Bayesian and Sensitivity Analysis Approaches to Modelling the Direct Solar Irradiance

Solar Energy ◽  
2005 ◽  
Author(s):  
Philippe Lauret ◽  
Mathieu David ◽  
Eric Fock ◽  
Laetitia Adelard
Keyword(s):  
Neural Network ◽  
Sensitivity Analysis ◽  
Solar Irradiance ◽  
Second Step ◽  
Practical Implementation ◽  
Bayesian Techniques ◽  
Systematic Methodology ◽  
Important Input ◽  
Input Variables ◽  
Selection Of

In this paper, emphasis is put on the design of a neural network to model the direct solar irradiance. Since unfortunately a neural network (NN) is not a statistician in-a-box, building a NN for a particular problem is a non trivial task. As a consequence, we argue that in order to properly model the direct solar irradiance, a systematic methodology must be employed. For this purpose, we propose a two-step approach to building the NN model. The first step deals with a probabilistic interpretation of the NN learning by using Bayesian techniques. The Bayesian approach to modelling offers significant advantages over the classical NN learning process. Among others, one can cite a) automatic complexity control of the NN using all the available data b) selection of the most important input variables. The second step consists in using a new sensitivity analysis-based pruning method in order to infer the optimal NN structure. We show that the combination of the two approaches makes the practical implementation of the Bayesian techniques more reliable.

Bayesian and Sensitivity Analysis Approaches to Modeling the Direct Solar Irradiance

2006 ◽  
Vol 128 (3) ◽  
pp. 394-405 ◽  
Author(s):  
Philippe Lauret ◽  
Mathieu David ◽  
Eric Fock ◽  
Alain Bastide ◽  
Carine Riviere
Keyword(s):  
Neural Network ◽  
Sensitivity Analysis ◽  
Solar Irradiance ◽  
Second Step ◽  
Practical Implementation ◽  
Bayesian Techniques ◽  
Systematic Methodology ◽  
Important Input ◽  
Input Variables ◽  
Selection Of

In this paper, emphasis is put on the design of a neural network (NN) to model the direct solar irradiance. Since, unfortunately, a neural network is not a statistician-in-a-box, building a NN for a particular problem is a nontrivial task. As a consequence, we argue that in order to properly model the direct solar irradiance, a systematic methodology must be employed. For this purpose, we propose a two-step approach to building the NN model. The first step deals with a probabilistic interpretation of the NN learning by using Bayesian techniques. The Bayesian approach to modeling offers significant advantages over the classical NN learning process. Among others, one can cite (i) automatic complexity control of the NN using all the available data and (ii) selection of the most important input variables. The second step consists of using a new sensitivity analysis-based pruning method in order to infer the optimal NN structure. We show that the combination of the two approaches makes the practical implementation of the Bayesian techniques more reliable.

scholarly journals Prediction Model of Iron Ore Pellet Ambient Strength and Sensitivity Analysis on the Influence Factors

Metals ◽  
2018 ◽  
Vol 8 (8) ◽  
pp. 593 ◽  
Author(s):  
Qiangjian Gao ◽  
Yingyi Zhang ◽  
Xin Jiang ◽  
Haiyan Zheng ◽  
Fengman Shen
Keyword(s):  
Neural Network ◽  
Sensitivity Analysis ◽  
Prediction Model ◽  
Bp Neural Network ◽  
Back Propagation ◽  
Influence Factors ◽  
Principal Component ◽  
Bp Network ◽  
Input Variables ◽  
Artificial Neural Network Ann

The Ambient Compressive Strength (CS) of pellets, influenced by several factors, is regarded as a criterion to assess pellets during metallurgical processes. A prediction model based on Artificial Neural Network (ANN) was proposed in order to provide a reliable and economic control strategy for CS in pellet production and to forecast and control pellet CS. The dimensionality of 19 influence factors of CS was considered and reduced by Principal Component Analysis (PCA). The PCA variables were then used as the input variables for the Back Propagation (BP) neural network, which was upgraded by Genetic Algorithm (GA), with CS as the output variable. After training and testing with production data, the PCA-GA-BP neural network was established. Additionally, the sensitivity analysis of input variables was calculated to obtain a detailed influence on pellet CS. It has been found that prediction accuracy of the PCA-GA-BP network mentioned here is 96.4%, indicating that the ANN network is effective to predict CS in the pelletizing process.

