Artificial Network Model Research Articles

Machine learning prediction of the optimal carrier concentration and band gap of quaternary thermoelectric materials via element feature descriptors

AbstractWhile the thermoelectric (TE) materials have attracted significant attention in recent years, the design and discovery of new TE materials with optimal carrier concentration and band gap remains a great challenge. Herein, we report the development of machine learning (ML) methods to predict TE materials with introducing physically meaningful simple descriptors. Specifically, we use the number of electrons, Pauling electronegativity and relative atomic mass as the basic physical variables and compute 242 descriptors in 64 categories to characterize the molecular information of a TE material. Multiple stepwise regression is employed to reduce the dimensionality in the developed ML models, and 5 and 4 important features for the band gap and carrier concentration is selected, respectively. The important features are used as input of a total number of 19 ML methods to select the optimal ML models for the prediction of band gap and carrier concentration, respectively. It is shown that the least square support vector machines method is the best model for the prediction of the band gap, while the back propagating artificial neutral network model exhibits the best performance in predicting the carrier concentration values. This work provides novel theoretical guidance for the rapid prediction properties of TE materials. The simple descriptors we defined can accurately predict the band gap and carrier concentration of quaternary TE materials.

Modified artificial neural network model with an explicit expression to describe flow behavior and processing maps of Ti2AlNb-based superalloy

The elevated-temperature deformation behavior of Ti2AlNb superalloy was observed by isothermal compression experiments in a wide range of temperatures (950–1200 °C) and strain rates (0.001–10 $${\mathrm{s}}^{-1}$$ ). The flow behavior is nonlinear, strongly coupled, and multivariable. The constitutive models, namely the double multivariate nonlinear regression model, artificial neural network model, and modified artificial neural network model with an explicit expression, were applied to describe the Ti2AlNb superalloy plastic deformation behavior. The comparative predictability of those constitutive models was further evaluated by considering the correlation coefficient and average absolute relative error. The comparative results show that the modified artificial network model can describe the flow stress of Ti2AlNb superalloy more accurately than the other developed constitutive models. The explicit expression obtained from the modified artificial neural network model can be directly used for finite element simulation. The modified artificial neural network model solves the problems that the double multivariate nonlinear regression model cannot describe the nonlinear, strongly coupled, and multivariable flow behavior of Ti2AlNb superalloy accurately, and the artificial neural network model cannot be embedded into the finite element software directly. However, the modified artificial neural network model is mainly dependent on the quantity of high-quality experimental data and characteristic variables, and the modified artificial neural network model has not physical meanings. Besides, the processing maps were applied to obtain the optimum processing parameters.

O29: PREDICTING RESPONSE TO NEOADJUVANT THERAPY IN OESOPHAGEAL ADENOCARCINOMA PRE-TREATMENT BIOPSIES

Abstract Aims We currently cannot predict which patients with locally advanced oesophageal adenocarcinoma will be amongst the 15-20% to gain a clinically important response to neoadjuvant therapy (NAT). This pilot study aimed to identify differentially expressed genes from oesophageal adenocarcinoma pre-treatment biopsies between responders and non-responders to NAT and develop methodology for predicting response. Method Response to NAT was assessed pathologically using Tumour Regression Grading (TRG). Pre-treatment formalin-fixed paraffin embedded samples were analysed with two nuclease protection assays (EdgeSeq, HTG = Oncology Biomarker Panel (OBP) and Precision Immuno-Oncology Panel (PIP)). Sequencing was performed on the NextSeq500 (Illumina). Result Whilst there was no difference in pre-treatment characteristics, responders (TRG1-2, n=26) had significantly better post-treatment pathology and overall survival than non-responders (TRG4-5, n=30). Genes up-regulated in responders were involved in regulating cell cycling, whereas genes up-regulated in non-responders were involved in cytokine signalling and the immune response. Neuronal artificial network models could predict response to NAT with overall accuracy of 73% and 68% for the OBP and PIP, respectively, which is promising considering the small sample size. As no model will be 100% accurate, we developed a model that could take patient's views into consideration with an adjustable probability threshold for classification. Conclusion This pilot study informs a biologically sound hypothesis for the basis of response to NAT and suggests prediction from pre-treatment biopsies may be possible using EdgeSeq. We now aim to validate these results in a larger study to inform a bespoke classifier of response to enable delivery of precision therapy. Take-home message In oesophageal adenocarcinoma, responders and non-responders to neoadjuvant therapy have different expression profiles. Through using EdgeSeq in larger studies, we may be able to predict which patients will respond to treatment, allowing for delivery of precision therapy.

