Artificial Neural Network Output Research Articles

MINIATURIZATION OF A PIFA ANTENNA FOR BIOMEDICAL APPLICATIONS USING ARTIFICIAL NEURAL NETWORKS

This work deals with the optimization of an inverted F dual-band implantable antenna operating in Medical Device Radiocommunications Service (MedRadio, 401-406 MHz) and Industrial Scientific Medical (ISM, 902-928 MHz) applications bands. Artificial neural networks (ANNs) are implemented to minimize the size of the initial design. The ANN's output with the physical and dielectric parameters of antenna as inputs is tested using COMSOL Multiphysics®. The obtained results regarding the return loss S11, resonant frequency and bandwidth of the antenna are presented and discussed. Indeed, the size of the antenna is reduced by 21.48% with respect to the initial size while preserving its initial good performance in both frequency bands.

Estimating source rock parameters using wireline data: An example from Dezful Embayment, South West of Iran

Availability and high potential of wireline data to correlate with geochemical properties convinced petroleum geologists to use them for source rock detection and calculating their richness. In this research, intelligent techniques and mathematical relationships were used to evaluate source rock potential and organic matter type indirectly. Artificial neural networks (ANN) and ΔLogR techniques were successfully used to model the relation between wireline logs and Total Organic Carbon (TOC) content. Furthermore, wireline data and TOC values were used as input data to model a second ANN for S2 parameter (present hydrocarbon potential) calculation. Owing to its great ability in the course of solving non-linear problems with overwhelming complexity, the back propagation method was used to train the networks using 70 points datasets. Predicted TOC contents were validated by Rock-Eval pyrolysis results of which revealed dependency of the ΔLogR accuracy to the organic richness and lithology, and therefore higher precision of the ANN outputs in compare to the ΔLogR results.Hydrogen Index (HI) and kerogen type were also effectively predicted using mathematical relationship between TOC and S2 factors. Reliability of all steps of the methodology was approved using a 16 members retest dataset for which testing predicted parameters against measured data demonstrated good correlation coefficients combined with negligible errors. Finally, applicability of the methodology was checked by applying it on two wells in the Dezful Embayment to evaluate geochemical and depositional properties of the Paleocene-Early Oligocene Pabdeh Formation.

Toward operational shortwave radiation modeling and retrieval over rugged terrain

Shortwave radiation (0.3–3.0μm) is a dominant component of land surface all-wave radiation budget. Reliable estimation of shortwave radiation throughout the globe is particularly important for an in-depth understanding of global changes. Unfortunately, a large number of existing space-based algorithms ignore the effects of topography by simply assuming that the surface is ideally flat. As pointed out by many studies, such neglect toward topographic effects leads to significant errors in derived radiation. This study proposes a shortwave topographic radiation model (SWTRM) by quantifying solar direct radiation shielding, occlusion of sky radiation, reflected radiation from nearby terrain, and invisibility of certain targets. To drive the SWTRM, an artificial neural network (ANN) approach was employed to generate multiple components of the shortwave radiation budget. The new model was run over the Tibetan Plateau region and used MODIS products coupled with digital elevation data (DEM). Topographically corrected shortwave fluxes were derived by coupling SWTRM and ANN outputs. The results show that: (1) the proposed SWTRM works well over mountainous regions; (2) the ANN-based shortwave model provides reasonable retrieval accuracy (RMSE <60W/m2, bias<13W/m2 for all shortwave radiation components). More importantly, the ANN model can simultaneously provide all radiation components that are indispensable for driving the SWTRM; (3) over mountainous areas, the induced error can exceed 600W/m2 for shortwave net flux. Hence, topographic effects cannot be neglected; and (4) topography and solar illumination angle are key modulators in deriving shortwave radiation from space, which control the magnitude and spatial distribution of shortwave radiation over mountainous regions.

Experimental evaluation, new correlation proposing and ANN modeling of thermal properties of EG based hybrid nanofluid containing ZnO-DWCNT nanoparticles for internal combustion engines applications

