Accurate Estimation Model Research Articles

Rice Yield Estimation Based on Vegetation Index and Florescence Spectral Information from UAV Hyperspectral Remote Sensing

Rice floret number per unit area as one of the key yield structure parameters is directly related to the final yield of rice. Previous studies paid little attention to the effect of the variations in vegetation indices (VIs) caused by rice flowering on rice yield estimation. Unmanned aerial vehicles (UAV) equipped with hyperspectral cameras can provide high spatial and temporal resolution remote sensing data about the rice canopy, providing possibilities for flowering monitoring. In this study, two consecutive years of rice field experiments were conducted to explore the performance of florescence spectral information in improving the accuracy of VIs-based models for yield estimates. First, the florescence ratio reflectance and florescence difference reflectance, as well as their first derivative reflectance, were defined and then their correlations with rice yield were evaluated. It was found that the florescence spectral information at the seventh day of rice flowering showed the highest correlation with the yield. The sensitive bands to yield were centered at 590 nm, 690 nm and 736 nm–748 nm, 760 nm–768 nm for the first derivative florescence difference reflectance, and 704 nm–760 nm for the first derivative florescence ratio reflectance. The florescence ratio index (FRI) and florescence difference index (FDI) were developed and their abilities to improve the estimation accuracy of models basing on vegetation indices at single-, two- and three-growth stages were tested. With the introduction of florescence spectral information, the single-growth VI-based model produced the most obvious improvement in estimation accuracy, with the coefficient of determination (R2) increasing from 0.748 to 0.799, and the mean absolute percentage error (MAPE) and the root mean squared error (RMSE) decreasing by 11.8% and 10.7%, respectively. Optimized by flowering information, the two-growth stage VIs-based model gave the best performance (R2 = 0.869, MAPE = 3.98%, RMSE = 396.02 kg/ha). These results showed that introducing florescence spectral information at the flowering stage into conventional VIs-based yield estimation models is helpful in improving rice yield estimation accuracy. The usefulness of florescence spectral information for yield estimation provides a new idea for the further development and improvement of the crop yield estimation method.

Assessment of protein model structure accuracy estimation in CASP14: Old and new challenges.

In CASP, blind testing of model accuracy estimation methods has been conducted on models submitted by tertiary structure prediction servers. In CASP14, model accuracy estimation results were evaluated in terms of both global and local structure accuracy, as in the previous CASPs. Unlike the previous CASPs that did not show pronounced improvements in performance, the best single-model method (from the Baker group) showed an improved performance in CASP14, particularly in evaluating global structure accuracy when compared to both the best single-model methods in previous CASPs and the best multi-model methods in the current CASP. Although the CASP14 experiment on model accuracy estimation did not deal with the structures generated by AlphaFold2, new challenges that have arisen due to the success of AlphaFold2 are discussed.

A review and tests of validation and sensitivity of geolocation models for marine fish tracking

AbstractUncertainties in fish tracking studies limit their integration into conservation and fisheries management plans. This is especially true for archival tagging studies that rely on geolocation models to infer fish tracks from recorded environmental variables. Hidden Markov Models (HMMs) are increasingly popular to geolocate marine fish equipped with archival tags; however, true errors and sensitivity of geolocation HMMs are seldom evaluated. In this study, we first review validation methods and implementations of geolocation HMMs to adapt to regional oceanography, fish species and tag data. We then use a case‐study to evaluate strengths and limitations of each validation approach and to illustrate the sensitivity of geolocation HMMs to implementation assumptions. Simulated and fixed tag locations are the most widely implemented methods, but less common methods relying on true fish tracking, that is double‐tagging or distance from recapture experiments, provide more informative estimates of model accuracy and precision. Results showed that model performance can be improved using simple assumptions when pre‐processing tag data rather than using a complex movement behaviour model. In addition, accelerometer show potential to further parameterise geolocation models. Overall, results from our case‐study and previous studies showed that current geolocation HMMs have average errors of ca. 30–50 and 120 km for demersal and large pelagic fish, respectively. We suggest that these errors are acceptable for investigations at the scale of fisheries management units.

