Root Mean Square Of Error Research Articles

Does the Extension Torque Control Differ between Injured and Uninjured Knees of ACL-Deficient Individuals?

Background: Knee extension torque control decreases after anterior cruciate ligament (ACL) rupture. There is a controversy in neuromuscular control changes on the uninvolved side. We intended to evaluate the steadiness and accuracy of quadriceps muscle control in the healthy and deficient sides of people with acute ACL rupture. Methods: In this cross-sectional study, thirteen men with ACL rupture (age: 27.8±7.0, body mass index (BMI): 24.7±2.25: 24.7, days from injury: 48.1±21.3) participated in the study. We measured quadriceps force control, which is quantitatively assessed by the standard deviation (SD) of joint torque for a predefined submaximal target. The accuracy of muscular control or performance of quadriceps is commonly quantified by the root mean square of error (RMSE) was also measured. A two-way analysis of variance was conducted to assess SD and RMSE of two levels of quadriceps contraction (30% of muscle voluntary contraction (MVC), 50% of MVC) across both healthy and deficient knees. Results: There was a significant main effect for SD and RMSE of MVC percentage (p<0.001). SD of quadriceps torque in 50% of MVC (1.44 ± 0.13) was higher than 30% of MVC (0.88 ± 0.1). In contrast, there was no significant main effect for SD and RMSE of knee condition. Conclusion: After unilateral ACL rupture, the neuromuscular system becomes defected and quadriceps muscle control is then reduced in the healthy side. Therefore, the healthy side is also vulnerable to ligamentous damage. Besides, with the increasing intensity of physical activities, neuromuscular control decreases and the risk of re-injury rises.

Examining the role of receptivity to green communication, altruism and openness to change on young consumers’ intention to purchase green apparel: A multi-analytical approach

The current study aims to examine the impact of receptivity to green communication, altruism and openness to change on young consumers' purchase intention towards green apparel in India. Using a two stage approach, the research model is assessed with data collected through an online survey from a sample of 549 young consumers. In the first step, structural equation modeling (SEM) is used to determine the linear relationship among factors in the theoretical model. In the second step, artificial neural network (ANN) analysis is used to determine the predictive accuracy of the significant factors obtained from the SEM. The results from the SEM analysis reveal that receptivity to green communication, altruism and openness to change, along with the three predictors of the theory of planned behavior (TPB) significantly affect consumers’ purchase intention towards green apparel. Further, there is an indirect effect of altruism and openness to change on purchase intention through attitude supporting the value-attitude-behavior cognitive hierarchy (VABH). Besides, perceived behavioral control is found to mediate the impact of openness to change on intention to purchase green apparel. The root mean square of error (RMSE) values obtained from the ANN analysis suggest that the models show good predictive accuracy. The study contributes to the existing literature by confirming the applicability of the TPB and the VABH as well as suggesting a robust predictive framework by using ANN for green apparel purchase intention.

Accuracy evaluation of surface registration algorithm using normal distribution transform in stereotactic body radiotherapy/radiosurgery: A phantom study

PurposeTo evaluate a feasibility of normal distribution transform (NDT) algorithm compared with the iterative closest point (ICP) method as a useful surface registration in stereotactic body radiotherapy (SBRT)/stereotactic radiosurgery (SRS).MethodsPoint cloud images using the 3D triangulation technology were obtained from a depth camera‐based optical imaging (OSI) system equipped in a radiosurgery room. Two surface registration algorithms, NDT and ICP, were used to measure and compare the discrepancy values between the reference and the current surfaces during the positioning of the patient. The performance evaluation was investigated by calculating the registration error and root‐mean‐square (RMS) values for the surface model, reposition, and target accuracy, which were analyzed statistically using a paired t‐test.ResultsFor surface model accuracy, the average of the registration error and RMS values were measured as 3.56 ± 2.20 mm and 6.98 ± 1.89 mm for ICP method, and 1.76 ± 1.32 mm and 3.58 ± 1.30 mm for NDT method (p < 0.05). For reposition accuracy, the average registration error and RMS values were calculated as 1.41 ± 0.98 mm and 2.53 ± 1.64 mm using ICP method, and 0.92 ± 0.61 mm and 1.75 ± 0.80 mm using NDT method (p = 0.005). The overall target accuracy using the NDT method reduced the average of the reposition error and overall RMS value by 0.71 and 1.32 mm, respectively, compared to the ICP method (p = 0.03).ConclusionsWe found that the surface registration algorithm based on NDT method provides more reliable accuracy in the values of surface model, reposition, and target accuracies than the classic ICP method. The NDT method in OSI systems offers reasonable accuracy in SBRT/SRS.