Thermal conductivity of EG based hybrid nanofluid containing ZnO-DWCNT nanoparticles was investigated experimentally at concentration of 0.045 to 1.9% and a temperature of 30–50 °C. ZnO particles (with an average diameter of 10–30 nm) and double wall carbon nanotubes (DWCNT) (internal diameter of 3–5 nm and 5–15 nm external diameter) were mix at a ratio of 90%: 10% and dispersed in ethylene glycol (EG) then its thermal conductivity was measured. The results showed that maximum relative thermal conductivity (TCR) at temperature of 50 °C and the concentration of 1.9%, equivalent to 24.9%. Economic evaluation and qualitative performance showed that nanofluids hybrid compared with ZnO and nanofluids containing MWCNT, in terms of increasing thermal conductivity (TCE) and economically, is quite effective. A new correlation to predict TCR in terms of concentration of nanoparticles and the temperature was proposed. This correlation has a coefficient of determination (R-squared) and the maximum error of 0.9826 and 2.9%, respectively. The greatest sensitivity was calculated at a maximum temperature and solid volume fraction. Based on the TCR data the artificial neural network (ANN) was developed. The best case ANN containing two hidden layer and 3 neurons in each layer was obtained. This ANN has an R-squared and MSE and was equal to 0.9966% AARD and 1.3127e-05 and 0.0489, respectively. The comparison between experimetnal data, correlation and ANN outputs shows the accuracy and capability of ANN in modeling the TCR data.

ANN modeling, cost performance and sensitivity analyzing of thermal conductivity of DWCNT–SiO2/EG hybrid nanofluid for higher heat transfer

In this study, the thermal conductivity of SiO2–DWCNT/ethylene glycol hybrid nanofluid has been experimentally investigated on 0.03–1.71% solid volume fraction and temperatures from 30 to 50 °C. SiO2 and DWCNT’s nanoparticles dispersed in EG as base fluid, and its thermal conductivity was measured. The thermal conductivity was obtained 38% more than ethylene glycol thermal conductivity at some temperatures. A new correlation (R 2 = 0.9925) was proposed to predict experimental thermal conductivity ratio as a function of volume concentration and temperature. Also an artificial neural network was designed for thermal conductivity ratio data predicting. The best artificial neural network topology has two hidden layers with five neurons in each layer. Comparing the experimental thermal conductivity ratio with artificial neural network outputs and the correlation shows the high capacity and accuracy of artificial neural network in thermal conductivity ratio data predicting.

Simulation of Wind-Battery Microgrid Based on Short-Term Wind Power Forecasting

The inherently intermittent and highly variable nature of wind necessitates the use of wind power forecasting tools in order to facilitate the integration of wind turbines in microgrids, among others. In this direction, the present paper describes the development of a short-term wind power forecasting model based on artificial neural network (ANN) clustering, which uses statistical feature parameters in the input vector, as well as an enhanced version of this approach that adjusts the ANN output with the probability of lower misclassification (PLM) method. Moreover, it employs the Monte Carlo simulation to represent the stochastic variation of wind power production and assess the impact of energy management decisions in a residential wind-battery microgrid using the proposed wind power forecasting models. The results indicate that there are significant benefits for the microgrid when compared to the naïve approach that is used for benchmarking purposes, while the PLM adjustment method provides further improvements in terms of forecasting accuracy.

A high performance gas–liquid two-phase flow meter based on gamma-ray attenuation and scattering

The ability to precisely estimate the void fraction of multiphase flow in a pipe is very important in the petroleum industry. In this paper, an approach based on our previous works is proposed for predicting the void fraction independent of flow regime and liquid phase density changes in gas–liquid two-phase flows. Implemented technique is a combination of dual modality densitometry and multi-beam gamma-ray attenuation techniques. The detection system is comprised of a single energy fan beam, two transmission detectors, and one scattering detector. In this work, artificial neural network (ANN) was also implemented to predict the void fraction percentage independent of the flow regime and liquid phase density changes. Registered counts in three detectors and void fraction percentage were utilized as the inputs and output of ANN, respectively. By applying the proposed methodology, the void fraction was estimated with a mean relative error of less than just 1.2480%.

Artificial Neural Network to Predict Vine Water Status Spatial Variability Using Multispectral Information Obtained from an Unmanned Aerial Vehicle (UAV).

Water stress, which affects yield and wine quality, is often evaluated using the midday stem water potential (Ψstem). However, this measurement is acquired on a per plant basis and does not account for the assessment of vine water status spatial variability. The use of multispectral cameras mounted on unmanned aerial vehicle (UAV) is capable to capture the variability of vine water stress in a whole field scenario. It has been reported that conventional multispectral indices (CMI) that use information between 500–800 nm, do not accurately predict plant water status since they are not sensitive to water content. The objective of this study was to develop artificial neural network (ANN) models derived from multispectral images to predict the Ψstem spatial variability of a drip-irrigated Carménère vineyard in Talca, Maule Region, Chile. The coefficient of determination (R2) obtained between ANN outputs and ground-truth measurements of Ψstem were between 0.56–0.87, with the best performance observed for the model that included the bands 550, 570, 670, 700 and 800 nm. Validation analysis indicated that the ANN model could estimate Ψstem with a mean absolute error (MAE) of 0.1 MPa, root mean square error (RMSE) of 0.12 MPa, and relative error (RE) of −9.1%. For the validation of the CMI, the MAE, RMSE and RE values were between 0.26–0.27 MPa, 0.32–0.34 MPa and −24.2–25.6%, respectively.