Comparison of four bio-inspired algorithms to optimize KNEA for predicting monthly reference evapotranspiration in different climate zones of China

Accurate estimation of reference crop evapotranspiration (ET0) is of great significance to crop water use and agricultural water resources management. This study evaluated the performance of four bio-inspired algorithm optimized kernel-based nonlinear extension of Arps decline (KNEA) models, namely KNEA with grasshopper optimization algorithm (GOA-KNEA), KNEA with grey wolf optimizer algorithm (GWO-KNEA), KNEA with particle swarm optimization algorithm (PSO-KNEA), and KNEA with salp swarm algorithm (SSA-KNEA), on estimating monthly ET0 across China. Monthly meteorological data [including maximum air temperature (Tmax), minimum air temperature (Tmin), extra-terrestrial solar radiation (Ra), relative humidity (RH), global solar radiation (Rs), wind speed (U)] during 1966–2015 from 51 weather stations across the seven different climate zones of China were used for model training and testing. Four different combinations of meteorological data were applied as model input, and results from the FAO-56 Penman-Monteith formula were used as a control. Results showed that the GWO-KNEA model overall performed better than the other three coupling models, of which the GWO-KNEA2 model (i.e., the model with input combination 2) was the best (on average R2 = 0.9814, RMSE = 0.2143 mm d−1). The convergence rate and the population size of the GWO-KNEA model were also superior to the other three models. Among input combinations, models with combination 2 had the best overall performance, while models with combination 3 were the worst on average. In terms of the importance of each meteorological parameter contributing to model accuracy, Rs was greater than Ra, RH, or U. Among different climate zones, station specific models in the semi-arid steppe of Inner Mongolia showed the best estimating performance in general, while models in the Qinghai-Tibetan Plateau overall performed relatively poorly. The GOA-KNEA and SSA-KNEA models with the combination 4 showed large increases (29.7% and 28.7%, respectively), indicating a problem of overfitting. Conversely, the GWO-KNEA model was the most stable model with the smallest increase in RMSE during the testing phase (4.4–15.7%) among all models. In summary, taking the model's estimation accuracy, stability, convergence rate, and climate environment into account, our results suggest that GWO-KNEA would be a suitable model for estimating ET0 in different climate zones of China and the most practical input combination would be combination 2.

Protein model accuracy estimation empowered by deep learning and inter-residue distance prediction in CASP14

The inter-residue contact prediction and deep learning showed the promise to improve the estimation of protein model accuracy (EMA) in the 13th Critical Assessment of Protein Structure Prediction (CASP13). To further leverage the improved inter-residue distance predictions to enhance EMA, during the 2020 CASP14 experiment, we integrated several new inter-residue distance features with the existing model quality assessment features in several deep learning methods to predict the quality of protein structural models. According to the evaluation of performance in selecting the best model from the models of CASP14 targets, our three multi-model predictors of estimating model accuracy (MULTICOM-CONSTRUCT, MULTICOM-AI, and MULTICOM-CLUSTER) achieve the averaged loss of 0.073, 0.079, and 0.081, respectively, in terms of the global distance test score (GDT-TS). The three methods are ranked first, second, and third out of all 68 CASP14 predictors. MULTICOM-DEEP, the single-model predictor of estimating model accuracy (EMA), is ranked within top 10 among all the single-model EMA methods according to GDT-TS score loss. The results demonstrate that inter-residue distance features are valuable inputs for deep learning to predict the quality of protein structural models. However, larger training datasets and better ways of leveraging inter-residue distance information are needed to fully explore its potentials.

Development of asphalt mixture density estimation model applicable to wide air void content range using ground penetrating radar

Density is one of the key properties that strongly affect the service quality and life of asphalt mixture. Ground penetration radar (GPR) coupled with suitable density estimation model has been recognized as an effective method to nondestructively estimate the density of asphalt mixture. However, the estimation accuracy of conventional models was questioned for asphalt mixture with high void content. To fill this gap, this study aims at developing an electromagnetic mixing model to predict the density of asphalt mixture with a wide range of air void content based on GPR measured data. In this study, the performance of four classical models, including Al-Qadi, Lahouar and Leng (ALL) model, Rayleigh model, Böttcher model and complex refractive index (CRIM) model was compared. Then, the influence of neighboring inclusion polarization linking with the air void content was taken into account to propose the density estimation model based on the electromagnetic mixing theory. Finally, the proposed model was validated using the laboratory measured density of the field cored samples. It was found that the density predicted by the proposed model is highly consistent with that measured in laboratory, with the average ratio of estimation results to test results at 1.004 and the standard deviation at 0.015. In addition, the proposed model showed superior prediction with currently available models in terms of density estimation for asphalt mixture with different aggregate gradations and air void contents. Based on the results of this study, it is recommended to using the proposed model as an alternative to the currently available models in estimating the density of asphalt mixture based on GPR collected data.