Potential of Multiway PLS (N-PLS) Regression Method to Analyse Time-Series of Multispectral Images: A Case Study in Agriculture

Recent literature reflects the substantial progress in combining spatial, temporal and spectral capacities for remote sensing applications. As a result, new issues are arising, such as the need for methodologies that can process simultaneously the different dimensions of satellite information. This paper presents PLS regression extended to three-way data in order to integrate multiwavelengths as variables measured at several dates (time-series) and locations with Sentinel-2 at a regional scale. Considering that the multi-collinearity problem is present in remote sensing time-series to estimate one response variable and that the dataset is multidimensional, a multiway partial least squares (N-PLS) regression approach may be relevant to relate image information to ground variables of interest. N-PLS is an extension of the ordinary PLS regression algorithm where the bilinear model of predictors is replaced by a multilinear model. This paper presents a case study within the context of agriculture, conducted on a time-series of Sentinel-2 images covering regional scale scenes of southern France impacted by the heat wave episode that occurred on 28 June 2019. The model has been developed based on available heat wave impact data for 107 vineyard blocks in the Languedoc-Roussillon region and multispectral time-series predictor data for the period May to August 2019. The results validated the effectiveness of the proposed N-PLS method in estimating yield loss from spectral and temporal attributes. The performance of the model was evaluated by the R2 obtained on the prediction set (0.661), and the root mean square of error (RMSE), which was 10.7%. Limitations of the approach when dealing with time-series of large-scale images which represent a source of challenges are discussed; however, the N–PLS regression seems to be a suitable choice for analysing complex multispectral imagery data with different spectral domains and with a clear temporal evolution, such as an extreme weather event.

Communication-Efficient Coordinated RSS-Based Distributed Passive Localization via Drone Cluster

Multi-UAV passive localization via received signal strength (RSS) is extremely important for wide applications such as rescue and battlefield combat. However, the energy consumption of UAVs is a key issue in this UAVs-enabled application. Usually, the communication overhead plays an important role in the energy consumption. To address this problem, we design two distributed methods for this multi-UAV system with considerable performance under low communication overhead. Firstly, a distributed majorize-minimization (DMM) method is proposed. To accelerate its convergence, a tight upper bound of the objective function from the primary one is derived. Furthermore, a distributed estimation scheme using the Fisher information matrix (DEF) is presented, only requiring one round of communication between edge UAVs and central UAV. Simulation results show that the proposed DMM outperforms the existing distributed iterative methods in terms of root of mean square error (RMSE) under low communication overhead. Moreover, the most communication-effective DEF with local search estimation performs much better than the proposed DMM in terms of RMSE, but has a higher computational complexity.

Evaluating the accuracy of electron pencil beam dosimetry based on Monte Carlo simulations using homogeneous and heterogeneous phantoms

PurposeThis study aimed to assess the accuracy of the electron pencil beam algorithm (PBA) regarding Monte Carlo (MC) simulations for irradiating homogeneous and heterogeneous phantoms. MethodsThe PBA program was reproduced in the MATLAB software package. The lateral scattering factors were calculated using MC simulations and imported to the PBA program. The line dose values (profile curves) in various depths for a simple water phantom, a slab phantom with different layers, and a heterogeneous phantom with rods equivalent to bone tissue were calculated using PBA and compared to the MC simulation results for 6, 10, and 12 MeV electron beams. The mean differences and root mean square of error (RMSE) were obtained between the PBA and MC calculations. Furthermore, gamma analysis was used to compare the profile doses of PBA calculations with MC simulations. ResultsThe mean differences between the PBA and MC dose calculations were less than 1% for 10 and 12 MeV electron energies, and lower than 2% for the 6 MeV electron irradiation in all the assessed phantoms. The profile dose calculations showed that the separation of two isodose curves was less than 2 mm in central regions of the fields and less than 3 mm in the fields’ edge/penumbra regions and inhomogeneity interfaces. The gamma pass rate values ranged from 94.1 to 100% (average of 96.4%) in different phantoms, depths, and energy ranges. ConclusionThe PBA provided acceptable accuracy for radiotherapy electron dose calculations in a water phantom, slab phantom (with different materials), and heterogeneous phantom compared to MC dose simulations, as a reference dose calculation method.