Value Analysis of Using Urinary Microalbumin in Artificial Intelligence Medical Institutions to Detect Early Renal Damage in Diabetes.

As the scale and depth of artificial intelligence network models continue to increase, their accuracy in albumin recognition tasks has increased rapidly. However, today's small medical datasets are the main reason for the poor recognition of artificial intelligence techniques in this area. The sample size in this article is based on the data analysis and research on urine albumin detection of diabetes in the EI database. It is assumed that the observation group has at least 20 mg UAER difference from the control group, and the standard deviation of the UAER change from baseline to 12 weeks is 30 mg. Therefore, the sample size of the two groups is 77 cases. Assuming that the rate of loss to follow-up during the follow-up period is 20%, at least 92 patients are needed. The final enrollment in this study is 100 patients. Studies have shown that DR is used as an indicator to diagnose NDRD, and its OR value is as high as 28.198, indicating that non-DR can be used as an indicator to distinguish DN from NDRD. The meta-analysis found that DR has a sensitivity of 0.65 and a specificity of 0.75 in distinguishing DN from NDRD in patients with type 2 diabetes, and it is emphasized that PDR is highly specific in the diagnosis of DN. Using a meta-analysis to systematically analyze 45 studies, it was found that the sensitivity of DR to diagnose DN was 0.67, the specificity was 0.78, and the specificity of PDR to predict DN was 0.99, indicating that DR is a good indicator for predicting DN, and the team's latest research has also verified this point of view. They have established a new model for diagnosing DN. In addition to including traditional proteinuria, glycosylated hemoglobin, FR, blood pressure, and other indicators into the diagnostic model, it will also include the presence or absence of DR. The final external verification accuracy rate of this model is 0.875.

On the reliability of predictions on Covid-19 dynamics: A systematic and critical review of modelling techniques.

Since the emergence of the novel 2019 coronavirus pandemic in December 2019 (COVID-19), numerous modellers have used diverse techniques to assess the dynamics of transmission of the disease, predict its future course and determine the impact of different control measures. In this study, we conducted a global systematic literature review to summarize trends in the modelling techniques used for Covid-19 from January 1st, 2020 to November 30th, 2020. We further examined the accuracy and precision of predictions by comparing predicted and observed values for cumulative cases and deaths as well as uncertainties of these predictions. From an initial 4311 peer-reviewed articles and preprints found with our defined keywords, 242 were fully analysed. Most studies were done on Asian (78.93%) and European (59.09%) countries. Most of them used compartmental models (namely SIR and SEIR) (46.1%) and statistical models (growth models and time series) (31.8%) while few used artificial intelligence (6.7%), Bayesian approach (4.7%), Network models (2.3%) and Agent-based models (1.3%). For the number of cumulative cases, the ratio of the predicted over the observed values and the ratio of the amplitude of confidence interval (CI) or credibility interval (CrI) of predictions and the central value were on average larger than 1 indicating cases of inaccurate and imprecise predictions, and large variation across predictions. There was no clear difference among models used for these two ratios. In 75% of predictions that provided CI or CrI, observed values fall within the 95% CI or CrI of the cumulative cases predicted. Only 3.7% of the studies predicted the cumulative number of deaths. For 70% of the predictions, the ratio of predicted over observed cumulative deaths was less or close to 1. Also, the Bayesian model made predictions closer to reality than classical statistical models, although these differences are only suggestive due to the small number of predictions within our dataset (9 in total). In addition, we found a significant negative correlation (rho = - 0.56, p = 0.021) between this ratio and the length (in days) of the period covered by the modelling, suggesting that the longer the period covered by the model the likely more accurate the estimates tend to be. Our findings suggest that while predictions made by the different models are useful to understand the pandemic course and guide policy-making, some were relatively accurate and precise while other not.