LASSO-Penalized Clusterwise Linear Regression Modeling With Local Least Angle Regression (L-LARS)

In clusterwise regression analysis, the goal is to predict a response variable based on a set of explanatory variables, each with cluster-specific effects. Nowadays, the number of candidates is typically large: whereas some of these variables might be useful, some others might contribute very little to the prediction. A well known method to perform variable selection is the LASSO, with calibration done by minimizing the Bayesian Information Criterion (BIC). However, current LASSO-penalized estimators have several disadvantages: only certain types of penalties are considered; the computations might involve approximate schemes and can be very time consuming, with overly complex tuning of the penalty term. This is usually due to the possibly large number of times the estimator must be evaluated for each plausible value of the tuning parameter(s). To ease such computation, we introduce an Expectation Maximization (EM) algorithm with closed-form updates working with a very general version of the LASSO penalty. Such an EM is based on an iterative scheme where the component specific LASSO regression coefficients are computed according to a coordinate descent update. Least Angle Regression is then used to perform covariate selection by evaluating the estimator only once. The advantages of our proposal, in terms of computation time reduction and accuracy of model estimation and selection, are shown by means of a simulation study and illustrated with a real data application.

Applicability of optimized hyperspectral indices for estimating Betalain content in Suaeda salsa

ABSTRACT Betalains (Bt) are a collective group of natural, edible, red-yellow plant pigments, and are an important biochemical parameter in vegetable growth. Betalains are believed to have fungicidal activity and their presence in plants is a key to explaining the physiological response and resistance of plants caused by different environmental stress factors or seasonal fluctuations. Hyperspectral remote sensing has proven potential to understand many biochemical processes widely used in investigating vegetation growth condition. To better understand vegetation containing betalain pigments, for growth monitoring and agronomic decision-making, a novel betalain source, Suaeda salsa (S. salsa) was studied by extracting betalain at different phenological periods, and registering the results with time-course measurements of canopy spectral reflectance via hyperspectral remote imaging. Partial least squares regression (PLSR) analysis was undertaken with three spectral transformation methods (the raw hyperspectral reflectance (R), first derivative reflectance (FDR) and second derivative reflectance (SDR). The variable importance in projection (VIP) score resulting from PLSR model was used to determine the key spectral wavelengths and reduce the dimensionality of the hyperspectral reflectance data matrix. The study results demonstrated that Bt content had a significant correlation with the effective wavelength (FD-R 436nm, R 847nm, R 848nm, R 860nm, R 871nm, R 971nm and SD-R 437nm, R 496nm, R 499nm, R 508nm, R 740nm, R 772nm, R 774nm, R 780nm) and optimized spectral indices (simple normalized difference spectral indices (NDSI(802nm,408nm), NDSI(781nm,776nm)), normalized polarization indices (NPDI(781nm,775nm), NPDI(781nm,776nm)), ratio spectral indices (RSI(781nm,776nm), RSI(781nm,775nm)), chlorophyll indices(CI(781nm,776nm), CI(781nm, 775nm)) derived from first derivative spectral reflectance. The PLSR-5 (constructed with the spectral indices) model resulted in an R 2 val (determination coefficient) of 0.79 and root mean square error (RMSEval) was 0.81 μg cm−2. Compared to PLSR-3 model built by sensitive bands, the PLSR-5 model estimation accuracy increased 15%. These results suggest the possibility of estimating Bt content using hyperspectral indices, which contain sufficient information for a successful estimation. Study findings will help researchers in deciding suitable Bt indices for S. salsa growth status and stress assessment, and will facilitate detection growth condition of the vegetables with betalain pigments.

Estimation of soil Na + content based on measured hyperspectral and Sentinel-2B data in northern Ningxia, China.

To explore the ability of different sensors to estimate soil Na+ content, we got the mea-sured soil spectra and Sentinel-2B image spectra of the typical soil samples from the northern area of Ningxia. We filtered the sensitive parameters from the spectra data by means of stepwise regression (SR) and principal component regression analysis (PCA). We established the models to estimate soil Na+ content based on the measured spectra and image data using partial least square regression (PLSR), support vector machine (SVM) and back propagation neural network model (BPNN). The results showed that, except for Band9, there was significant correlation between the resampling data and the image data. The estimation accuracy of models based on SR-screening was generally higher than the PCA (excluding SVM model). The PCA-SVM model was the best image estimation model for soil Na+ content, with a prediction accuracy of 0.792. The SR-BPNN model was the best measured estimation model, with a prediction accuracy of 0.908. The estimating accuracy of the SR-PLSR image-spectra-based model increased from 0.481 to 0.798 after calibrated by the resampled measured spectrum model, which effectively enhanced the accuracy in estimating the soil Na+ content at large scale. We successfully made the spatial transformation of soil Na+ content from point to surface. Our results provided a scientific reference for Sentinel-2B image to monitor Na+ content in salinized soil.