Option Volatility Investment Strategy: The Combination of Neural Network and Classical Volatility Prediction Model

This study focuses on the volatility prediction and option volatility investment. By investigating the traditional Volatility Prediction Model and machine learning algorithms, this study tries to merge these two aspects together. This work setup a bridge of previous financial studies and machine learning studies by proposing an algorithm integrating neural network and three traditional volatility models, called “Quantile based neural network and model integration combination algorithm.” The algorithm effectively lowers the volatility prediction error (measured by root of mean square error, shorted for RMSE: 0.319724) and beat the Wavenet (RMSE: 0.44) which is the benchmark and surpasses integrated model (RMSE: 0.348346) in test set. In terms of option investment strategy, this paper constructs a CSI 300 index option portfolio which hedges the underlying asset price risk and exposes the volatility risk. Then propose the “Option strategy of volatility prediction with dynamic thresholds.” With the new algorithm above, the strategy improves the return‐risk ratio in test set (measured by Sharpe ratio: 1.99–2.07).

Modelling and prediction of hydrolysis index of gluten-free cookies from cardaba banana starch vis-?-vis response surface methodology and support vector machine.

The increase in the onset of celiac disease among the world populace had increased the demand for gluten-free products. Therefore, this study aimed at modelling and predicting the hydrolysis index of gluten-free cookies using response surface methodology (RSM) and support vector machine (SVM). The baking temperature (150 -180 ?) and baking time (15-25 min) were varied using a central composite design. The obtained result revealed that both modelling approaches (RSM and SVM) accurately predict the hydrolysis index of the gluten-free cookies owing to their higher coefficient of determinant (R2 &gt; 0.9). The predictive capability assessment of response surface methodology and support vector machine revealed the superiority of support vector machine (0.9658, 0.9329, 0.059) in predicting the hydrolysis index of the gluten-free cookies over response surface model (0.9613, 0.9241, 0.063) owing to its high correlation coefficient (R), Coefficient of determinant (R2) and lower mean square of error as well as root mean square of error (RMSE).

Statistical Analysis of Flow Parameters for the Graphical Simulation Outputs

System dynamics simulation software, in general, depicts graphical interpretations. The values of the parameters, on the other hand, are required for prediction. The goal of this research is to develop a novel multivariate model that can predict flow parameters while simulating flow under various scenarios. The project involves looking for variations in the streamline and constructing a new multivariate model for each elliptic cylinder system's velocity magnitude. Furthermore, the flow zones were split into three groups based on streamline behavior. As a result, utilizing simulation outputs, new models for flow zones are developed using linear and semiparametric regression. The best fitted model for each flow region was determined using mean square error (MSE), root of mean square error (RMSE), and mean absolute percentage error (MAPE). Based on the fitted smoothing curve of the velocity magnitude, a summary statistic and variability may be assessed. The presented models can be used to predict magnitude in any point of fluid flow using these models.

Highly effective agro-waste based functional green adsorbents for toxic chromium(VI) ion removal from wastewater

In this investigation, two novel and economically competitive adsorbents i.e., acid treated activated biochar (ATAB) and green synthesized graphene oxide iron-nanoparticles (GS-GO@FeNPs) were synthesized from waste-tea-leaves to exterminate chromium (Cr(VI)) from the contaminated water to ensure sustainable reuse of large-scale industrially produced tea waste. The adsorbents were characterized by several instrumental techniques whereas the adsorption experiments were performed in a batch process using Diphenylcarbazide methods. The characterization profile of the adsorbents suggested substantial Cr(VI) sequestration potential. The biosorption data was found to be better fitted with a pseudo-second order (PSO) kinetic model having significant regression co-efficient and providing negligible root square mean error (RSME). Isotherm studies demonstrated Langmuir and Freundlich models perfectly matched with the experimental data for both adsorbents. The maximum adsorption capacities were found to be 369.00 and 387.59 mg/g for ATAB and GS-GO@FeNPs, respectively. The outcomes of the artificial neural network (ANN) modelling provided a good correlation (R2 = 0.999) between simulated and experimental results of the adsorption process parameters. The adsorption process was endothermic in nature and its feasibility and spontaneity were confirmed by thermodynamic models. The adsorption mechanisms of Cr(VI) can be suggested as a complex process dominated by chemisorption phenomenon owing to ion-exchange, electrostatic attraction, surface complexation formation and film or intra-particle diffusion. The substrates displayed outstanding renewal performances by means of 1.0 M HCl/H2SO4 as eluent, and the adsorbents could be reused for several cycles. In essence, the adsorbents fabricated in the present study could be a promising, eco-friendly and cheap bio-filter for the remediation of Cr(VI) from aquatic environments to safe-guard the public health.