High-velocity powder compaction: An experimental investigation, modelling, and optimization

Dynamic compaction of Aluminum powder using gas detonation forming technique was investigated. The experiments were carried out on four different conditions of total pre-detonation pressure. The effects of the initial powder mass and grain particle size on the green density and strength of compacted specimens were investigated. The relationships between the mentioned powder design parameters and the final features of specimens were characterized using Response Surface Methodology (RSM). Artificial Neural Network (ANN) models using the Group Method of Data Handling (GMDH) algorithm were also developed to predict the green density and green strength of compacted specimens. Furthermore, the desirability function was employed for multi-objective optimization purposes. The obtained optimal solutions were verified with three new experiments and ANN models. The obtained experimental results corresponding to the best optimal setting with the desirability of 1 are 2714 kg m3 and 21.5 MPa for the green density and green strength, respectively, which are very close to the predicted values.

ARTIFICIAL NEURAL NETWORK MODELING OF THE WATER QUALITY INDEX FOR THE EUPHRATES RIVER IN IRAQ

This study was aimed to investigate the development and evaluation of artificial intelligence techniques by using multilayer neural network. Levenberg–Marquardt back propagation (LMA) training algorithm was applied for calculating drinking water quality index (WQI) for Euphrates river (IRAQ). The transfer functions in the artificial network model were tangent sigmoid and linear for hidden and output layers, respectively. Eleven neurons presented for good prediction for results of (WQI) with a coefficient of correlation >0.97 and statistically calculated WQI values, inferring that the model predictions explain 94% of the variation in the calculated WQI scores. The WQI score of the Euphrates was 142 considered as poor. The analysis of sensitivity revealed that the total dissolved solids (TDS) is the highest effective variable with the relative importance of (26.3%), followed by electrical conductivity (EC) (23.1%), pH (17.3%), calcium (Ca) (0.149), chlorides (Cl) (11.2%), Hardness (5.7%), Temperature (1.3%), respectively. It can be concluded that the model presented in this study gives a useful alternate to WQI assessment, which use sub indices formulae.

A Diversity of Intrinsic Timescales Underlie Neural Computations.

Neural processing occurs across a range of temporal scales. To facilitate this, the brain uses fast-changing representations reflecting momentary sensory input alongside more temporally extended representations, which integrate across both short and long temporal windows. The temporal flexibility of these representations allows animals to behave adaptively. Short temporal windows facilitate adaptive responding in dynamic environments, while longer temporal windows promote the gradual integration of information across time. In the cognitive and motor domains, the brain sets overarching goals to be achieved within a long temporal window, which must be broken down into sequences of actions and precise movement control processed across much shorter temporal windows. Previous human neuroimaging studies and large-scale artificial network models have ascribed different processing timescales to different cortical regions, linking this to each region’s position in an anatomical hierarchy determined by patterns of inter-regional connectivity. However, even within cortical regions, there is variability in responses when studied with single-neuron electrophysiology. Here, we review a series of recent electrophysiology experiments that demonstrate the heterogeneity of temporal receptive fields at the level of single neurons within a cortical region. This heterogeneity appears functionally relevant for the computations that neurons perform during decision-making and working memory. We consider anatomical and biophysical mechanisms that may give rise to a heterogeneity of timescales, including recurrent connectivity, cortical layer distribution, and neurotransmitter receptor expression. Finally, we reflect on the computational relevance of each brain region possessing a heterogeneity of neuronal timescales. We argue that this architecture is of particular importance for sensory, motor, and cognitive computations.

