Data-driven flow forecasting models, such as Artificial Neural Networks (ANNs), are increasingly used for operational flood warning systems. In this research, we systematically evaluate different machine learning techniques (random forest and decision tree) and compare them with classical methods of the NAM rainfall run-off model for the Vésubie River, Nice, France. The modeled network is trained and tested using discharge, precipitation, temperature, and evapotranspiration data for about four years (2011–2014). A comparative investigation is executed to assess the performance of the model by using Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and a correlation coefficient (R). According to the result, Feed Forward Neural Network (FFNN) (a type of ANN) models are less efficient than NAM models. The precision parameters correlation coefficient of ANN is 0.58 and for the NAM model is 0.76 for the validation dataset. In all machine learning models, the decision tree which performed best had a correlation coefficient of 0.99. ANN validation data prediction is good compared to the training, which is the opposite in the NAM model. ANN can be improved by fitting more input variables in the training dataset for a long period.
The environmental issues we are currently facing require long-term prospective efforts for sustainable growth. Renewable energy sources seem to be one of the most practical and efficient alternatives in this regard. Understanding a nation's pattern of energy use and renewable energy production is crucial for developing strategic plans. No previous study has been performed to explore the dynamics of power consumption with the change in renewable energy production on a country-wide scale. In contrast, a number of deep learning algorithms demonstrated acceptable performance while handling sequential data in the era of data-driven predictions. In this study, we developed a scheme to investigate and predict total power consumption and renewable energy production time series for eleven years of data using a Recurrent Neural Network (RNN). The dynamics of the interaction between the total annual power consumption and renewable energy production are investigated through extensive Exploratory Data Analysis (EDA) and a feature engineering framework. The performance of the model is found satisfactory through the comparison of the predicted data with the observed data, visualization of the distribution of the errors and Root Mean Squared Error (RMSE) value of 0.084. Higher performance is achieved through the increase in the number of epochs and hyperparameter tuning. The proposed framework can be used and transferred to investigate the trend of renewable energy production and power consumption and predict the future scenarios for different communities. Incorporation of the cloud-based platform into the proposed pipeline may lead to real-time forecasting.
River flow prediction is a pivotal task in the field of water resource management during the era of rapid climate change. The highly dynamic and evolving nature of the climatic variables, e.g., precipitation, has a significant impact on the temporal distribution of the river discharge in recent days, making the discharge forecasting even more complicated for diversified water-related issues, e.g., flood prediction and irrigation planning. In order to predict the discharge, various physics-based numerical models are used using numerous hydrologic parameters. Extensive lab-based investigation and calibration are required to reduce the uncertainty involved in those parameters. However, in the age of data-driven predictions, several deep learning algorithms showed satisfactory performance in dealing with sequential data. In this research, Long Short-term Memory (LSTM) neural network regression model is trained using over 80 years of daily data to forecast the discharge time series up to seven days ahead of time. The performance of the model is found satisfactory through the comparison of the predicted data with the observed data, visualization of the distribution of the errors, and R2 value of 0.93 with one day lead time. Higher performance is achieved through the increase in the number of epochs and hyperparameter tuning. This model can be transferred to other locations with proper feature engineering and optimization to perform univariate predictive analysis and potentially be used to perform real-time river discharge prediction.
Distribution of the water flow path and residence time (HRT) in the hyporheic zone is a pivotal aspect in anatomizing the transport of environmental contaminants and the metabolic rates at the groundwater and surface water interface in fluvial habitats. Due to high variability in material distribution and composition in streambed and subsurface media, a pragmatic model setup in the laboratory is strenuous. Moreover, investigation of an individual streamline cannot be efficiently executed in laboratory experiments. However, an automated generation of water flow paths, i.e., streamlines in the hyporheic zone with a range of different streambed configurations could lead to a greater insight into the behavior of hyporheic water flow. An automated approach to quantifying the water flow in hyporheic zone is developed in this study where the surface water modeling tool, HER-RAS, and subsurface water flow modelling code, MIN3P, are coupled. A 1m long stream with constant water surface elevation of 2 cm to generate hydraulic head gradients and a saturated subsurface computational space with the dimensions of x:y:z = 1:0.1:0.1 m is considered to analyze the hyporheic exchange. Response in the hyporheic streamlines and residence time due to small-scale changes in the gravel-sand streambed were analyzed. The outcomes of the model show that the size, shape, and distribution of the gravel and sand portions have a significant influence on the hyporheic flow path and HRT. A high number and length of the hyporheic flow path are found in case of the highly elevated portion of gravel pieces. With the increase in the base width of gravel pieces, the length of hyporheic flow path and HRT decreases. In the case of increased amounts of gravel and sand portions on the streambed, both the quantity and length of the hyporheic flow path are reduced significantly.
Residence time of water flow is an important factor in subsurface media to determine the fate of environmental toxins and the metabolic rates in the ecotone between the surface stream and groundwater. Both numerical and lab-based experimentation can be used to estimate the residence time. However, due to high variability in material composition in subsurface media, a pragmatic model set up in the laboratory to trace particles is strenuous. Nevertheless, the selection and inclusion of input parameters, execution of the simulation, and generation of results as well as post-processing of the outcomes of a simulation take a considerable amount of time. To address these challenges, an automated particle tracing method is developed where the numerical model, i.e., flow and reactive transport code, MIN3P, and MATLAB code for tracing particles in saturated porous media, is used. A rectangular model domain is set up considering a fully saturated subsurface media under steady-state conditions in MIN3P. Streamlines and residence times of the particles are computed with a variety of seeding locations covering the whole model surface. Sensitivity analysis for residence time is performed over the varying spatial discretization and computational time steps. Moreover, a comparative study of the outcomes with Paraview is undertaken to validate the automated model (R2 = 0.997). The outcome of the automated process illustrates that the computed residence times are highly dependent on the accuracy of the integration method, the value of the computational time step, ∆t, spatial discretization, stopping criterion for the integration process of streamlines, location, and amount of seed points. The automated process can be highly beneficial in obtaining insights into subsurface flow dynamics with high variability in the model setup instead of laboratory-based experimentation in a computationally efficient manner.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.