Improve real-time GNSS orbit with epoch-independent undifferenced ambiguity resolution

AR (Ambiguity Resolution) is one of the critical issues in high-accuracy GNSS (Global Navigation Satellite System) applications. Commonly, the double-difference AR technique is employed in GNSS network data processing. With the demand of high-accuracy, low-latency and high-integrity GNSS orbit, the orbit estimation in sequential mode becomes a candidate of that in batch mode. Unlike orbit estimation with double-difference AR in either batch or sequential mode, researches on real-time orbit estimation with undifferenced AR in sequential mode is still limited. In this study, real-time orbit determination with epoch-independent undifferenced AR in sequential mode is assessed. With one-week observation data from about 100 global distributed IGS (International GNSS Service) network stations, simulated real-time Galileo orbit determination are conducted with GFZ post-processed solution as reference for validation. Numerical results indicate that, first, the 3D error RMS (Root Mean Square) of fixed orbit is about 30% smaller than that of float orbit; second, the Galileo orbit error RMS reach 3.66, 3.60 and 3.09 cm for the radial, tangential and normal directions respectively for the fixed solution; third, small jumps exist in the real-time orbit and may be attributed to the temporal variations of UPD (Uncalibrated Phase Delay) and wrongly fixed ambiguities. In summary, the orbit quality is comparable to that of IGS real-time solutions, and further improvements may be achieved with improved AR constraint strategy and background orbit model.

The Surgeon's Role in Inducing and Controlling Motion Errors During Intraocular Membrane Peeling Procedures

Objectives:To understand the surgeon’s role in inducing and correcting movement inaccuracies during intraocular membrane peeling procedures.Materials and Methods:Optical sensors were used to record movement errors during actuation at the distal tip of 23-gauge pneumatic forceps both when the handle was handheld and when fixed with no human contact. Movements were also recorded at the proximal part of the forceps shaft (near the sclerotomy site) and compared to movement recorded at the distal end. The root mean square (RMS) and range values of the signals obtained from the sensors were calculated before and after applying high (7-13 Hz) and low (<5 Hz) frequency filters.Results:Comparison of RMS and range values of movement errors at the distal end of the forceps during actuation when the forceps handle was fixed and handheld showed that without human contact, these values were significantly lower in the X axis at all frequencies and in the Z axis at high frequencies compared to handheld (p<0.05), while there were no significant differences in the Y axis. Comparison of values from the distal and proximal ends of the forceps showed that when the forceps were fixed, RMS and range values were significantly higher for movement errors at the distal end compared to the proximal end at all frequencies (p<0.05). There was significant positive correlation between the extent of actuation and the RMS and range values for high-frequency movement errors but not low-frequency errors in all three axes with the fixed pneumatic handle (r=0.21-0.51, p<0.05).Conclusion:Surgeon- and non-surgeon-related errors are apparent in all axes, but skilled surgeons correct these errors through visual feedback, resulting in better correction in the visible planes. Sclerotomy sites provide a pivoting and stabilizing point for the shaft of the forceps and it is likely that skilled surgeons make use of the sclerotomy point to dampen motion errors, a skill worth teaching to beginners.

Effects of relative humidity on ethanol vapour releases from hydrophilic filmbased sachet in active food packaging

