RESUMEN
The effective reduction of hazardous organic pollutants in wastewater is a pressing global concern, necessitating the development of advanced treatment technologies. Pollutants such as nitrophenols and dyes, which pose significant risks to both human and aquatic health, making their reduction particularly crucial. Despite the existence of various methods to eliminate these pollutants, they are not without limitations. The utilization of nanomaterials as catalysts for chemical reduction exhibits a promising alternative owing to their distinguished catalytic activity and substantial surface area. For catalytically reducing the pollutants NaBH4 has been utilized as a useful source for it because it reduces the pollutants quiet efficiently and it also releases hydrogen gas as well which can be used as a source of energy. This paper provides a comprehensive review of recent research on different types of nanomaterials that function as catalysts to reduce organic pollutants and also generating hydrogen from NaBH4 methanolysis while also evaluating the positive and negative aspects of nanocatalyst. Additionally, this paper examines the features effecting the process and the mechanism of catalysis. The comparison of different catalysts is based on size of catalyst, reaction time, rate of reaction, hydrogen generation rate, activation energy, and durability. The information obtained from this paper can be used to steer the development of new catalysts for reducing organic pollutants and generation hydrogen by NaBH4 methanolysis.
Asunto(s)
Contaminantes Ambientales , Contaminantes Químicos del Agua , Purificación del Agua , Humanos , Aguas Residuales , Purificación del Agua/métodos , Catálisis , Hidrógeno , Contaminantes Químicos del Agua/análisisRESUMEN
The study explores the spatio-temporal variation of water quality parameters in the Hooghly estuary, which is considered an ecologically-stressed shallow estuary and a major distributary for the Ganges River. The estimated parameters are chlorophyll-a, total suspended matter (TSM), and chromophoric dissolved organic matter (CDOM). The Sentinel-3 OLCI remote sensing imageries were analyzed for the duration of October 2018 to February 2019. We observed that the water quality of the Hooghly estuaries is comparatively low-oxygenated, mesotrophic, and phosphate-limited. Ongoing channel dredging for maintaining shipping channel depth keeps the TSM in the estuary at an elevated level, with the highest amount of TSM observed during March of 2019 (41.59g m-3) at station A, upstream point. Since the pre-monsoon season, TSM data shows a decreasing trend towards the mouth of the estuary. Chl-a concentration is higher during pre-monsoon than monsoon and post-monsoon periods, with the highest value observed in April at 1.09 mg m-3 in station D during the pre-monsoon period. The CDOM concentration was high in the middle section (January-February) and gradually decreased towards the estuary's head and mouth. The highest CDOM was found in February at locations C and D during the pre-monsoon period. Every station shows a significant correlation among CDOM, TSM, and Chl-a measured parameters. Based on our satellite data analysis, it is recommended that SNAP C2RCC be regionally used for TSM, Chl-a, and CDOM for water quality product retrieval and in various algorithms for the Hooghly estuary monitoring.
Asunto(s)
Monitoreo del Ambiente , Calidad del Agua , Estuarios , Clorofila A , RíosRESUMEN
The massive environmental noise interference and insufficient effective sample degradation data of the intelligent fault diagnosis performance methods pose an extremely concerning issue. Realising the challenge of developing a facile and straightforward model that resolves these problems, this study proposed the One-Dimensional Convolutional Neural Network (1D-CNN) based on frequency-domain signal processing. The Fast Fourier Transform (FFT) analysis is initially utilised to transform the signals from the time domain to the frequency domain; the data was represented using a phasor notation, which separates magnitude and phase and then fed to the 1D-CNN. Subsequently, the model is trained with White Gaussian Noise (WGN) to improve its robustness and resilience to noise. Based on the findings, the proposed model successfully achieved 100% classification accuracy from clean signals and simultaneously achieved considerable robustness to noise and exceptional domain adaptation ability. The diagnosis accuracy retained up to 97.37%, which was higher than the accuracy of the CNN without training under noisy conditions at only 43.75%. Furthermore, the model achieved an accuracy of up to 98.1% under different working conditions, which was superior to other reported models. In addition, the proposed model outperformed the state-of-art methods as the Signal-to-Noise Ratio (SNR) was lowered to -10 dB achieving 97.37% accuracy. In short, the proposed 1D-CNN model is a promising effective rolling bearing fault diagnosis.
