scholarly journals Daily river flow forecasting using wavelet ANN hybrid models

2010 ◽  
Vol 13 (1) ◽  
pp. 49-63 ◽  
Author(s):  
Niranjan Pramanik ◽  
Rabindra K. Panda ◽  
Adarsh Singh
Keyword(s):  
Neural Network ◽  
Stream Flow ◽  
Network Models ◽  
Hybrid Models ◽  
Wavelet Coefficients ◽  
Flow Data ◽  
Neural Network Models ◽  
Time Step ◽  
Frequency Bands ◽  
Ann Models

Advance time step stream flow forecasting is of paramount importance in controlling flood damage. During the past few decades, artificial neural network (ANN) techniques have been used extensively in stream flow forecasting and have proven to be a better technique than other forecasting methods such as multiple regression and general transfer function models. This study uses discrete wavelet transformation functions to preprocess the time series of the flow data into wavelet coefficients of different frequency bands. Effective wavelet coefficients are selected from the correlation analysis of the decomposed wavelet coefficients of all frequency bands with the observed flow data. Neural network models are proposed for 1-, 2- and 3-day flow forecasting at a site of Brahmani River, India. The effective wavelet coefficients are used as input to the neural network models. Both the wavelet and ANN techniques are employed to form a loose type of wavelet ANN hybrid model (NW). The hybrid models are trained using Levenberg–Marquart (LM) algorithm and the results are compared with simple ANN models. The results revealed that the predictabilities of NW models are significantly superior to conventional ANN models. The peak flow conditions are predicted with better accuracy using NW models than compared to ANN models.

Neural Network Models

2016 ◽  
pp. 368-395
Author(s):  
Eliano Pessa
Keyword(s):  
Neural Network ◽  
Statistical Physics ◽  
Optimization Theory ◽  
Network Models ◽  
Cooperative Interaction ◽  
New Paradigm ◽  
Neural Network Models ◽  
Ann Models ◽  
Artificial Neural Network Ann ◽  
The One

The Artificial Neural Network (ANN) models gained a wide popularity owing to a number of claimed advantages such as biological plausibility, tolerance with respect to errors or noise in the input data, learning ability allowing an adaptability to environmental constraints. Notwithstanding the fact that most of these advantages are not typical only of ANNs, engineers, psychologists and neuroscientists made an extended use of ANN models in a large number of scientific investigations. In most cases, however, these models have been introduced in order to provide optimization tools more useful than the ones commonly used by traditional Optimization Theory. Unfortunately, just the successful performance of ANN models in optimization tasks produced a widespread neglect of the true – and important – objectives pursued by the first promoters of these models. These objectives can be shortly summarized by the manifesto of connectionist psychology, stating that mental processes are nothing but macroscopic phenomena, emergent from the cooperative interaction of a large number of microscopic knowledge units. This statement – wholly in line with the goal of statistical mechanics – can be readily extended to other processes, beyond the mental ones, including social, economic, and, in general, organizational ones. Therefore this chapter has been designed in order to answer a number of related questions, such as: are the ANN models able to grant for the occurrence of this sort of emergence? How can the occurrence of this emergence be empirically detected? How can the emergence produced by ANN models be controlled? In which sense the ANN emergence could offer a new paradigm for the explanation of macroscopic phenomena? Answering these questions induces to focus the chapter on less popular ANNs, such as the recurrent ones, while neglecting more popular models, such as perceptrons, and on less used units, such as spiking neurons, rather than on McCulloch-Pitts neurons. Moreover, the chapter must mention a number of strategies of emergence detection, useful for researchers performing computer simulations of ANN behaviours. Among these strategies it is possible to quote the reduction of ANN models to continuous models, such as the neural field models or the neural mass models, the recourse to the methods of Network Theory and the employment of techniques borrowed by Statistical Physics, like the one based on the Renormalization Group. Of course, owing to space (and mathematical expertise) requirements, most mathematical details of the proposed arguments are neglected, and, to gain more information, the reader is deferred to the quoted literature.

Development of Neural Network Models for a Crude Oil Distillation Column

2012 ◽  
Author(s):  
Khairiyah Mohd. Yusof ◽  
Fakhri Karray ◽  
Peter L. Douglas
Keyword(s):  
Neural Network ◽  
Steady State ◽  
Crude Oil ◽  
Distillation Column ◽  
Network Models ◽  
Ann Model ◽  
Neural Network Models ◽  
Ann Models ◽  
Artificial Neural Network Ann ◽  
Crude Oil Distillation

This paper discusses the development of artificial neural network (ANN) models for a crude oil distillation column. Since the model is developed for real time optimisation (RTO) applications they are steady state, multivariable models. Training and testing data used to develop the models were generated from a reconciled steady-state model simulated in a process simulator. The radial basis function networks (RBFN), a type of feedforward ANN model, were able to model the crude tower very well, with the root mean square error for the prediction of each variable less than 1%. Grouping related output variables in a network model was found to give better predictions than lumping all the variables in a single model; this also allowed the overall complex, multivariable model to be simplified into smaller models that are more manageable. In addition, the RBFN models were also able to satisfactorily perform range and dimensional extrapolation, which is necessary for models that are used in RTO.