Active food package incorporating an ethanol vapour-controlled release sachet has been known for its efficacies to delay microbial proliferation in fresh fruit and vegetable. High humidity inside the package could be utilized as a stimulus for conditional releases as a means to stabilize the sachet prior to being used. The present research was undertaken to investigate the effects of relative humidity on ethanol vapour release from the hydrophilic film-based sachet. The prototype 4-side sealed sachets were made of either ethylene vinyl acetate (EVA) or laminated film comprising EVA and Nylon/PE (designated as ENP). A gas chromatogram equipped with a flame-ionized detector (FID-GC) was employed to analyze ethanol vapour concentration levels released from both sachet types and accumulated in headspaces of sealed glass beakers having different relative humidity (RH) levels. For a given RH level, the concentrations in the headspaces containing the ENPbased sachets were lower than those containing the EVA-based sachets. Delays of ethanol vapour release up to 24 h were observed in the ENP-based sachet system, whilst these did not occur among EVA-based sachets. Both sachets could release ethanol vapour with faster rates and subsequently higher concentrations accumulated at the very high relative humidity level (90-99% RH), compared to lower RH levels (60-89% RH). However, the release rates and concentration levels accumulated in 60-75% RH were not different from those in 80-89% RH. Extents of water vapour uptake by films were relatively small when the films were kept at the lower RH levels, but these became exponentially increased when the RH levels were ≥90%RH. Experimental data on water vapour uptakes were well predicted by an exponential model (R2 0.92-0.99; and root mean square of errors (RMSE) 0.004-0.054). Overall, experiment findings indicate that the ENP film caused delayed ethanol vapour releases from the sachet. The relative humidity levels had significant effects on the releases from hydrophilic film-based sachets

Design and field evaluation of cooperative adaptive cruise control with unconnected vehicle in the loop

To fully harvest the benefits of vehicular automation and connectivity in the mixed traffic, a cooperative longitudinal control strategy named Cooperative Adaptive Cruise Control with Unconnected vehicle in the loop (CACCu) has been proposed. When encountering an unconnected preceding vehicle, CACCu enables a Connected and Automated Vehicle (CAV) to benefit from communicating with a connected vehicle further ahead, rather than completely falling back to Adaptive Cruise Control (ACC). To validate the feasibility of CACCu, this study developed and tested a CACCu system with real vehicles in the field. A speed-command-based CACCu controller is designed and parameterized for optimizing the anticipated string stability. The experiment was conducted with two automated vehicles equipped with Mobileye sensors and Wi-Fi modules. The car-following performance of CACCu, in comparison with ACC and human driving (as the ego vehicle), was evaluated in the real-traffic scenarios constructed using NGSIM vehicle trajectory data. Over the 6 test runs for each control method, it was found that CACCu reduced 10.8% acceleration Root Mean Square (RMS), 60.8% spacing error RMS and 6.2% fuel consumption from ACC’s, indicating advantages of CACCu in control accuracy, ride comfort and energy efficiency. Compared with human driving, CACCu also reduced 17.6% acceleration and 13.4% fuel consumption. More importantly, the CACCu was able to efficiently avoid the traffic disturbance amplifications that frequently happened to ACC and human driving, which means the string stability has been significantly improved by the CACCu.

Retrieval of water quality parameters from hyperspectral images using a hybrid feedback deep factorization machine model

Environmental protection of water resources is of critical importance to daily life of human beings. In recent years, monitoring the variation of water quality using remote sensing techniques has become prevalent. Unmanned aerial vehicle (UAV) based remote sensing techniques have been applied to quantitative retrieval of concentrations of water quality parameters including phosphorus, nitrogen, chemical oxygen demand (COD), biochemical oxygen demand (BOD), and chlorophyll a (Chl-a), successfully and efficiently. In this study, a novel method with deep factorization machine, spatial distribution pattern analysis, and probabilistic analysis engaged, named hybrid feedback deep factorization machine (HF-DFM), has been developed to quantitatively estimate concentrations of water quality parameters based on hyperspectral reflectance data on large scale effectively. Our proposed method is a unified model for quantifying concentrations of water quality parameters with an end to end structure, which integrates UAV based optical remote sensing techniques and deep learning to estimate concentrations of water quality parameters. Furthermore, our proposed model was applied to real-time quantitative monitoring the variation of water quality of Mazhou River, Shenzhen, Guangdong, China. Finally, we evaluate the performance of proposed model on a real-world dataset in terms of root of mean squared error (RMSE), mean absolute percent error (MAPE), and coefficient of determination (R2). The experimental results show that our proposed model outperforms other state-of-the-art models with respect to RMSE, MAPE, and R2, where resulting MAPEs for quantifying all water quality parameters range from 8.78% to 12.36%, and resulting R2s range from 0.81 to 0.93. It can serve as a useful tool for decision makers in effectively monitoring water quality of urban rivers.

