A novel hybrid BPNN model based on adaptive evolutionary Artificial Bee Colony Algorithm for water quality index prediction

Ecological Indicators, Journal Year: 2023, Volume and Issue: 146, P. 109882 - 109882

Published: Jan. 9, 2023

With the accelerated industrialization and urbanization process, water pollution in rivers is being increasingly worsened, has caused a series of ecological environmental issues. The prediction river quality index (WQI) prerequisite for prevention management. However, data non-smooth non-linear, strong coupling relationship between different parameters that influence each other observed, making it an inevitable problem to accurately predict parameters. To this end, combination machine learning intelligent optimization algorithms was hereby used break dilemma. Specifically, Back Propagation Neural Network (BPNN) model established using Artificial Bee Colony (ABC) algorithm, with three adaptive evolutionary strategies, i.e., dynamic factors, probability selection gradient initialization combined form Adaptive Evolutionary (AEABC) algorithm. experimental results algorithm demonstrate AEABC-BPNN only requires 14 iterations converge case. predictions WQI can reduce error evaluation indicators mean square (MSE) 0.2745, which at least 25.2% lower than those rest compared, absolute percentage (MAPE) 7.58%. In four WQIs, interval coverage (PICP) reaches 100%. Besides, robustness testing experiments were also designed verify still outperforms terms accuracy when guided by historical data. proposed plays pivotal role management lakes, scientific significance future protection.

Language: Английский

---

Applications of machine learning to water resources management: A review of present status and future opportunities

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: 441, P. 140715 - 140715

Published: Jan. 11, 2024

Water is the most valuable natural resource on earth that plays a critical role in socio-economic development of humans worldwide. used for various purposes, including, but not limited to, drinking, recreation, irrigation, and hydropower production. The expected population growth at global scale, coupled with predicted climate change-induced impacts, warrants need proactive effective management water resources. Over recent decades, machine learning tools have been widely applied to resources management-related fields often shown promising results. Despite publication several review articles applications water-related fields, this paper presents first time comprehensive techniques management, focusing achievements. study examines potential advanced improve decision support systems sectors within realm which includes groundwater streamflow forecasting, distribution systems, quality wastewater treatment, demand consumption, marine energy, drainage flood defence. This provides an overview state-of-the-art approaches industry how they can be ensure supply sustainability, quality, drought mitigation. covers related studies provide snapshot industry. Overall, LSTM networks proven exhibit reliable performance, outperforming ANN models, traditional established physics-based models. Hybrid ML exhibited great forecasting accuracy across all showing superior computational power over ANNs architectures. In addition purely data-driven physical-based hybrid models also developed prediction performance. These efforts further demonstrate Machine powerful practical tool management. It insights, predictions, optimisation capabilities help enhance sustainable use development, healthy ecosystems human existence.

Language: Английский

---

Prediction of Future Land Use/Land Cover Changes Using a Coupled CA-ANN Model in the Upper Omo–Gibe River Basin, Ethiopia

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(4), P. 1148 - 1148

Published: Feb. 20, 2023

Land use/land cover change evaluation and prediction using spatiotemporal data are crucial for environmental monitoring better planning management of land use. The main objective this study is to evaluate changes the time period 1991–2022 predict future CA-ANN model in Upper Omo–Gibe River basin. Landsat-5 TM 1991, 1997, 2004, Landsat-7 ETM+ 2010, Landsat-8 (OLI) 2016 2022 were downloaded from USGS Earth Explorer Data Center. A random forest machine learning algorithm was employed LULC classification. classification result evaluated an accuracy assessment technique assure correctness method employing kappa coefficient. Kappa coefficient values indicate that there strong agreement between classified reference data. Using MOLUSCE plugin QGIS model, predicted. Artificial neural network (ANN) cellular automata (CA) methods made available modeling via plugin. Transition potential computed, predicted model. An overall 86.53% value 0.82 obtained by comparing actual with simulated same year. findings revealed 2037, agricultural (63.09%) shrubland (5.74%) showed significant increases, (−48.10%) grassland (−0.31%) decreased. From 2037 2052, built-up area (2.99%) a increase, (−2.55%) decrease. 2052 2067, projected simulation (3.15%) (0.32%) increased, (−1.59%) (−0.56%) decreases. According study’s findings, drivers expansion areas land, which calls thorough investigation additional models give planners policymakers clear information on their effects.