Neural Network Classification of Ice-Crystal Images Observed by an Airborne Cloud Imaging Probe

ABSTRACT In the atmosphere, cloud particles have different shapes. The study of cloud particle shapes plays an important role in understanding cloud precipitation processes, radiative transfer, and weather modification. The image resolution and data quality of cloud probes affect the accuracy of the classification of particle shapes. To solve the occlusion of the photosensitive edge of the particle image and achieve automatic, high-precision ice-crystal classification of airborne Cloud Imaging Probe (CIP) ice-crystal images, this study uses a traditional image processing algorithm for data quality control and applies artificial intelligence algorithms to classify ice-crystal images. At present, there are mainly two types of ice-crystal classification methods, one classifies the shape of ice crystals using a pattern parameterization scheme, and the other uses an artificial intelligence network model to classify the shape. Combined with data quality control, the dataset was tested on eight models, and the TL-EfficientNet-b6 model was found to be the most accurate. Therefore, the TL-EfficientNet-b6 classifier model was used in this study, which is a newly developed convolutional neural network (CNN) based on a transfer learning method. Experimental results show that the TL-EfficientNet-b6 model can reach 100% in the single-class precision of tiny and hexagonal ice crystals, and the average precision can reach 98%. These results are more accurate than those using traditional classification methods. This method could be valuable in cloud microphysics research and weather modification.

Production Decline Prediction of Shale Gas using Hybrid Models

Hybrid models have frequently been used for shale gas production decline prediction by manipulating the unique strength of each of the known decline models. The use of a combination of models provides a more precise predicting model for forecasting time series data as compared to an individual model. In this study, the forecasting performance of decline curve hybrid models and ANN-ARIMA hybrid models are evaluated and compared with Arps’, Duong’s, the Power Law Exponential Decline, Autoregressive Integrated Moving Average (ARIMA) and Artificial Neutral Network (ANN) models, respectively. The variable used to assess the models was the respective flow rate, q(t) monitored over a period of time (T). The results have shown that the single model approach can outperform hybrid models. The average deviation of the two best models indicates a central tendency of the production data around the mean. Subsequently, the spread in the data between the actual and predicted values is found to be less. It can thus be concluded that the ARIMA and ANN models have the best forecasting accuracy for production decline in shale gas compared to the other models.

Influence of water-to-binder ratio on the optimum percentage of nano-TiO2 addition in terms of compressive strength of mortars: A laboratory and virtual experimental study based on ANN model

This research aims to study the effect of water-to-binder ratio (w/b) as a potential key factor that changes the optimum percentage of nano-TiO2 addition in terms of compressive strength of mortars. Data from previous literature were collected and analyzed. An artificial neuron network (ANN) model was trained and validated to predict the compressive strength of mortars as a function of the level of nano-TiO2, w/b, cement content, aggregate-to-binder ratio, and age. Then, a virtual experimental campaign evaluating 45,015 case scenarios, and using the ANN model, was performed to study the effect of the w/b on the optimum percentage of nano-TiO2. Results suggest that the higher the w/b and the lower the age, the higher the optimum percentage of nano-TiO2. Laboratory experimental compressive strength results are in agreement with these trends. Thermogravimetric analysis and 3D X-Ray scans show that the use of nano-TiO2 presents a double positive effect on compressive strength, reducing the porosity and increasing the hydration product content when the initial porosity of the reference cement paste is high (due to high w/b). However, when the porosity of reference samples is very low (due to low w/b), the use of nano-TiO2 may lead to a reduction of hydration products content due to the lack of space for the CH to grow. Since the initial porosity was already very low, the porosity reduction due to the use of nano-TiO2 would not be enough to overcome the negative effect on hydration. Therefore, it may lead to a decrease in the compressive strength of mortars. These results explain the dependence of the optimum percentage of nano-TiO2 on w/b and age.