A kinetic analysis of the aflatoxin detoxification potential of lactic acid bacteria in Terxine (a cereal-based food)

Aflatoxin B1 (AFB1) is a hazardous component that can seriously threaten the public health. Terxine is a component used in traditional soup and found in the western mountainous regions of Iran. Several microorganisms have been reported to bind or degrade aflatoxins (AFs) in foods and feeds. This research aimed to investigate the effect of Terxine fermentation byLactobacillus plantarumstrains AF1 and LU5 on AFB1. Fermentation was carried out, and pH, lactic acid and AFB1 amount and microbial count were further determined. In addition, the kinetic experimental data of AFB1 by L. plantarumAF1 and LU5 (obtained at 37°C) were fitted to the zero-order, first-order and parabolic diffusion models. According to the coefficient of determination (R2) and root mean square of errors (RMSE), the zero-order model best described AF degradation. The growth of Lactobacillus strains was increased by the rise in the fermentation time; in this regard, the number of L. plantarum AF1 increased from 4.2 to 5.1 log cfu/g and that of L. plantarum LU5 increased from 4.1 to 5.2 log cfu/g in the first 8h, reaching 7.2 and 7.4 log cfu/g in the next 8h, respectively. The results also showed that the amount of lactic acid increased whereas the pH value decreased during the 24h fermentation. Both microorganisms reduced the amount of AFB1 while L. plantarum AF1 was more effective. Therefore, L. plantarumstrains AF1 and LU5 can be effectively used to reduce AFB1 in fermented foods.

Bayesian spatial modeling of poverty risk in Kelantan

Poverty data vary in rural areas, in certain states or regions, and some urban areas. These areal data tends to have spatial autocorrelation. A Bayesian hierarchical model is commonly used to estimates the risks using a combination of available covariate data and a set of spatial random effects. These random effects are commonly modelled by conditional autoregressive (CAR) prior distributions, a type of Markov random field model. Spatial autocorrelation between the random effects, ϕ in CAR models is induced by a n × n neighbourhood matrix, W. However, many studies assumed that the W is fixed when fitting the model. Therefore, this study evaluates the performance of the Poisson-log linear Leroux Conditional Autoregressive (CAR) model with m-nearest neighbourhood weight matrices using a simulation study. This study creates simulated poverty data for 66 districts of Kelantan with different scenarios that related with random effects and covariate. A Poisson log-linear Leroux CAR model with m = 1, 5 and 10 nearest neighbours are applied to the simulated poverty data. The performance of the models is evaluated using bias, Root Mean Square of Error (RMSE) and Deviance Information Criterion (DIC). The results show that the choice of m = 1, 5 and 10 neighbourhood matrices and scenarios do not affect the bias for either the regression parameter β or the risk Rk and RMSE for the risk Rk . Nevertheless, there is the dissimilarity of the performance of the models in the RMSE of regression parameter β. The results suggest that the Poisson log-linear Leroux CAR model with the m = 5 nearest neighbours performed overall best for simulated poverty data. It consistently gives good results across different strength of spatial autocorrelation of random effects ϕ and covariate. The model also gives the lowest DIC in all the scenarios, indicating a better fitting model than other models. The findings of this study give guidance in choosing the suitable m-nearest neighbourhood matrices to estimate the poverty risk in Kelantan.

Neural network-based estimation of lighting condition in indoor environment with improved brain storm algorithm

Purpose For decades, continuous research work is going on to maximize the power harvested from the sun; however, there is only a limited analysis on exploiting the microwatt output power from indoor lightings. Microelectronic system has power demand in the µW range, and therefore, indoor photovoltaics would be appropriate for micro-energy harvesting appliances. “Energy harvesting is defined as the transfer process by which energy source is acquired from the ambient energy, stored in energy storage element and powered to the target systems”. The theory of energy harvesting is: gathering energy from surroundings and offering technological solutions such as solar energy harvesting, wind energy collection and vibration energy harvesting. “The solar cell or photovoltaic cell (PV), is a device that converts light into electric current using the photoelectric effect”. Factors such as light source, temperature, circuit connection, light intensity, angle and height can manipulate the functions of PV cells. Among these, the most noticeable factor is the light intensity that has a major impact on the operations of solar panels. Design/methodology/approach This paper aims to design an enhanced prediction model on illuminance or irradiance by an optimized artificial neural network (ANN). The input attributes or the features considered here are temperatures, maxim, TSL, VI, short circuit current, open-circuit voltage, maximum power point (MPP) voltage, MPP current and MPP power, respectively. To enhance the performance of the prediction model, the weights of ANN are optimally tuned by a new self-improved brain storm optimization (SI-BSO) model. Findings The superiority of the implemented work is compared and proved over the conventional models in terms of error analysis and prediction analysis. Accordingly, the presented approach was analysed and its superiority was proved over other conventional schemes such as ANN, ANN-Levenberg–Marquardt (LM), adaptive-network-based fuzzy inference system (ANFIS) and brainstorm optimization (BSO). In addition, analysis was held with respect to error measures such as mean absolute relative error (MARE), mean square root error (MSRE), mean absolute error and mean absolute percentage error. Moreover, prediction analysis was also performed that revealed the betterment of the presented model. More particularly, the proposed ANN + SI-BSO model has attained minimal error for all measures when compared to the existing schemes. More particularly, on considering the MARE, the adopted model for data set 1 was 23.61%, 48.12%, 79.39% and 90.86% better than ANN, ANN-LM, ANFIS and BSO models, respectively. Similarly, on considering data set 2, the MSRE of the implemented model was 99.87%, 70.69%, 99.57% and 94.74% better than ANN, ANN-LM, ANFIS and BSO models, respectively. Thus, the enhancement of the presented ANN + SI-BSO scheme has been validated effectively. Originality/value This work has established an improved illuminance/irradiance prediction model using the optimization concept. Here, the attributes, namely, temperature, maxim, TSL, VI, Isc, Voc, Vmpp, Impp and Pmpp were given as input to ANN, in which the weights were chosen optimally. For the optimal selection of weights, a novel ANN + SI-BSO model was established, which was an improved version of the BSO model.