scholarly journals A Modified Kennard-Stone Algorithm for Optimal Division of Data for Developing Artificial Neural Network Models

2012 ◽  
Vol 7 (1) ◽  
Author(s):  
Agus Saptoro ◽  
Moses O. Tadé ◽  
Hari Vuthaluru
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Network Models ◽  
Data Representation ◽  
Data Splitting ◽  
Neural Network Models ◽  
Data Division ◽  
Ann Models ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Abstract This paper proposes a method, namely MDKS (Kennard-Stone algorithm based on Mahalanobis distance), to divide the data into training and testing subsets for developing artificial neural network (ANN) models. This method is a modified version of the Kennard-Stone (KS) algorithm. With this method, better data splitting, in terms of data representation and enhanced performance of developed ANN models, can be achieved. Compared with standard KS algorithm and another improved KS algorithm (data division based on joint x - y distances (SPXY) method), the proposed method has also shown a better performance. Therefore, the proposed technique can be used as an advantageous alternative to other existing methods of data splitting for developing ANN models. Care should be taken when dealing with large amount of dataset since they may increase the computational load for MDKS due to its variance-covariance matrix calculations.

Learning Nonlinear Constitutive Laws Using Neural Network Models Based on Indirectly Measurable Data

2020 ◽  
Vol 87 (8) ◽  
Author(s):  
Xin Liu ◽  
Fei Tao ◽  
Haodong Du ◽  
Wenbin Yu ◽  
Kailai Xu
Keyword(s):  
Neural Network ◽  
Damage Model ◽  
Network Models ◽  
Great Accuracy ◽  
Constitutive Laws ◽  
Constitutive Law ◽  
Ann Model ◽  
Neural Network Models ◽  
Plane Shear ◽  
Ann Models

Abstract Artificial neural network (ANN) models are used to learn the nonlinear constitutive laws based on indirectly measurable data. The real input and output of the ANN model are derived from indirect data using a mechanical system, which is composed of several subsystems including the ANN model. As the ANN model is coupled with other subsystems, the input of the ANN model needs to be determined during the training. This approach integrates measurable data, mechanics, and ANN models so that the ANN models can be trained without direct data which is usually not available from experiments. Two examples are provided as an illustration of the proposed approach. The first example uses two-dimensional (2D) finite element (FE) analysis to train an ANN model to learn the nonlinear in-plane shear constitutive law. The second example applies a continuum damage model to train an ANN model to learn the damage accumulation law. The results show that the trained ANN models achieve great accuracy based on the proposed approach.

scholarly journals Forecasting PM10 levels using ANN and MLR: A case study for Sakarya City

2018 ◽  
Vol 20 (2) ◽  
pp. 281-290 ◽  
Keyword(s):  
Neural Network ◽  
Air Pollutants ◽  
Regression Models ◽  
Network Models ◽  
Linear Regression Models ◽  
Neural Network Models ◽  
Prediction Ability ◽  
Topographic Features ◽  
Ann Models

In this study, potential of neural network to estimate daily mean PM10 concentration levels in Sakarya city, Turkey as a case study was examined to achieve improved prediction ability. The level and distribution of air pollutants in a particular region is associated with changes in meteorological conditions affecting air movements and topographic features. Thus, meteorological variables data for a two-year period for Sakarya city which is located in most industrialized and crowded part of Turkey were selected as input. Neural network models and multiple linear regression models have been statistically evaluated. The results of the study showed that ANN models were accurate enough for prediction of PM10 levels

scholarly journals Prediction of EV arrivals at Battery Swapping Station using Hybrid Neural Network models

2021 ◽  
Vol 2062 (1) ◽  
pp. 012021
Author(s):  
Manikanta Suri ◽  
Neha Raj ◽  
K Sireesha
Keyword(s):  
Neural Network ◽  
Performance Metrics ◽  
Pearson Correlation ◽  
Back Propagation ◽  
Network Models ◽  
Absolute Error ◽  
Hybrid Models ◽  
Time Data ◽  
Neural Network Models ◽  
Local Optima

Abstract There is an enormous increase in demand for Electric Vehicles (EV) in the present era, as they are environment-friendly when compared to conventional vehicles. Battery Swapping Stations (BSS) are gaining a lot of attention from the EV sector as it is like the gasoline stations. Forecasting of EV arrivals at BSS helps in optimally scheduling the depleted batteries to different charging piles without affecting the State of Health of the battery. Back Propagation Neural Network (BPNN) is widely used in the prediction of real-time data. Training of BPNN using metaheuristic algorithms such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) helps to overcome the local optima problem in BPNN. Thus, in the present work forecasting on the EV arrivals is carried out using GA-BPNN and PSO-BPNN hybrid models. Finally, a comparative study is carried out among BPNN, GA-BPNN, and PSO-BPNN models using the performance metrics such as Mean Square Error (MSE), Mean Absolute Error (MAE) and Pearson Correlation Coefficient (PCC). From the results, it was obtained that GA-BPNN model is preferred in forecasting the EV arrivals at BSS as the model has less overfitting. The hybrid models have been simulated in MATLAB/Simulink software.