Generalized Linear Distributed Lag Models

We propose distributed generalized linear models for the purpose of incorporating lagged effects. The model class provides a more accurate statistical measure of the relationship between the dependent variable and a series of covariates. The estimators from the proposed procedure are shown to be consistent. Simulation studies not only confirm the asymptotic properties of the estimators, but exhibit the adverse effects of model misspecification in terms of accuracy of model estimation and prediction. The application is illustrated by analyzing the presidential election data of 2016.

The Importance of Time Metric Precision When Implementing Bivariate Latent Change Score Models

The literature on latent change score models does not discuss the importance of using a precise time metric when structuring the data. This study examined the influence of time metric precision on model estimation, model interpretation, and parameter estimate accuracy in bivariate LCS (BLCS) models through simulation. Longitudinal data were generated with a panel study where assessments took place during a given time window with variation in start time and measurement lag. The data were analyzed using precise time metric, where variation in time was accounted for, and then analyzed using coarse time metric indicating only that the assessment took place during the time window. Results indicated that models estimated using the coarse time metric resulted in biased parameter estimates as well as larger standard errors and larger variances and covariances for intercept and slope. In particular, the coupling parameter estimates—which are unique to BLCS models—were biased with larger standard errors. An illustrative example of longitudinal bivariate relations between math and reading achievement in a nationally representative survey of children is then used to demonstrate how results and conclusions differ when using time metrics of varying precision. Implications and future directions are discussed.

PaMIR: Parametric Model-Conditioned Implicit Representation for Image-Based Human Reconstruction.

Modeling 3D humans accurately and robustly from a single image is very challenging, and the key for such an ill-posed problem is the 3D representation of the human models. To overcome the limitations of regular 3D representations, we propose Parametric Model-Conditioned Implicit Representation (PaMIR), which combines the parametric body model with the free-form deep implicit function. In our PaMIR-based reconstruction framework, a novel deep neural network is proposed to regularize the free-form deep implicit function using the semantic features of the parametric model, which improves the generalization ability under the scenarios of challenging poses and various clothing topologies. Moreover, a novel depth-ambiguity-aware training loss is further integrated to resolve depth ambiguities and enable successful surface detail reconstruction with imperfect body reference. Finally, we propose a body reference optimization method to improve the parametric model estimation accuracy and to enhance the consistency between the parametric model and the implicit function. With the PaMIR representation, our framework can be easily extended to multi-image input scenarios without the need of multi-camera calibration and pose synchronization. Experimental results demonstrate that our method achieves state-of-the-art performance for image-based 3D human reconstruction in the cases of challenging poses and clothing types.

Hyperspectral Estimation of Apple Canopy Chlorophyll Content Using an Ensemble Learning Approach

HighlightsMonitored the canopy chlorophyll content of apple trees using hyperspectral reflectance information.Constructed support vector machine combination regression model (C-SVR) based on five-fold cross validation and support vector machine regression approach.Compared estimation accuracy of ensemble learning models (C-SVR, RF), machine learning models (SVR, ANN), and PLSR models for apple canopy chlorophyll content.Abstract. Rapidly and effective monitoring of the canopy chlorophyll content (CCC) of apple trees is of great significance for crop stress monitoring in precision agriculture. This study attempted to use hyperspectral vegetation indices (VIs) to estimate the CCC of apple trees based on ensemble learning approach. In this study, vegetation indices combined by any two wavelengths from 400 to 1100 nm were constructed to calculate the correlation coefficient with the CCC in apple. We constructed a partial least squares regression model (PLSR), artificial neural network regression model (ANN), support vector machine regression (SVR), random forest regression (RF) model and support vector machine combination regression model (C-SVR) based on combinations of VIs to improve the estimation accuracy in apple CCC. The results showed that the correlation coefficients between NDVI (949,695), OSAVI (828,705), RDVI (741,725), RVI (716,707), DVI (572,532), and apple CCC were all above 0.76. The CCC estimation model using the RF and C-SVR approach constructed by the NDVI (949,695), OSAVI (828,705), RDVI (741,725), RVI (716,707), and DVI (572,532) achieved the better estimation results, and the R2V, RMSEV, and RPDV values of models were 0.76, 0.131(mg . g-1), 2.04 and 0.78, 0.127(mg . g-1), 2.12, respectively. Compared with the PLSR, ANN, and SVR model, the R2V and RPDV values of C-SVR model were increased by 4%, 1.2%, 3.8%, and 5.0%, 28.4%, 7.1%, respectively. The results show that using C-SVR approach to estimating the apple CCC can realize high accuracy of quantitative estimation. Ensemble learning approach is an effective method for monitoring the nutrient status of fruit trees based on hyperspectral technique. Keywords: Apple tree canopy, Chlorophyll content, Crop stress monitoring, Ensemble learning, Hyperspectral, Vegetation index.

