 2024, Номер 439, С. 140809 - 140809

Опубликована: Янв. 24, 2024

Язык: Английский

---

Multiedge Graph Convolutional Network for House Price Prediction

Journal of Construction Engineering and Management, Год журнала: 2023, Номер 149(11)

Опубликована: Сен. 5, 2023

Accurate house price prediction allows construction investors to make informed decisions about the housing market and understand growth opportunities for development risks rewards of different projects. Machine learning (ML) models have been utilized as predictors, reducing decision-making costs, increasing reliability. To further improve reliability existing this study develops a hybrid multiedge graph convolutional network (GCN) that considers various relationships between records. The developed GCN receives richer input from multidependency information thus provides more reliable accounts changes based on neighborhood, building age, number bedrooms. Compared other ML approaches, predictor displayed good accuracy while providing valuable insights into factors affect price, such desirability neighborhoods age.

Язык: Английский

---

Unveiling the SALSTM-M5T model and its python implementation for precise solar radiation prediction

Energy Reports, Год журнала: 2023, Номер 10, С. 3402 - 3417

Опубликована: Окт. 11, 2023

The world is increasingly embracing cleaner and more sustainable energy sources, with solar playing a crucial role in mitigating greenhouse gas emissions addressing climate change. Accurate radiation predictions are vital for optimizing resource utilization identifying suitable locations power plants. Therefore, our study introduces new model to advance renewable systems. In this paper, we suggest novel hybrid model, Self-attention (SA) mechanism-long short-term memory neural network (LSTM)-M5Tree (SALSTM5T) predicting radiation. SALSTM-M5T combines the advantages of Self-attention- LSTM (SALSTM) M5Tree models. component captures long-term dependencies. SA improves accuracy M5T models by focusing on relevant input features at different time steps. utilizes K-fold cross-validation overcome limitations traditional methods determining size training testing data. By combining SALSTM models, research presents framework accurate prediction. Existing techniques developed using cross-validation. Furthermore, paper emphasizes practical applications prediction, such as areas plants production. Our concluded that self-attention mechanism improved efficiency analyzing series data these can attend important features. centralized root mean square error (CRMSE) SALSTM-M5T, LSTM-M5T, LSTM, ANN, was 0.04, 0.17, 0.25, 0.49, 0.70, respectively. models' correlation coefficients were 0.99, 0.98, 0.96, 0.89, 0.82, contributes advancing planning decision-making. main innovation current article development capabilities which self-attention, techniques, proposed an effective approach results show its superiority over other terms suitability, be useful field energy, site selection optimization aligns advancement digital sensors enhance

Язык: Английский

---

Hazard Susceptibility Mapping with Machine and Deep Learning: A Literature Review

Remote Sensing, Год журнала: 2024, Номер 16(18), С. 3374 - 3374

Опубликована: Сен. 11, 2024

With the increase in climate-change-related hazardous events alongside population concentration urban centres, it is important to provide resilient cities with tools for understanding and eventually preparing such events. Machine learning (ML) deep (DL) techniques have increasingly been employed model susceptibility of This study consists a systematic review ML/DL applied air pollution, heat islands, floods, landslides, aim providing comprehensive source reference both modelling approaches. A total 1454 articles published between 2020 2023 were systematically selected from Scopus Web Science search engines based on queries selection criteria. extracted categorised using ad hoc classification. Consequently, general approach was consolidated, covering data preprocessing, feature selection, modelling, interpretation, map validation, along examples related global/continental data. The most frequently across various hazards include random forest, artificial neural networks, support vector machines. also provides, per hazard, definition, requirements, insights into used, including state-of-the-art novel

Язык: Английский

---

Machine Learning and Multicriteria Analysis for Prediction of Compressive Strength and Sustainability of Cementitious Materials

Case Studies in Construction Materials, Год журнала: 2024, Номер unknown, С. e04080 - e04080

Опубликована: Дек. 1, 2024

Язык: Английский

---