Application of Artificial Neural Network for Prediction of Key Indexes of Corn Industrial Drying by Considering the Ambient Conditions

Uncontrollable ambient conditions are the main factors limiting the self-adaption control of an industrial drying system. To achieve the goal of accurate control of the drying process, the influence of the ambient conditions on the drying behavior should be taken into consideration when modeling the drying process. Present work introduced an industrial drying system with a loading capacity of 50 t, two artificial neural network prediction models with (IANN) and without (OANN) considering the ambient conditions were established using artificial neural network modeling approach. The ambient conditions on the moisture content (MC), exergy efficiency of the heat exchanger (ηex,h) and specific recovered radiant energy (Er) of the drying process were also investigated. The results showed that the ηex,h and Er increase with the increase of ambient temperature while the drying time decrease with the increase of the ambient temperature. The IANN model has a better prediction performance that that of OANN model. An optimal architecture of 9-2-12-3 artificial neuron network model was developed and the best prediction performance of the artificial neural network (ANN) model were found at a training epoch number of 30, and a momentum coefficient of 0.4, where the coefficient of determination of moisture content, exergy efficiency of heat exchanger, and the specific recovered radiant energy, respectively are 0.998, 0.992, and 0.980, indicating that the model has an excellent prediction performance and can be used in engineering practice.

New artificial network model to estimate biological activity of peat humic acids

PurposeThis article focuses on new method to estimate biological activity of peat humic acids (HAs) using artificial neural network (ANN) to process spectroscopic measurements in infrared and visible ranges. Conventional approaches generally rely on biological models and direct detection of chemical substances related to bioactivity. These methods proved to be accurate and reliable, but at the expense of speed and simplicity. Materials and methodsRecently, a conception of quantitative structure-activity relationship (QSAR) has been introduced and successfully implemented to predict effects of HAs on toxicity of polycyclic aromatic hydrocarbons. Our research stems from this conception, but employs multilayer perceptron (MLP) model to improve overall performance. The developed MLP model allowed us to estimate biological activity of the complete vertical peat cores collected from oligotrophic peat bog, located in southern taiga zone of West Siberia (north-eastern spurs of the Great Vasyugan Mire, 56°58′ N 82о36’ E). In total, 42 samples taken from the cores were collected. The protocol included spectroscopy (in infrared and visible ranges) and biological model with peritoneal activated macrophages as a reference method to directly measure biological activity of HAs. Resultsand discussion. Numerical experiments confirmed consistency of the measured and estimated bioactivity, coefficient of determination R2 = 0.97. These experiments also showed that the MLP model significantly outperforms conventional linear multiple regression models, mainly due to essential nonlinearity of structure-activity relationships. ConclusionsOur research demonstrates that biological activity of HAs extracted from peat samples can be estimated using an artificial neural network model trained on infrared and visible spectra.

606-P: Artificial Neuronal Network Model for Three-Month Prognosis in Diabetic Foot Syndrome

Background: To make a reliable prognosis on wound healing in patients with diabetic foot syndrome (DFS) is extremely difficult. We used the artificial neuronal network (ANN) to identify the most significant variables which affect DFS healing process. We also aimed at providing data for designing a digital application which would help practitioners to predict the course of DFS. Methods: We enrolled 213 DFS patients and examined them using a variety of diagnostic tests. The patients were followed for three months. In the initial model, we assessed 35 clinical and biochemical variables. Subsequently, with the help of traditional statistics we reduced their number to twelve, and after conducting the sensitivity analysis, we found out that the final number of significant variables is as low as six. Results: The most significant variables in predicting the outcome of DFS treatment were: probe-to-bone test result, presence of blood flow in Doppler probe, prior amputation within the foot, erythrocyte sedimentation rate, and the area and duration of the ulceration. The area under the ROC curve was 0.87 (Figure). The total accuracy was 85%, sensitivity 94.6%, specificity 66% and F1 score 89%. Conclusions: ANN can be used in the prediction of the DFS course. The algorithm, which is the source of a digital application, is particularly useful in identifying individuals with diabetic foot ulcerations who fail to be healed in three months. Disclosure A.A. Poradzka: Other Relationship; Self; Boehringer Ingelheim (Canada) Ltd., Urgo Medical. L. Czupryniak: None.