Accuracy of Model Estimation versus Tanaka and Johnston Arch Length Analysis.

This study examines how accurate pediatric dentists are at estimating dental arch lengths by comparing their model estimations (guesstimating the arch length without measuring) to the Tanaka and Johnston mixed dentition arch length analysis. This study consisted of two parts, a survey of practitioners and a model estimating and measuring component. The survey was designed and given to 44 pediatric dentists to determine how many were practicing orthodontics and using arch length analyses routinely. Then 18 pediatric dentists and 13 pediatric dental residents examined 20 sets of mixed dentition models and estimated how much space was available. These estimations were compared to the calculated gold standard, the Tanaka and Johnston arch length analysis of the same models. More than half of the dentists surveyed that practice comprehensive orthodontics use arch length estimates. Pediatric dentists and pediatric dental residents are just as good as each other at estimating arch length. Pediatric dentists and pediatric dental residents underestimated arch length by -3.6 and -3.1 mm, respectively. More research needs to be done to determine if model estimation is a clinically acceptable way to judge arch length.

Integration of Crop Growth Model and Random Forest for Winter Wheat Yield Estimation From UAV Hyperspectral Imagery

Accurate and timely crop yield estimation is critical for food security and sustainable development. The rapid development of unmanned aerial vehicles (UAVs) offers a new approach to acquire high spatio-temporal resolution imagery of farmland at a low cost. In order to realize the full potential of UAV platform and sensor, machine learning has been introduced to estimate crop yield, but the shortages of field measurements have troubled researchers. In this article, the CW-RF model, a new wheat yield estimation model suitable for the North China plain, was established using random forest, and the crop growth model (the CERES-wheat model) was chosen to simulate abundant training samples for random forest at field plot scale. According to CERES-wheat model simulation, the leaf area index (LAI) and leaf nitrogen content (LNC) at the wheat jointing and heading stages were selected as the most sensitive parameters, and were retrieved from UAV hyperspectral imagery using the directional second derivative and angular insensitivity vegetation index methods, respectively. Then the retrieved LAI and LNC results were input into the CW-RF model to estimate winter wheat yield. The field validation in Luohe, Henan showed that the root-mean-squared error of the retrieved LAI and LNC were 6.27% and 12.17% at jointing stages, 9.21% and 13.64% at heading stages, respectively. The RMSE of estimated yield was 1,008.08 kg/ha, and the mean absolute percent error of estimated yield was 9.36%, demonstrating the available of the CW-RF model in wheat yield estimation at field plot scale. Apart from Luohe, validations in some other fields (e.g., Xiaotangshan, Beijing), prove the wide applicability of the CW-RF model. In addition, the UAV hyperspectral data were found to significantly improve the retrieval accuracy, and further improve CW-RF model estimation accuracy. In conclusion, this article showed that the CERES-Wheat model simulation can be important data source for machine learning-based wheat yield estimation model at field plot scale, and the hyperspectral sensor mounted on a UAV is a feasible remote sensing data acquisition mode for winter wheat growth monitoring and yield estimation.

Machine learning model development for predicting aeration efficiency through Parshall flume

This study compares several advanced machine learning models to obtain the most accurate method for predicting the aeration efficiency (E20) through the Parshall flume. The required dataset is obtained from the laboratory tests using different flumes fabricated in National Institute Technology Kurukshetra, India. Besides, the potential of K Nearest Neighbor (KNN), Random Forest Regression (RFR), and Decision Tree Regression (DTR) models are evaluated to predict the aeration efficiency. In this way, several input combinations (e.g. M1-M15) are provided using the laboratory parameters (e.g. W/L, S/L, Fr, and Re). Different predictive models are obtained based on those input combinations and machine learning models proposed in the present study. The predictive models are assessed based on several performance metrics and visual indicators. Results show that the KNN-M11 model (), which includes W/L, S/L, and Fr as predictive variables outperforms the other predictive models. Furthermore, an enhancement is observed in KNN model estimation accuracy compared to the previously developed empirical models. In general, the predictive model dominated in the present study provides adequate performance in predicting the aeration efficiency in the Parshall flume.