Asunto(s)
Algoritmos , Redes Neurales de la Computación , Análisis de Fourier , Procesamiento de Señales Asistido por Computador , Relación Señal-RuidoRESUMEN
The massive destruction and loss caused by the 2004 Sumatra-Andaman tsunami were attributed to the lack of knowledge on tsunami and low regional detection and communication systems for early warning in that region. This study aimed to identify locations at risk of impending tsunami from Andaman Sea for the safety of community and proper development planning at the coastal areas by providing an updated and revised inundation maps. The last study on this area was conducted several years ago which open the possibility to new findings. Generated by tsunami simulation models, the maps illustrate the extent and level of inundation to which the coastal community and infrastructure would be subjected. As a result of coastal changes and availability of better topographic data, the existing inundation maps for the coastal areas of northwest Peninsular Malaysia at risk to impending tsunami from the Andaman Sea are revised. This paper documented the computational setup leading to the generation of the revised inundation maps. The tsunami simulation model TUNA was used to simulate the generation, propagation, and subsequent run-up and inundation of tsunamis triggered by earthquakes of moment magnitudes (Mw) 8.5, 9.0, and 9.25 along the Sunda Trench. From the simulations, it was found that at Mw 9.25, Balik Pulau, Pulau Pinang would be subjected to inundation of as far as 3.47 km with 5.40-m-deep inundation at the highest section.
Asunto(s)
Terremotos , Tsunamis , Monitoreo del Ambiente , Indonesia , MalasiaRESUMEN
In the recent decade, deep eutectic solvents (DESs) have occupied a strategic place in green chemistry research. This paper discusses the application of DESs as functionalization agents for multi-walled carbon nanotubes (CNTs) to produce novel adsorbents for the removal of 2,4-dichlorophenol (2,4-DCP) from aqueous solution. Also, it focuses on the application of the feedforward backpropagation neural network (FBPNN) technique to predict the adsorption capacity of DES-functionalized CNTs. The optimum adsorption conditions that are required for the maximum removal of 2,4-DCP were determined by studying the impact of the operational parameters (i.e., the solution pH, adsorbent dosage, and contact time) on the adsorption capacity of the produced adsorbents. Two kinetic models were applied to describe the adsorption rate and mechanism. Based on the correlation coefficient (R2) value, the adsorption kinetic data were well defined by the pseudo second-order model. The precision and efficiency of the FBPNN model was approved by calculating four statistical indicators, with the smallest value of the mean square error being 5.01 × 10-5. Moreover, further accuracy checking was implemented through the sensitivity study of the experimental parameters. The competence of the model for prediction of 2,4-DCP removal was confirmed with an R2 of 0.99.