Artificial Neural Network Modelling of Heat Transfer to Canned Particulate Fluids under Axial Rotation Processing

2010 ◽  
Vol 6 (3) ◽  
Author(s):  
Mritunjay Dwivedi ◽  
Hosahalli S. Ramaswamy
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Artificial Neural Network ◽  
Cross Validation ◽  
Network Models ◽  
Neural Network Models ◽  
Axial Mode ◽  
Ann Models ◽  
Hidden Layer ◽  
Dimensionless Correlations

Artificial neural network models were developed for the overall heat transfer coefficient (U) and the fluid to particle heat transfer coefficient hfp in canned Newtonian fluids with and without particles, and the model performances were compared with the dimensionless correlations for both free and fixed axial modes of agitation. Part of the experimental data were used for training and testing, and a portion was used for cross validation. The average errors (RMS), associated with predicted hfp and U values in fixed and free axial mode were a function of the ANN variables: number of hidden layers, number of neurons in each hidden layer, learning rule, transfer function and number of learning runs. RMS values not significantly different with number of hidden layers between one and three, and the associated RMS was minimal with a high R2 value with one hidden layer and 8 neurons. The combination of the Delta-rule and TanH transfer function also gave the lowest RMS and the highest R2. The highest R2 was achieved for the data set with 85% used for training and testing and 15 % for the cross validation in both modes of rotation, and therefore this combination was used for the development of neural network models. Mean relative errors (MRE) for ANN models were much lower compared with MRE associated with dimensionless correlations; 75-78% lower for hfp and 66% lower for U in fixed and free axial mode with particulate in liquid. Without particulates, in comparison with dimensionless correlations, the MRE for ANN models were 37% lower in end-over-end mode and 76% lower for free axial mode. Overall, ANN models yielded much higher R2 values than dimensionless correlations. The ANN coefficient matrix is included so that the models can be implemented in a spreadsheet.

scholarly journals Predictive Ability of Improved Neural Network Models to Simulate Pollutant Dispersion

2014 ◽  
Vol 2014 ◽  
pp. 1-12 ◽  
Author(s):  
Khandaker M. A. Hossain
Keyword(s):  
Neural Network ◽  
Predictive Ability ◽  
Network Models ◽  
Pollutant Dispersion ◽  
Neural Network Models ◽  
Coastal Cities ◽  
Ann Models ◽  
Using Data ◽  
Artificial Neural Network Ann ◽  
Urban Atmospheres

This paper describes the ability of artificial neural network (ANN) models to simulate the pollutant dispersion characteristics in varying urban atmospheres at different regions. ANN models are developed based on twelve meteorological (including rainfall/precipitation) and six traffic parameters/variables that have significant influence on emission/pollutant dispersion. The models are trained to predict concentration of carbon monoxide and particulate matters in urban atmospheres using field meteorological and traffic data. Training, validation, and testing of ANN models are conducted using data from the Dhaka city of Bangladesh. The models are used to simulate concentration of pollutants as well as the effect of rainfall on emission dispersion throughout the year and inversion condition during the night. The predicting ability and robustness of the models are then determined by using data of the coastal cities of Chittagong and Dhaka. ANN models based on both meteorological and traffic variables exhibit the best performance and are capable of resolving patterns of pollutant dispersion to the atmosphere for different cities.

scholarly journals Improving streamflow forecast using optimal rain gauge network-based input to artificial neural network models

2017 ◽  
Vol 49 (5) ◽  
pp. 1559-1577 ◽  
Author(s):  
Sajal Kumar Adhikary ◽  
Nitin Muttil ◽  
Abdullah Gokhan Yilmaz
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Network Models ◽  
Rain Gauge ◽  
Streamflow Forecasting ◽  
Neural Network Models ◽  
Ann Models ◽  
Artificial Neural ◽  
Input Variables ◽  
Rain Gauge Network

Abstract Accurate streamflow forecasting is of great importance for the effective management of water resources systems. In this study, an improved streamflow forecasting approach using the optimal rain gauge network-based input to artificial neural network (ANN) models is proposed and demonstrated through a case study (the Middle Yarra River catchment in Victoria, Australia). First, the optimal rain gauge network is established based on the current rain gauge network in the catchment. Rainfall data from the optimal and current rain gauge networks together with streamflow observations are used as the input to train the ANN. Then, the best subset of significant input variables relating to streamflow at the catchment outlet is identified by the trained ANN. Finally, one-day-ahead streamflow forecasting is carried out using ANN models formulated based on the selected input variables for each rain gauge network. The results indicate that the optimal rain gauge network-based input to ANN models gives the best streamflow forecasting results for the training, validation and testing phases in terms of various performance evaluation measures. Overall, the study concludes that the proposed approach is highly effective to achieve the enhanced streamflow forecasting and could be a viable option for streamflow forecasting in other catchments.

Sign in / Sign up

Export Citation Format

Share Document