ANN Modeling of Motional Resistance for Micro Disk Resonator

This article describes how modeling is an integral part of design and development of any system that provides the theoretical characterization of the system and helps in understanding the relations between various parameters of the system, before the system is developed. The capability of an Artificial Neural Network (ANN) to model the complex relations between a set of inputs and outputs is exploited to model the motional resistance and resonance frequency for a contour mode disk resonator. The solution was to develop a multilayer feed forward neural network. The data set required to train the ANN is obtained by developing an electrical equivalent model and through the MEMS simulation software Coventorware. The network is trained using a Levenberg Marquardt algorithm. The number of hidden layers and the number of neurons in each hidden layer is optimized using a genetic algorithm. The ANN model developed an efficient model of the motional resistance and resonance frequency of the disk resonator. The ANN output is compared with the output of an electrical equivalent model and a reported fabricated structure.

Prediction of temperature and HAZ in thermal-based processes with Gaussian heat source by a hybrid GA-ANN model

Thermal-based processes with Gaussian heat source often produce excessive temperature which can impose thermally-affected layers in specimens. Therefore, the temperature distribution and Heat Affected Zone (HAZ) of materials are two critical factors which are influenced by different process parameters. Measurement of the HAZ thickness and temperature distribution within the processes are not only difficult but also expensive. This research aims at finding a valuable knowledge on these factors by prediction of the process through a novel combinatory model. In this study, an integrated Artificial Neural Network (ANN) and genetic algorithm (GA) was used to predict the HAZ and temperature distribution of the specimens. To end this, a series of full factorial design of experiments were conducted by applying a Gaussian heat flux on Ti-6Al-4V at first, then the temperature of the specimen was measured by Infrared thermography. The HAZ width of each sample was investigated through measuring the microhardness. Secondly, the experimental data was used to create aGA-ANN model. The efficiency of GA in design and optimization of the architecture of ANN was investigated. The GA was used to determine the optimal number of neurons in hidden layer, learning rate and momentum coefficient of both output and hidden layers of ANN. Finally, the reliability of models was assessed according to the experimental results and statistical indicators. The results demonstrated that the combinatory model predicted the HAZ and temperature more effective than a trial-and-error ANN model.

Empirical study and model development of thermal conductivity improvement and assessment of cost and sensitivity of EG-water based SWCNT-ZnO (30%:70%) hybrid nanofluid

In this research, the effects of temperature and solid volume fraction on thermal conductivity of SWCNT-ZnO (30%–70%)/EG-water (60%–40%) hybrid nano fluid are studied experimentally. In order to present a model, two methods of correlation and artificial neural network modeling were utilized. From the results of the test it was revealed that the increment of temperature and solid volume fraction of nanoparticles in the base fluid would result in thermal conductivity enhancement (TCE). Investigating the sensitivity of relative thermal conductivity of the nanofluid to concentration variations is more than its sensitivity to temperature changes. A comparison of cost and TCE in terms of solid volume fraction signified that using hybrid nanofluid is more convenient from the aspects of cost and TCE improvement. With purpose of modeling the experimental data, a new correlation was presented where its R-squared was only 0.9918 which is an acceptable error. Also an artificial neural network was designed in order to model the experimental data, R-squared and mean-squared errors of this model were 0.9972 and 0.000037887 respectively. The outputs of artificial neural network were exceptionally accurate; this is an indication of the capabilities of such a model.

Noninvasive Continuous Blood Pressure Measurement Method Based on EEMD and ANN

Blood pressure is an important index to measure the function of human cardiovascular system. In order to solve the problem of non-invasive continuous measurement of blood pressure in electronic sphygmomanometer, a noninvasive blood pressure measurement method based on EEMD (ensemble empirical mode decomposition) and ANN (artificial neural networks) were proposed. In the experiment, a total of 19 500 pulse wave signals from THE MIMIC DATABASE were analyzed and subsequently the pulse wave was decomposed by EEMD. Furthermore, 10 characteristic parameters of the 4th layer decomposition signal were extracted as the input of ANN. The blood pressure corresponding to the pulse wave was taken as the output of ANN to train the BP (blood pressure) model. The error analysis of the model was carried out. The results indicated that the error of the model meets the standards of the American Association for the advancement of medical instrumentation (AAMI). Therefore, this method can be employed in noninvasive continuous measurement of blood pressure.