Support Vector Machine to Predict Electricity Consumption in the Energy Management Laboratory

Predicted electricity consumption is needed to perform energy management. Electricity consumption prediction is also very important in the development of intelligent power grids and advanced electrification network information. we implement a Support Vector Machine (SVM) to predict electrical loads and results compared to measurable electrical loads. Laboratory electrical loads have their own characteristics when compared to residential, commercial, or industrial, we use electrical load data in energy management laboratories to be used to be predicted. C and Gamma as searchable parameters use GridSearchCV to get optimal SVM input parameters. Our prediction data is compared to measurement data and is searched for accuracy based on RMSE (Root Square Mean Error), MAE (Mean Absolute Error) and MSE (Mean Squared Error) values. Based on this we get the optimal parameter values C 1e6 and Gamma 2.97e-07, with the result RSME (Root Square Mean Error) ; 0.37, MAE (meaning absolute error); 0.21 and MSE (Mean Squared Error); 0.14.

Convolutional Neural Networks to Estimate Dry Matter Yield in a Guineagrass Breeding Program Using UAV Remote Sensing.

Forage dry matter is the main source of nutrients in the diet of ruminant animals. Thus, this trait is evaluated in most forage breeding programs with the objective of increasing the yield. Novel solutions combining unmanned aerial vehicles (UAVs) and computer vision are crucial to increase the efficiency of forage breeding programs, to support high-throughput phenotyping (HTP), aiming to estimate parameters correlated to important traits. The main goal of this study was to propose a convolutional neural network (CNN) approach using UAV-RGB imagery to estimate dry matter yield traits in a guineagrass breeding program. For this, an experiment composed of 330 plots of full-sib families and checks conducted at Embrapa Beef Cattle, Brazil, was used. The image dataset was composed of images obtained with an RGB sensor embedded in a Phantom 4 PRO. The traits leaf dry matter yield (LDMY) and total dry matter yield (TDMY) were obtained by conventional agronomic methodology and considered as the ground-truth data. Different CNN architectures were analyzed, such as AlexNet, ResNeXt50, DarkNet53, and two networks proposed recently for related tasks named MaCNN and LF-CNN. Pretrained AlexNet and ResNeXt50 architectures were also studied. Ten-fold cross-validation was used for training and testing the model. Estimates of DMY traits by each CNN architecture were considered as new HTP traits to compare with real traits. Pearson correlation coefficient r between real and HTP traits ranged from 0.62 to 0.79 for LDMY and from 0.60 to 0.76 for TDMY; root square mean error (RSME) ranged from 286.24 to 366.93 kg·ha for LDMY and from 413.07 to 506.56 kg·ha for TDMY. All the CNNs generated heritable HTP traits, except LF-CNN for LDMY and AlexNet for TDMY. Genetic correlations between real and HTP traits were high but varied according to the CNN architecture. HTP trait from ResNeXt50 pretrained achieved the best results for indirect selection regardless of the dry matter trait. This demonstrates that CNNs with remote sensing data are highly promising for HTP for dry matter yield traits in forage breeding programs.