Asunto(s)
Nanotubos de Carbono/química , Redes Neurales de la Computación , Fenoles/química , Solventes/química , Contaminantes Químicos del Agua/química , Adsorción , Algoritmos , Cinética , Modelos Teóricos , Purificación del AguaRESUMEN
In this study, the analysis of the extreme sea level was carried out by using 10 years (2007-2016) of hourly tide gauge data of Karachi port station along the Pakistan coast. Observations revealed that the magnitudes of the tides usually exceeded the storm surges at this station. The main observation for this duration and the subsequent analysis showed that in June 2007 a tropical Cyclone "Yemyin" hit the Pakistan coast. The joint probability method (JPM) and the annual maximum method (AMM) were used for statistical analysis to find out the return periods of different extreme sea levels. According to the achieved results, the AMM and JPM methods erre compatible with each other for the Karachi coast and remained well within the range of 95% confidence. For the JPM method, the highest astronomical tide (HAT) of the Karachi coast was considered as the threshold and the sea levels above it were considered extreme sea levels. The 10 annual observed sea level maxima, in the recent past, showed an increasing trend for extreme sea levels. In the study period, the increment rates of 3.6 mm/year and 2.1 mm/year were observed for mean sea level and extreme sea level, respectively, along the Karachi coast. Tidal analysis, for the Karachi tide gauge data, showed less dependency of the extreme sea levels on the non-tidal residuals. By applying the Merrifield criteria of mean annual maximum water level ratio, it was found that the Karachi coast was tidally dominated and the non-tidal residual contribution was just 10%. The examination of the highest water level event (13 June 2014) during the study period, further favored the tidal dominance as compared to the non-tidal component along the Karachi coast.
RESUMEN
Multi-walled carbon nanotubes (CNTs) functionalized with a deep eutectic solvent (DES) were utilized to remove mercury ions from water. An artificial neural network (ANN) technique was used for modelling the functionalized CNTs adsorption capacity. The amount of adsorbent dosage, contact time, mercury ions concentration and pH were varied, and the effect of parameters on the functionalized CNT adsorption capacity is observed. The (NARX) network, (FFNN) network and layer recurrent (LR) neural network were used. The model performance was compared using different indicators, including the root mean square error (RMSE), relative root mean square error (RRMSE), mean absolute percentage error (MAPE), mean square error (MSE), correlation coefficient (R2) and relative error (RE). Three kinetic models were applied to the experimental and predicted data; the pseudo second-order model was the best at describing the data. The maximum RE, R2 and MSE were 9.79%, 0.9701 and 1.15 × 10-3, respectively, for the NARX model; 15.02%, 0.9304 and 2.2 × 10-3 for the LR model; and 16.4%, 0.9313 and 2.27 × 10-3 for the FFNN model. The NARX model accurately predicted the adsorption capacity with better performance than the FFNN and LR models.
Asunto(s)
Mercurio/química , Nanotubos de Carbono/química , Redes Neurales de la Computación , Purificación del Agua/métodos , Adsorción , Solventes/químicaRESUMEN
Examining driving behaviour is crucial for traffic operations because of its influence on driver safety and the potential for increased risk of accidents, injuries, and fatalities. Approximately 95% of severe traffic collisions can be attributed to human error. With the progress in artificial intelligence in recent decades, notable advancements have been achieved in computer capabilities, communication systems and data collection technology. This increase has significantly influenced our capacity to replicate driver behaviour and comprehend underlying driving mechanisms in diverse situations. Traffic microsimulation facilitates an understanding of traffic performance inside a given road network. Among the microsimulation software packages, Verkehr In Städten - SIMulationsmodell (VISSIM) has garnered significant attention owing to its notable ability to accurately replicate traffic circumstances with high dependability in real-world scenarios. Given the diverse applicability of VISSIM-based schemes, this review systematically examines the applications of the VISSIM-based driving-behaviour models within different research contexts, revealing their utility. This review is designed to provide guidance for researchers in selecting the most suitable methodological approach tailored to their specific research objectives and constraints when utilising VISSIM. Five important aspects, including calibration, driving behaviour, incident, and heterogeneous traffic simulation, as well as utilisation of artificial intelligence with VISSIM, are assessed, which could yield substantial advantages in advancing more precise and authentic driving-behaviour modelling in VISSIM.