A Framework for Underground Gas Storage System Reliability Assessment Considering Functional Failure of Repairable Components

Abstract As one of the most important means of nature gas peak shaving and energy strategic reserving, the reliability assessment of underground gas storage (UGS) system is necessary. Although many methods have been proposed for system reliability assessment, the functional heterogeneity of components and the influence of hydrothermal parameters on system reliability are neglected. To overcome these problems, we propose and apply a framework to assess UGS system reliability. Combining two-layer Monte Carlo simulation (MCS) technique with hydrothermal calculation, the framework integrates dynamic functional reliability of components into system reliability evaluation. To reflect the state transition process of repairable components and their impact on system reliability, the Markov model is introduced at system level. In order to improve the calculation speed, artificial neural network (ANN) model based on off-line MCS is established to replace the on-line MCS at components level. The proposed framework is applied to the reliability assessment and operation optimization of an UGS under different operation conditions. Compared with the traditional single-layer MCS method, the proposed method can not only reflect the variation of UGS reliability with hydrothermal parameters and operation time, but also can improve evaluation efficiency significantly.

Artificial Neural Network Model for Predicating Resilient Modulus of Silty Subgrade Soil

Recently machine learning is gaining acceptance in different civil engineering applications. In this study, an Artificial Neural Network (ANN) model is proposed to predict resilient modulus (MR) of a silty subgrade soil for pavement designs. A silty subgrade soil was compacted at the maximum dry density (γdopt) and optimum moisture content (OMC) according to the standard Proctor compaction. The resilient modulus test was then conducted on at least replicate samples of three groups of samples. The first group of samples were tested directly after compaction, the second group and third groups, after compaction at the standard Proctor effort were left in open air to dry over time or exposed to wetting to gain moisture. The testing results were then used to develop the ANN model. This model predicts MR of the soil based on water content (Wc), ratio of dry density over the maximum dry density at the optimum moisture content (γd/γdopt) and octahedral shear stress (τoct). After the ANN model architecture is set, the strengths and weaknesses of the developed model are examined by comparing the predicted versus measured MR values with respect to goodness-of-fit statistics. In addition, a sensitivity analysis of the model input parameters is performed.

Predicting GHG emissions from subway lines in the planning stage on a city level

The construction of urban rail transit (URT) infrastructures requires extensive input of resources and energy, thus generating considerable greenhouse gas (GHG) emissions. Despite efforts to quantify and mitigate the carbon footprint (CFP) during the URT’s construction stage, the magnitude of this CFP is largely dependent on design parameters, such as route selection and buried depth, that require confirmation during the planning stage. Therefore, the best means by which to control the CFP of URT construction is for government and design consultancies to have mitigation awareness in the determination of early proposal parameters. To achieve this goal, it is important to identify a link between the construction CFP and these parameters. In this paper, an artificial neuro network (ANN) model is established to predict the GHG emissions from the construction of the planning URT lines of the Fuzhou subway using training data from the in-service lines. GHG emissions from the construction of the planning lines amount to 6.46 Mt CO2.eq., averaging 53024.13 t CO2.eq. per unit line length. Buried depth is identified as the primary factor that determines the GHG intensity of URT stations and tunnel sections, and two simplified equations are derived to facilitate the estimation. In addition, the payback periods of the stations are evaluated. For lines 2, 4, 5, 6, it is estimated to require 16.52, 11.69, 21.60 and 12.67 years, respectively, before the initial construction CFPs are balanced by those mitigated by URT operation. Carbon-inefficient stations are attributed to either low service levels or overconstruction.