Primary User Emulation Detection Algorithm Based on Distributed Sensor Networks

In this paper, we propose a novel primary user emulation (PUE) detection approach which employs a distributed sensor network, where each sensor node operates as an independent PUE detector. Distributed nodes collaborate in order to obtain the final detection results for the whole network. A voting algorithm is used to improve the performance of energy detection, while the classification is conducted by the nearest node in order to improve the efficiency of the detector. As a result of voting, if a potential primary user exists, then the features of the unknown user is compared with entries from the database in order to obtain a solid detection match. An artificial neural network (ANN) is used for the classification of an unknown user. To assess the accuracy of the detection result, we implement a reliability check at the output of ANN. The proposed algorithm is validated via computer simulations as well as by experimental hardware implementations using the Universal Software Radio Peripheral (USRP) software-defined radio (SDR) platform. The experiment results show that the distributed network detector detects the PUE 180–200%, depending on the number of primary users, faster than single node detector.

High-Power LED Photoelectrothermal Analysis Based on Backpropagation Artificial Neural Networks

As an electroluminescentdevice, the coupling relationship between light-emitting diode (LED) input currents, optical power, and LED junction temperature is a complicated multiphysics process. In this paper, a simplified LED photoelectron-thermal (PET) model by artificial neural network (ANN), which can translatemultiphysics field issue into a singlephysics field problem, ismentioned to study the coupling relationship. In the first, an LED lumens, optical power, and electric power at different temperatures are monitored in a temperature controlling integrating sphere. Then, a backpropagation (BP) ANN is trained by these data to construct an LED Photo-Electron-Thermal (PET) relationship. In addition, LED luminaire thermal analyzing is performed using a finite-element method on the outputs of the BP ANN. Finally, the advantage of this method in terms of saving computing resources and computing time is analyzed by comparing the degrees of freedom in different models. The result shows that at least 6.7 times computing resource are saved by this method, which will reduce the LED thermal management system analyzing time greatly. Finally, the application and extension of this model are discussed.

Multilevel inverter with space vector modulation: intelligence direct torque control of induction motor

An Intelligence direct torque control (DTC) of induction motor using multilevel inverter with space vector modulation (SVM) is proposed in this study. The novelty of the proposed intelligence technique is an artificial neural network (ANN), which is utilised for the DTC of the induction motor. Here, ANN has been trained by using the target parameters such as reference quadrature and direct axis voltage with the corresponding input parameters change in motor torque and flux. The change in motor torque and flux parameters are identified by using the multilevel inverter output voltage and current. The back propagation process is used for the training process of the ANN. By using the ANN outputs, the SVM technique develops the control pulses for the multilevel inverter. Then the proposed intelligence technique is implemented in the MATLAB/simulink platform and the effectiveness is analysed by comparing with the different case studies and different methods. The comparison results demonstrate the superiority of the proposed approach and confirm its potential to solve the problem.

Experimental evaluation, sensitivity analyzation and ANN modeling of thermal conductivity of ZnO-MWCNT/EG-water hybrid nanofluid for engineering applications

The thermal conductivity of ZnO-MWCNT / EG-water hybrid nanofluid was experimentally investigated at solid volume fractions of 0.02–1% and temperatures of 30–50°C. The thermal conductivity of a nanofluid was measured in the 50%:50% mixture of ethylene glycol (EG) and water. The results showed that the maximum thermal conductivity ratio (TCR) is obtained at 50 ˚C and at the solid volume fraction of 1%. A new correlation was proposed for prediction of experimental TCR. The coefficient of determination (R-squared) of this correlation and its maximum regression error were equal to 0.9864 and 2.6% respectively. The sensitivity measurement results showed that the hybrid nanofluid shows the highest sensitivity at the highest temperature and solid volume fraction. An artificial neural network (ANN) was developed based on the experimental TCR data. In the best case, an ANN with two hidden layers and four neurons in each layer was obtained. Comparisons of experimental TCR with ANN and correlation output indicated high capability and accuracy of ANN in modeling of TCR data.