RESUMEN
A modeling framework utilizing the coactive neuro-fuzzy inference system (CANFIS) has been developed for multi-lead time groundwater level (GWL) forecasting in four different wells located in Texas and Florida, USA. Various model input combinations, including GWL, precipitation, temperature, and surface water level variables, have been derived based on proposed correlation analysis using singular spectrum analysis (SSA) remainders. The models have been trained on data subsets of varying lengths to identify the optimal training data duration. Additionally, we have introduced the bagging ensemble learning method to enhance the performance of the CANFIS model. As part of a comprehensive model evaluation process, the best-performing CANFIS model for each forecasting scenario has undergone uncertainty analysis using bootstrap sampling. Our results reveal that the CANFIS model performs satisfactorily for daily forecasting but leaves room for improvement in monthly forecasting, particularly for two-month and three-month ahead forecasts. Moreover, we have identified several optimal input combinations, highlighting the significance of the temperature variable in monthly forecasting. Furthermore, our findings indicate that additional training data does not necessarily lead to improved performance. The ensemble CANFIS model has demonstrated significant performance enhancement, particularly for monthly forecasting. Finally, the CANFIS model uncertainty analysis has shown satisfactory results for daily forecasting scenarios, while monthly forecasting models exhibit higher uncertainties, particularly during periods with distinctly different GWL fluctuation patterns.
RESUMEN
Accurate river streamflow prediction is pivotal for effective resource planning and flood risk management. Traditional river streamflow forecasting models encounter challenges such as nonlinearity, stochastic behavior, and convergence reliability. To overcome these, we introduce novel hybrid models that combine extreme learning machines (ELM) with cutting-edge mathematical inspired metaheuristic optimization algorithms, including Pareto-like sequential sampling (PSS), weighted mean of vectors (INFO), and the Runge-Kutta optimizer (RUN). Our comparative assessment includes 20 hybrid models across eight metaheuristic categories, using streamflow data from the Aswan High Dam on the Nile River. Our findings highlight the superior performance of mathematically based models, which demonstrate enhanced predictive accuracy, robust convergence, and sustained stability. Specifically, the PSS-ELM model achieves superior performance with a root mean square error of 2.0667, a Pearson's correlation index (R) of 0.9374, and a Nash-Sutcliffe efficiency (NSE) of 0.8642. Additionally, INFO-ELM and RUN-ELM models exhibit robust convergence with mean absolute percentage errors of 15.21% and 15.28% respectively, a mean absolute errors of 1.2145 and 1.2105, and high Kling-Gupta efficiencies values of 0.9113 and 0.9124, respectively. These findings suggest that the adoption of our proposed models significantly enhances water management strategies and reduces any risks.
RESUMEN
This work reports silver nanoparticles (AgNPs) supported on biopolymer carboxymethyl cellulose beads (Ag-CMC) serves as an efficient catalyst in the reduction process of p-nitrophenol (p-NP) and methyl orange (MO). For Ag-CMC synthesis, first CMC beads were prepared by crosslinking the CMC solution in aluminium nitrate solution and then the CMC beads were introduced into AgNO3 solution to adsorb Ag ions. Field emission scanning electron microscopy (FE-SEM) analysis suggests the uniform distribution of Ag nanoparticles on the CMC beads. The X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD) analysis revealed the metallic and fcc planes of AgNPs, respectively, in the Ag-CMC catalyst. The Ag-CMC catalyst exhibits remarkable reduction activity for the p-NP and MO dyes with the highest rate constant (kapp) of a chemical reaction is 0.519 and 0.697 min-1, respectively. Comparative reduction studies of Ag-CMC with CMC, Fe-CMC and Co-CMC disclosed that Ag-CMC containing AgNPs is an important factore in reducing the organic pollutants like p-NP and MO dyes. During the recyclability tests, the Ag-CMC also maintained high reduction activity, which suggests that CMC protects the AgNPs from leaching during dye reduction reactions.