scholarly journals Why Biopolymer Packaging Materials are Better

2019 ◽  
Vol 23 (2) ◽  
pp. 366-384 ◽  
Author(s):  
Nadeeshika Silva ◽  
Dagnija Blumberga
Keyword(s):  
Sensitivity Analysis ◽  
Polymeric Materials ◽  
Packaging Material ◽  
Second Step ◽  
Packaging Materials ◽  
Mechanical And Physical Properties ◽  
Input Variables ◽  
Two Phases ◽  
The Ideal

Abstract The upcoming packaging material trend is bio-polymeric materials since it has shown tremendous potential in practical scenarios. Even though there have been experiments performed regarding material developments, there is still no confirmation about how uncertain the developments will be. A few statistical approaches were carried out in this work to identify the role of biopolymers as a packaging material based on their thermo-mechanical and physical properties and potential compared to other packaging materials. To determine the potential of biopolymer, it is compared with other package materials currently in demand. There are three main steps in the research. The first stage is an analysis of selected different packaging materials based on Multi-criteria decision making (MCDM) technique. The material properties are analysed through the criteria of TOPSIS analysis. The ideal and negative ideal alternatives have been identified. Biopolymers have an outcome as the final best alternative among others. To confirm the TOPSIS results and its uncertainties, a sensitivity analysis is performed. This sensitivity analysis was performed in two phases. The first step is a regression analysis of the weighted parameters and input variables of the TOPSIS scheme. The second step is the variation of weights in a unitary variation ratio to identify the order of the TOPSIS results at each variation. Finally, all the results have concluded that the research intention has been fulfilled by performing TOPSIS and the sensitivity analysis has also confirmed this decision.

scholarly journals Classification of Bidikmisi Scholarship Acceptance using Neural Network Based on Hybrid Method of Genetic Algorithm

2021 ◽  
Vol 5 (2) ◽  
pp. 396-404
Author(s):  
N Cahyani ◽  
Sinta Septi Pangastuti ◽  
K Fithriasari ◽  
Irhamah Irhamah ◽  
N Iriawan
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Optimal Number ◽  
Connection Weight ◽  
The Neural Network ◽  
Network Method ◽  
Input Variables ◽  
Human Brains ◽  
Selection Of

A Neural network is a series of algorithms that endeavours to recognize underlying relationships in a set of data through processes that mimic the way human brains operate. In the case of classification, this method can provide a fit model through various factors, such as the variety of the optimal number of hidden nodes, the variety of relevant input variables, and the selection of optimal connection weights. One popular method to achieve the optimal selection of connection weights is using a Genetic Algorithm (GA), the basic concept is to iterate over Darwin's evolution. This research presents the Neural Network method with the Backpropagation Neural Network (BPNN) and the combined method of BPNN with GA, where GA is used to initialize and optimize the connection weight of BPNN. Based on accuracy value, the BPNN method combined with GA provides better classification, which is 90.51%, in the case of Bidikmisi Scholarship classification in East Java.

scholarly journals SELECTION OF INPUT VARIABLES OF NONLINEAR AUTOREGRESSIVE NEURAL NETWORK MODEL FOR TIME SERIES DATA FORECASTING

2020 ◽  
Vol 13 (2) ◽  
pp. 116-124
Author(s):  
Hermansah Hermansah ◽  
Dedi Rosadi ◽  
Abdurakhman Abdurakhman ◽  
Herni Utami
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Network Model ◽  
Neural Network Model ◽  
Time Series Data ◽  
Series Data ◽  
Input Variables ◽  
Nonlinear Autoregressive ◽  
Selection Of ◽  
Autoregressive Neural Network

NARNN is a type of ANN model consisting of a limited number of parameters and widely used for various applications. This study aims to determine the appropriate NARNN model, for the selection of input variables of nonlinear autoregressive neural network model for time series data forecasting, using the stepwise method. Furthermore, the study determines the optimal number of neurons in the hidden layer, using a trial and error method for some architecture. The NARNN model is combined in three parts, namely the learning method, the activation function, and the ensemble operator, to get the best single model. Its application in this study was conducted on real data, such as the interest rate of Bank Indonesia. The comparison results of MASE, RMSE, and MAPE values with ARIMA and Exponential Smoothing models shows that the NARNN is the best model used to effectively improve forecasting accuracy.

Sign in / Sign up

Export Citation Format

Share Document