Coupling an artificial neuron network daylighting model and building energy simulation for vacuum photovoltaic glazing

Window plays an essential role in the indoor environment and building energy consumption. As an innovative building integrated photovoltaic (BIPV) window, the vacuum PV glazing was proposed to provide excellent thermal performance and utilize renewable energy. However, the daylighting performance of the vacuum PV glazing and the effect on energy consumption have not been thoroughly investigated. Most whole building energy simulation used the daylighting calculation based on Daylight Factor (DF) method, which fails to address realistic calculation for direct sunlight through complex glazing materials. In this study, a RADIANCE model was developed and validated to adequately represent the daylight behaviour of a vacuum cadmium telluride photovoltaic glazing with a three-layer structure. However, RADIANCE will consume too many computational resources for a whole year simulation. Therefore, an artificial neuron network (ANN) model was trained based on the weather conditions and the RADIANCE simulation results to predict the interior illuminance. Subsequently, a preprocessing coupling method is proposed to determine the lighting consumption of a typical office with the vacuum PV glazing. The performance evaluation of the ANN model indicates that it can predict the illuminance level with higher accuracy than the daylighting calculation methods in EnergyPlus. Therefore, the ANN model can adequately address the complex daylighting response of the vacuum PV glazing. The proposed coupling method showed a more reliable outcome than the simulations sole with EnergyPlus. Furthermore, the computational cost can be reduced dramatically by the ANN daylighting prediction model in comparison with the RADIANCE model. Compared with the lighting consumption determined by the ANN-based coupling method, the two approaches in EnergyPlus, the split-flux method and the DElight method, tend to underestimate the lighting consumption by 5.3% and 9.7%, respectively.

A sustainable approach to utilize olive pips for the sorption of lead ions: Numerical modeling with aid of artificial neural network

The ability of olive pips in the removal of lead ions from polluted water was investigated using three-layer artificial neural network and 114 batch experiments. The experiments revealed that the highest removal (99%) was achieved at pH 5, agitation time- 60 min, biosorbent dose- 0.3 g/100 mL of liquid phase, speed- 200 rpm at room temperature and 50 mg/L initial concentration. The transfer functions in the artificial network model were linear and tangent sigmoid for output and hidden layers, respectively where eight neurons presented a good description for results of batch experiments with coefficient of correlation >0.997. The analysis of sensitivity proved that the initial pH of solution had more influence in the present sorption process with relative importance of 34%. The outputs of artificial network were integrated with the solute transport equation for description the propagation of lead ions through fixed bed column packed with olive pips. For breakthrough curves, a satisfactory concurrence was observed between outputs of model and measured concentrations because Willmott's index was >0.97.

Models for estimating daily photosynthetically active radiation in oceanic and mediterranean climates and their improvement by site adaptation techniques

In this work Photosynthetically Active Radiation (PAR) in oceanic and mediterranean climates is modeled. Twenty-two different models have been developed and tested: eleven Multilinear Regression (MR) models and eleven Artificial Neuron Network (ANN) models, using combinations of variables such as Global Horizontal Irradiance (GHI), Global Extraterrestrial Irradiance (G0), Temperature (T) and Relative Humidity (RH). Data provided by Satellite Application Facility on Climate Monitoring (CM SAF) are used to develop and train the models, while the models have been validated using field data from four stations located in Spain, covering the different study climates.According to the results, zones with different climate conditions need different models, both for the case of MR and ANN. The results show the need of including the GHI in all models in order to obtain accurate estimates; in fact, the presence of more variables only improves slightly the results in mediterranean climate, while in oceanic climate no improvement is observed.On the other hand, comparing MR and ANN models, ANN models did not show better results than those of MR models in no one of the cases studied. Regarding the climate, both types of models are clearly better for the mediterranean case than for the oceanic one. In order to improve the performance of the model for oceanic climate a correction based on the site adaptation technique was carried out. The good results obtained by this technique fully justify its use.The best proposed models provide better performance than other models which are restricted to certain locations. Besides, the clustering technique based on the PAR variable, used in this work, allows obtaining useful models for a whole region. Finally, another advantage of this methodology is that there is no need of ground measurements for its development, except for the site adaptation technique.