Asunto(s)
Nanopartículas del Metal , Plata , Plata/química , Nanopartículas del Metal/química , Carboximetilcelulosa de Sodio , Biopolímeros , Colorantes/químicaRESUMEN
To maintain human health and purity of drinking water, it is crucial to eliminate harmful chemicals such as nitrophenols and azo dyes, considering their natural presence in the surroundings. In this particular research study, the application of machine learning techniques was employed in order to make an estimation of the performance of reduction catalysis in the context of ecologically detrimental nitrophenols and azo dyes contaminants. The catalyst utilized in the experiment was Ag@CMC, which proved to be highly effective in eliminating various contaminants found in water, like 4-nitrophenol (4-NP). The experiments were carefully conducted at various time intervals, and the machine learning procedures used in this study were all employed to forecast catalytic performance. The evaluation of the performance of such algorithms were done by means of Mean Absolute Error. The noteworthy findings of this research indicated that the ADAM and LSTM algorithm exhibited the most favourable performance in the case of toxic compounds i.e. 4-NP. Moreover, the Ag@CMC catalyst demonstrated an impressive reduction efficiency of 98 % against nitrophenol in just 8 min. Thus, based on these compelling results, it can be concluded that Ag@CMC works as a highly effective catalyst for practical applications in real-world scenarios.
RESUMEN
[This corrects the article DOI: 10.1016/j.heliyon.2023.e15740.].
RESUMEN
Sick Building Syndrome (SBS) is an illness among workers linked to time spent in a building. This study aimed to investigate the Indoor Air Quality (IAQ) and symptoms of Sick Building Syndrome (SBS) among administrative office workers. The IAQ parameters consist of ventilation performance indicators, and physical and chemical parameters were measured using specified instruments for three days during weekdays. The SBS symptoms were assessed by a questionnaire adopted from the Industry Code of Practice of Indoor Air Quality (ICOP-IAQ) 2010 among 19 employees from the office in East Coast Malaysia. Relationship between past symptoms and present symptoms which are draught (past symptoms) with feeling heavy headed (present symptoms) (r = 0.559, p < 0.05), room temperature too high (past symptoms) was highly correlated with feeling heavy headed (present symptoms) (r = 0.598, p < 0.01) and cough (present symptoms) (r = 0.596, p < 0.01). Room temperature (past symptoms) has a positive medium relationship with cough (present symptoms) (r = 0.477, p < 0.05) and scaling itching scalp or ears (present symptoms) has a relationship between stuffy bad air (r = 0.475, p < 0.05) and dry air (r = 0.536, p < 0.05). There was a significant association between RH with drowsiness (χ2 = 7.090, p = 0.049) and dizziness (χ2 = 7.090, p = 0.049). The association was found between temperature and SBS symptoms between temperature with headache (χ2 = 7.574, p = 0.051), feeling heavy-headed (χ2 = 8.090, p = 0.046), and skin rash itchiness (χ2 = 7.451, p = 0.044). Air movement also showed a positive association with symptoms of feeling heavy-headed (x2 = 8.726, p = 0.021). PM10 has positive significance with SBSS which are feeling heavy-headed (χ2 = 7.980, p = 0.023), and eyer's irritation (χ2 = 7.419, p = 0.038). The conclusion of this study showed that there were positive significant between temperature and relative humidity toward SBSS.
RESUMEN
Groundwater, the world's most abundant source of freshwater, is rapidly depleting in many regions due to a variety of factors. Accurate forecasting of groundwater level (GWL) is essential for effective management of this vital resource, but it remains a complex and challenging task. In recent years, there has been a notable increase in the use of machine learning (ML) techniques to model GWL, with many studies reporting exceptional results. In this paper, we present a comprehensive review of 142 relevant articles indexed by the Web of Science from 2017 to 2023, focusing on key ML models, including artificial neural networks (ANN), adaptive neuro-fuzzy inference systems (ANFIS), support vector regression (SVR), evolutionary computing (EC), deep learning (DL), ensemble learning (EN), and hybrid-modeling (HM). We also discussed key modeling concepts such as dataset size, data splitting, input variable selection, forecasting time-step, performance metrics (PM), study zones, and aquifers, highlighting best practices for optimal GWL forecasting with ML. This review provides valuable insights and recommendations for researchers and water management agencies working in the field of groundwater management and hydrology.