Language: Английский

---

Urban river water quality monitoring based on self-optimizing machine learning method using multi-source remote sensing data

Ecological Indicators, Journal Year: 2022, Volume and Issue: 146, P. 109750 - 109750

Published: Dec. 2, 2022

Urban rivers are complex ecosystems that directly determine the living environment of human beings. Monitoring urban river water quality indexes is a challenge in evaluation. The purpose this study was to propose multi-source remote sensing inversion method based on small number samples solve problem scale inconsistency among data, so as achieve large-scale and efficient quality. Since there very important nonlinear relationships must be solved between simple ground point data inversion, novel self-optimizing machine learning monitoring proposed, which can automatically find optimal parameters model from samples, reduce training time. Meanwhile, order strengthen correlation feature enhancement used for generating input data. Moreover, quantity quality, spatial mapping consistency information since these have different characteristics. experimental results show unmanned aerial vehicle (UAV) images, R2 chlorophyll (Chla), turbidity (TUB), ammonia nitrogen (NH3-N) reached 0.917, 0.877 0.846, respectively. Using satellite image, Chla, TUB, NH3-N reach 0.827, 0.679 0.779, This provides new way realize integration air-space-ground inland future.

Language: Английский

---

Suspended sediment load prediction modelling based on artificial intelligence methods: The tropical region as a case study

Heliyon, Journal Year: 2023, Volume and Issue: 9(8), P. e18506 - e18506

Published: July 20, 2023

The impact of the suspended sediment load (SSL) on environmental health, agricultural operations, and water resources planning, is significant. deposit SSL restricts streamflow region, affecting aquatic life migration finally causing a river course shift. As result, data sediments their fluctuations are essential for number authorities especially decision makers. prediction often difficult due to issues such as site-specific data, models, lack several substantial components use in prediction, complexity its pattern. In past two decades, many machine learning algorithms have shown huge potential prediction. However, these models did not provide very reliable results, which led conclusion that accuracy should be improved. order solve concerns, this research proposes Long Short-Term Memory (LSTM) model proposed was applied Johor River located Malaysia. study allocated flow period 2010 2020. current research, four alternative models—Multi-Layer Perceptron (MLP) neural network, Support Vector Regression (SVR), Random Forest (RF), Short-term were investigated predict load. attained high correlation value between predicted actual (0.97), with minimum RMSE (148.4 ton/day MAE (33.43 ton/day).and can thus generalized application similar rivers around world.

Language: Английский

---

A novel problem-solving method by multi-computational optimisation of artificial neural network for modelling and prediction of the flow erosion processes

Engineering Applications of Computational Fluid Mechanics, Journal Year: 2024, Volume and Issue: 18(1)

Published: Jan. 7, 2024

This research aims to forecast, using various criteria, the flow of soil erosion that will occur at a particular geographical location. As for training dataset, 80% dataset from sample sites, four hybrid algorithms, namely heap-based optimizer (HBO), political (PO), teaching-learning based optimization (TLBO), and backtracking search algorithm (BSA) combined with artificial neural network (ANN) was used create an susceptibility model establishes unique original approach. After it confirmed be successful, algorithms were applied map this area, demonstrating integrity results. The AUC values computed every optimisation in study. optimal estimated accuracy indices populations 450 determined 0.9846 BSA-MLP databases. maximum value HBO-MLP databases different swarm sizes 0.9736. A size 350–300 is considered forecasting mapping models. With same constraints, TLBO-MLP scenario 0.996. 150 conditions train PO-MLP model, 0.9845. According these findings, worked best 50 150, respectively.