Searching for turbulence models by artificial neural network

Artificial neural network (ANN) is tested as a tool for finding a new subgrid model of the subgrid-scale (SGS) stress in large-eddy simulation. ANN is used to establish a functional relation between the grid-scale (GS) flow field and the SGS stress without any assumption of the form of function. Data required for training and test of ANN are provided by direct numerical simulation (DNS) of a turbulent channel flow. It is shown that ANN can establish a model similar to the gradient model. The correlation coefficients between the real SGS stress and the output of ANN are comparable to or larger than similarity models, but smaller than a two-parameter dynamic mixed model.

Performance predicting of 2D and 3D submerged hydrofoils using CFD and ANNs

In the present study, hydrodynamic performance of 2D and 3D submerged hydrofoils in terms of various geometries were simulated by computational fluid dynamic (CFD). Then, by selecting optimal artificial neural networks (ANN) hydrodynamic performance of hydrofoils are predicted. For this purpose, a finite volume method based on Navier–Stokes equation solver available in OpenFOAM, open-source CFD software, was used. After mesh size analyzing, to verify computational procedure, numerical results were compared with experimental ones which appropriate accuracy was observed. In this simulation, environmental and geometrical conditions such as, angle of attack, Reynolds number (Re), aspect ratio (AR) and taper ratio (TR) of hydrofoils are relevant on performance criteria of lift to drag ratio (LDR). To select a proper feed-forward ANNs to predict the performance of 2D and 3D hydrofoils under considered conditions based on the iterative algorithm, ANN architecture analysis was conducted. According to CFD results, larger value of AR and lower TR lead to greater LDR for 3D hydrofoils. Meanwhile, ANNs output showed that the maximum mean square error in predicting the LDR of 2D and 3D submerged hydrofoils are 0.0043 and 0.0035, respectively. In addition, based on the ANN weights and bias, two set of equations for predicting LDR of considered 2D and 3D submerged hydrofoils were proposed.

Comparison between Possibilistic c-Means (PCM) and Artificial Neural Network (ANN) Classification Algorithms in Land use/ Land cover Classification

There are several statistical classification algorithms available for land use/land cover classification. However, each has a certain bias or compromise. Some methods like the parallel piped approach in supervised classification, cannot classify continuous regions within a feature. On the other hand, while unsupervised classification method takes maximum advantage of spectral variability in an image, the maximally separable clusters in spectral space may not do much for our perception of important classes in a given study area. In this research, the output of an ANN algorithm was compared with the Possibilistic c-Means an improvement of the fuzzy c-Means on both moderate resolutions Landsat8 and a high resolution Formosat 2 images. The Formosat 2 image comes with an 8m spectral resolution on the multispectral data. This multispectral image data was resampled to 10m in order to maintain a uniform ratio of 1:3 against Landsat 8 image. Six classes were chosen for analysis including: Dense forest, eucalyptus, water, grassland, wheat and riverine sand. Using a standard false color composite (FCC), the six features reflected differently in the infrared region with wheat producing the brightest pixel values. Signature collection per class was therefore easily obtained for all classifications. The output of both ANN and FCM, were analyzed separately for accuracy and an error matrix generated to assess the quality and accuracy of the classification algorithms. When you compare the results of the two methods on a per-class-basis, ANN had a crisper output compared to PCM which yielded clusters with pixels especially on the moderate resolution Landsat 8 imagery.

Artificial Neural Network Modelling of Total Dissolved Solid (Elzawia city – Libya, as a case study)

In this study Mathematical, and statistical methods to simulate an aquifer water quality parameters have been considered. It is necessary to measured water quality for groundwater zonation maps. Therefore, more accurate maps produced more essential role in the discussionmanagement. Water samples were collected from thirty wells at Elzawi city located around 45km west of Tripoli and analysed for water quality parameters including: EC(Electrical Conductivity), TDS(Total Dissolved Solids), Ca (Calcium), Mg (Magnesium), and pH using standard methods. The wells location andwater level at wells observed by Global positioning system (GPS) type Garmin GPS 12XL. An artificial neural network (ANN) models were investigated to predict the TDS in water of Elzawi city wells. The input variables were the wells longitude, latitude, EC, Ca, Mg, and pH while the TDS in water was the output. The Levenberg–Marquardt (LM) algorithm and Back propagation used for training of the feed forward ANN. The ANN models performance were compared using the coefficient of determination(E), mean absolute percentage error (MAPE %), and 95% confidence interval(CI95%).It has been a good agreement between actual data and the ANN outputs for training, validation and testing data sets at the forth model (ANN4) while based on all inputs . ANN4 model performed superior to the other models in predicting TDS with high E=0.94 and lowest MAPE= 5.6% and the result average within the range of observed 95CI%. Also the ANN models could be successfully applied and provide high accuracy and reliability for water quality parameters forecasting.