Asunto(s)
Monitoreo del Ambiente , Agua Subterránea , Monitoreo del Ambiente/métodos , Predicción , Aprendizaje Automático , Redes Neurales de la Computación , Alcanos/químicaRESUMEN
To ease water scarcity, dynamic programming, stochastic dynamic programming, and heuristic algorithms have been applied to solve problem matters related to water resources. Development, operation, and management are vital in a reservoir operating policy, especially when the reservoir serves a complex objective. In this study, an attempt via metaheuristic algorithms, namely the Harris Hawks Optimisation (HHO) Algorithm and the Opposite Based Learning of HHO (OBL-HHO) are made to minimise the water deficit as well as mitigate floods at downstream of the Klang Gate Dam (KGD). Due to trade-offs between water supply and flood management, the HHO and OBL-HHO models have configurable thresholds to optimise the KGD reservoir operation. To determine the efficacy of the HHO and OBL-HHO in reservoir optimisation, reliability, vulnerability, and resilience are risk measures evaluated. If inflow categories are omitted, the OBL-HHO meets 71.49% of demand compared to 54.83% for the standalone HHO. The HHO proved superior to OBL-HHO in satisfying demand during medium inflows, achieving 38.60% compared to 20.61%, even though the HHO may have experienced water loss at the end of the storage level. The HHO is still a promising method, as proven by its reliability and resilience indices compared to other published heuristic algorithms: at 62.50% and 1.56, respectively. The Artificial Bee Colony (ABC) outcomes satisfied demand at 61.36%, 59.47% with the Particle Swarm Optimisation (PSO), 55.68% with the real-coded Genetic Algorithm (GA), and 23.5 percent with the binary GA. For resilience, the ABC scored 0.16, PSO scored 0.15, and real coded GA scored 0.14 whilst the binary-GA has the worst failure recovery algorithm with 0.09.
RESUMEN
Cities are growing geographically in response to the enormous increase in urban population; consequently, comprehending growth and environmental changes is critical for long-term planning. Urbanization transforms naturally permeable surfaces into impermeable surfaces, causing an increase in urban land surface temperature, leading to the phenomenon known as urban heat islands. The urban heat islands are noticeable across Malaysia's rural communities and villages, particularly in Kuala Lumpur. These effects must be addressed to slow, if not halt, climate change and meet the Paris Agreement's 2030 goal. The study posits an application of thermal remote sensing utilizing a space-borne satellite-based technique to demonstrate urban evolution for urban heat island analysis and its relationship to land surface temperature. The urban heat island (UHI) was analyzed by converting infrared radiation into visible thermal images utilizing thermal imaging from remote sensing techniques. The heat island is validated by reference to the characteristics of the normalized difference vegetation index (NDVI), which define the land surface temperature (LST) of distinct locations. Based on the digital information from the satellite, the highest temperature difference between urban and rural regions for a few chosen cities in 2013 varied from 10.8 to 25.5 °C, while in 2021, it ranged from 16.1 to 26.73 °C, highlighting crucial temperature changes. The results from ANOVA test has substantially strengthened the credibility of the significant temperature changes. Some notable reveals are as follows: The Sungai Batu area, due to its rapid development and industry growth, was more vulnerable to elevated urban heat due to reduced vegetation cover; therefore, higher relative vulnerability. Contrary, the Bukit Ketumbar area, which region lies in the woodland region, experienced the lowest, with urban heat islands reading from 2013 at -0.3044 and 0.0154 in 2021. It shows that despite having urban heat islands increase two-fold from 2013 to 2021, increasing the amount of vegetation coverage is a simple and effective way of reducing the urban heat island effect, as evidenced by the low urban heat islands in the Bukit Ketumbar woodland region. The study findings are critical for advising municipal officials and urban planners to decrease urban heat islands by investing in open green spaces.