Deep Dichromatic Model Estimation Under AC Light Sources.

The dichromatic reflection model has been popularly exploited for computer vison tasks, such as color constancy and highlight removal. However, dichromatic model estimation is an severely ill-posed problem. Thus, several assumptions have been commonly made to estimate the dichromatic model, such as white-light (highlight removal) and the existence of highlight regions (color constancy). In this paper, we propose a spatio-temporal deep network to estimate the dichromatic parameters under AC light sources. The minute illumination variations can be captured with high-speed camera. The proposed network is composed of two sub-network branches. From high-speed video frames, each branch generates chromaticity and coefficient matrices, which correspond to the dichromatic image model. These two separate branches are jointly learned by spatio-temporal regularization. As far as we know, this is the first work that aims to estimate all dichromatic parameters in computer vision. To validate the model estimation accuracy, it is applied to color constancy and highlight removal. Both experimental results show that the dichromatic model can be estimated accurately via the proposed deep network.

Neural network-based model predictive tracking control of an uncertain robotic manipulator with input constraints

This paper proposes a neural network-based model predictive control (MPC) method for robotic manipulators with model uncertainty and input constraints. In the presented NN-based MPC structure, two groups of radial basis function neural networks (RBFNNs) are considered for online model estimation and effective optimization. The first group of RBFNNs is introduced as a predictive model for the robotic system with online learning strategies for handling the system uncertainty and improving the model estimation accuracy. The second one is developed for solving the optimization problem. By taking into account an actor–critic scheme with different weights and the same activation function, adaptive learning strategies are established for balancing between optimal tracking performance and predictive system stability. In addition, aiming at guaranteeing the input constraints, a nonquadratic cost function is adopted for the NN-based MPC. The ultimately uniformly boundedness (UUB) of all variables is verified through the Lyapunov approach. Simulation studies are conducted to explain the effectiveness of the proposed method.

Parameter Identification Method Based on Mixed-Integer Quadratic Programming and Edge Computing in Power Internet of Things

With the rapid development of power Internet of Things, its scale is becoming larger and larger. Many advanced applications depend on the accuracy of network model and state estimation, and the accuracy of network model and state estimation largely depends on network parameter error. Therefore, a parameter identification and estimation method based on mixed-integer quadratic programming (MIQP) and edge computing is proposed. Firstly, a “cloud-tube-edge-end” architecture of power Internet of Things is proposed, and the edge computing layer collects terminal data and conducts data analysis, which greatly reduces the computing pressure of cloud center. In this architecture, the local state estimation is used to limit the branch with error data in a specific range to prevent the measurement errors in other ranges from affecting the local estimation process. Then, the parameter identification model is transformed into MIQP model, and a penalty factor is introduced into the optimization model to identify the parameter error and measurement error in the process of minimizing the objective function. Finally, data encryption, identity authentication, and other methods are used in edge computing to achieve network security protection, so as to avoid network attacks and information leakage in the process of data transmission. The proposed method is tested and analyzed in IEEE 14-bus test system. The results show that the proposed method can accurately determine and identify the error data in a certain probability in the actual operation of the power grid, which is convenient for the controller to find out the wrong data in time and determine the source of the error data, so as to set a reasonable data value.

Experimental Evaluation of Different Microturbojet EGT Modeling Approaches

Microturbojets are one of the main thrust sources in several high-performance unmanned aerial vehicles (UAVs). In many applications these vehicles operate in conditions that may demand performance at the limits of microturbojets’ capabilities. These conditions can eventually result in component degradation and engine malfunction. Safety and reliability demand to monitor and diagnose the health of the motor during the operation of the vehicle. One of the main microturbojet health indicators is the exhaust gas temperature (EGT). The scope of this article is to evaluate the estimation accuracy and computational cost of several EGT models through extensive experimental data. The evaluated models include classical, data-based, and novel approaches: an adiabatic nozzle representation, a nozzle model with a lineal correction, a nozzle model including heat transfer, a genetic programming model, a feed-forward neural network, a polynomic approximation and a multiphysics model. This evaluation provides information regarding the advantages and disadvantages of each model. The results show that the best-suited model depends mainly on the application and available identification data.