Language: Английский

---

A novel interval decomposition correlation particle swarm optimization-extreme learning machine model for short-term and long-term water quality prediction

Journal of Hydrology, Journal Year: 2023, Volume and Issue: 625, P. 130034 - 130034

Published: Aug. 11, 2023

Language: Английский

---

Monthly rainfall forecasting modelling based on advanced machine learning methods: tropical region as case study

Engineering Applications of Computational Fluid Mechanics, Journal Year: 2023, Volume and Issue: 17(1)

Published: Aug. 10, 2023

Existing forecasting methods employed for rainfall encounter many limitations, because the difficulty of underlying mathematical proceeding in dealing with patterning and imitation data. This study attempts to provide a robust methodology detecting nonlinearity pattern by integrating several optimizer algorithms an Artificial Neural Network (ANN). The Bee Colony, Particle Swarm Optimization, Imperialism Competitive Algorithm have been integrated improve optimize internal parameters ANN method. In Malaysia, real-world case was set up, model created using 54 years (1967–2020) worth local monthly artificial neural network method is being utilized real-time. A variety types were evaluated various input information goal producing accurate forecasts. Statistical analysis conducted statistical indicators evaluate model's accuracy rainfall. revealed that based on integration Imperial (ICA-ANN) outperformed other predictive models. results confirmed proposed promising high accuracy.

Language: Английский

---

Enhancing sediment transport predictions through machine learning-based multi-scenario regression models

Results in Engineering, Journal Year: 2023, Volume and Issue: 20, P. 101585 - 101585

Published: Nov. 14, 2023

Machine learning is one effective way of increasing the accuracy sediment transport models. captures patterns in sequence both structured and unstructured data uses it for forecasting. In this research, different regression models were train to forecast using 8 years measured collected Sg. Linggui suspended station. Data from scenarios used where each scenario indicates number lags. Seven models, namely, Linear Regression, Regression Trees, Support Vector Machines, Gaussian Process Kernel Approximation, Ensemble Neural Network trained compared. The evaluated Root Mean Square Error (RMSE) Coefficient Determination (R2). best-performing two types chosen they tested test find Relative Percentage (RPE) predicted data. Exponential model performs much better than other terms RMSE R2 values. When exponential all 3 are compared, seems have a better-performing but only by very small margin, after testing data, result shows has less RPE compared Hence, can be deduced that gaussian process overall RSME, R2, RPE.

Language: Английский

---

Evaluation of the water quality of a high Andean lake using different quantitative approaches

Ecological Indicators, Journal Year: 2023, Volume and Issue: 154, P. 110924 - 110924

Published: Sept. 8, 2023

This study assessed a high Andean lake's trophic state and water quality using methodologies with eutrophication indexes. Water samples were collected at six points in the lake, monthly frequency, for three winter summer months. Dissolved oxygen, pH, phosphates, nitrates, transparency, chlorophyll-a, fecal coliforms, biological oxygen demand (BOD), temperature, turbidity determined each point. The of lake was categorized by applying Organization Economic Cooperation Development (OECD) index, Carlson's index (CTSI) (TRIX). In addition, National Sanitation Foundation (NSF-WQI), Canadian Quality Index (CCME-WQI) Oregon (OWQI) used to evaluate quality. Results indicated that had level eutrophication, suggesting an excessive accumulation nutrients water. CTSI TRIX showed hyper-eutrophic state, while according OECD methodology, related phosphorus transparency hypereutrophic, chlorophyll, it varied from mesotrophic eutrophic. NSF classified average quality, CCME fair OWQI as very poor. Therefore, andean indexes presented significant differences based on physicochemical characteristics. human influence identified main cause including tourism agriculture. These results suggest measures should be taken reduce activity area control pollution lake.

Language: Английский

---