RESUMEN
The hydropower Plant in Terengganu is one of the major hydroelectric dams currently operated in Malaysia. For better operating and scheduling, accurate modelling of natural inflow is vital for a hydroelectric dam. The rainfall-runoff model is among the most reliable models in predicting the inflow based on the rainfall events. Such a model's reliability depends entirely on the reliability and consistency of the rainfall events assessed. However, due to the hydropower plant's remote location, the cost associated with maintaining the installed rainfall stations became a burden. Therefore, the study aims to create a continuous set of rainfall data before, during, and after the construction of a hydropower plant and simulate a rainfall-runoff model for the area. It also examines the reliability of alternative methods by combining rainfall data from two sources: the general circulation model and tropical rainfall measuring mission. Rainfall data from ground stations and generated data using inverse distance weighted method will be compared. The statistical downscaling model will obtain regional rainfall from the general circulation model. The data will be divided into three stages to evaluate the accuracy of the models in capturing inflow changes. The results revealed that rainfall data from TRMM is more correlated to ground station data with R2 = 0.606, while SDSM data has R2 = 0.592. The proposed inflow model based on GCM-TRMM data showed higher precision compared to the model using ground station data. The proposed model consistently predicted inflow during three stages with R2 values ranging from 0.75 to 0.93.
RESUMEN
Accurate water level prediction for both lake and river is essential for flood warning and freshwater resource management. In this study, three machine learning algorithms: multi-layer perceptron neural network (MLP-NN), long short-term memory neural network (LSTM) and extreme gradient boosting XGBoost were applied to develop water level forecasting models in Muda River, Malaysia. The models were developed using limited amount of daily water level and meteorological data from 2016 to 2018. Different input scenarios were tested to investigate the performance of the models. The results of the evaluation showed that the MLP model outperformed both the LSTM and the XGBoost models in predicting water levels, with an overall accuracy score of 0.871 compared to 0.865 for LSTM and 0.831 for XGBoost. No noticeable improvement has been achieved after incorporating meteorological data into the models. Even though the lowest reported performance was reported by the XGBoost, it is the faster of the three algorithms due to its advanced parallel processing capabilities and distributed computing architecture. In terms of different time horizons, the LSTM model was found to be more accurate than the MLP and XGBoost model when predicting 7 days ahead, demonstrating its superiority in capturing long-term dependencies. Therefore, it can be concluded that each ML model has its own merits and weaknesses, and the performance of different ML models differs on each case because these models depends largely on the quantity and quality of data available for the model training.
RESUMEN
Due to excessive streamflow (SF), Peninsular Malaysia has historically experienced floods and droughts. Forecasting streamflow to mitigate municipal and environmental damage is therefore crucial. Streamflow prediction has been extensively demonstrated in the literature to estimate the continuous values of streamflow level. Prediction of continuous values of streamflow is not necessary in several applications and at the same time it is very challenging task because of uncertainty. A streamflow category prediction is more advantageous for addressing the uncertainty in numerical point forecasting, considering that its predictions are linked to a propensity to belong to the pre-defined classes. Here, we formulate streamflow prediction as a time series classification with discrete ranges of values, each representing a class to classify streamflow into five or ten, respectively, using machine learning approaches in various rivers in Malaysia. The findings reveal that several models, specifically LSTM, outperform others in predicting the following n-time steps of streamflow because LSTM is able to learn the mapping between streamflow time series of 2 or 3 days ahead more than support vector machine (SVM) and gradient boosting (GB). LSTM produces higher F1 score in various rivers (by 5% in Johor, 2% in Kelantan and Melaka and Selangor, 4% in Perlis) in 2 days ahead scenario. Furthermore, the ensemble stacking of the SVM and GB achieves high performance in terms of F1 score and quadratic weighted kappa. Ensemble stacking gives 3% higher F1 score in Perak river compared to SVM and gradient boosting.