Non-crossing Quantile Regression Neural Network as a Calibration Tool for Ensemble Weather Forecasts

Despite the maturity of ensemble numerical weather prediction(NWP),the resulting forecasts are still,more often than not,under-dispersed.As such,forecast calibration tools have become popular.Among those tools,quantile regression(QR)is highly competitive in terms of both flexibility and predictive performance.Nevertheless,a long-standing problem of QR is quantile crossing,which greatly limits the interpretability of QR-calibrated forecasts.On this point,this study proposes a non-crossing quantile regression neural network(NCQRNN),for calibrating ensemble NWP forecasts into a set of reliable quantile forecasts without crossing.The overarching design principle of NCQRNN is to add on top of the conventional QRNN structure another hidden layer,which imposes a non-decreasing mapping between the combined output from nodes of the last hidden layer to the nodes of the output layer,through a triangular weight matrix with positive entries.The empirical part of the work considers a solar irradiance case study,in which four years of ensemble irradiance forecasts at seven locations,issued by the European Centre for Medium-Range Weather Forecasts,are calibrated via NCQRNN,as well as via an eclectic mix of benchmarking models,ranging from the naïve climatology to the state-of-the-art deep-learning and other non-crossing models.Formal and stringent forecast verification suggests that the forecasts post-processed via NCQRNN attain the maximum sharpness subject to calibration,amongst all competitors.Furthermore,the proposed conception to resolve quantile crossing is remarkably simple yet general,and thus has broad applicability as it can be integrated with many shallow-and deep-learning-based neural networks.

Regression Method for Rail Fastener Tightness Based on Center-Line Projection Distance Feature and Neural Network

In the railway system,fasteners have the functions of damping,maintaining the track distance,and adjusting the track level.Therefore,routine maintenance and inspection of fasteners are important to ensure the safe operation of track lines.Currently,assessment methods for fastener tightness include manual observation,acoustic wave detection,and image detection.There are limitations such as low accuracy and efficiency,easy interference and misjudgment,and a lack of accurate,stable,and fast detection methods.Aiming at the small deformation characteristics and large elastic change of fasteners from full loosening to full tightening,this study proposes high-precision surface-structured light technology for fastener detection and fastener deformation feature extraction based on the center-line projection distance and a fastener tightness regression method based on neural networks.First,the method uses a 3D camera to obtain a fastener point cloud and then segments the elastic rod area based on the iterative closest point algorithm registration.Principal component analysis is used to calculate the normal vector of the segmented elastic rod surface and extract the point on the centerline of the elastic rod.The point is projected onto the upper surface of the bolt to calculate the projection distance.Subsequently,the mapping relationship between the projection distance sequence and fastener tightness is established,and the influence of each parameter on the fastener tightness prediction is analyzed.Finally,by setting up a fastener detection scene in the track experimental base,collecting data,and completing the algorithm verification,the results showed that the deviation between the fastener tightness regression value obtained after the algorithm processing and the actual measured value RMSE was 0.2196 mm,which significantly improved the effect compared with other tightness detection methods,and realized an effective fastener tightness regression.

Nonparametric Statistical Feature Scaling Based Quadratic Regressive Convolution Deep Neural Network for Software Fault Prediction

The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software with defects negatively impacts operational costs and finally affects customer satisfaction. Numerous approaches exist to predict software defects. However, the timely and accurate software bugs are the major challenging issues. To improve the timely and accurate software defect prediction, a novel technique called Nonparametric Statistical feature scaled QuAdratic regressive convolution Deep nEural Network (SQADEN) is introduced. The proposed SQADEN technique mainly includes two major processes namely metric or feature selection and classification. First, the SQADEN uses the nonparametric statistical Torgerson–Gower scaling technique for identifying the relevant software metrics by measuring the similarity using the dice coefficient. The feature selection process is used to minimize the time complexity of software fault prediction. With the selected metrics, software fault perdition with the help of the Quadratic Censored regressive convolution deep neural network-based classification. The deep learning classifier analyzes the training and testing samples using the contingency correlation coefficient. The softstep activation function is used to provide the final fault prediction results. To minimize the error, the Nelder–Mead method is applied to solve non-linear least-squares problems. Finally, accurate classification results with a minimum error are obtained at the output layer. Experimental evaluation is carried out with different quantitative metrics such as accuracy, precision, recall, F-measure, and time complexity. The analyzed results demonstrate the superior performance of our proposed SQADEN technique with maximum accuracy, sensitivity and specificity by 3%, 3%, 2% and 3% and minimum time and space by 13% and 15% when compared with the two state-of-the-art methods.

SRAFE:Siamese Regression Aesthetic Fusion Evaluation for Chinese Calligraphic Copy

Evaluation of calligraphic copy is the core of Chinese calligraphy appreciation and in-heritance.However,previous aesthetic evaluation studies often focussed on photos and paintings,with few attempts on Chinese calligraphy.To solve this problem,a Siamese regression aesthetic fusion method is proposed,named SRAFE,for Chinese calligraphy based on the combination of calligraphy aesthetics and deep learning.First,a dataset termed Evaluated Chinese Calligraphy Copies(E3C)is constructed for aesthetic evalu-ation.Second,12 hand‐crafted aesthetic features based on the shape,structure,and stroke of calligraphy are designed.Then,the Siamese regression network(SRN)is designed to extract the deep aesthetic representation of calligraphy.Finally,the SRAFE method is built by fusing the deep aesthetic features with the hand‐crafted aesthetic features.Experimental results show that scores given by SRAFE are similar to the aesthetic evaluation label of E3C,proving the effectiveness of the authors’method.

Forecasting model of residential load based on general regression neural network and PSO-Bayes least squares support vector machine

Firstly,general regression neural network(GRNN) was used for variable selection of key influencing factors of residential load(RL) forecasting.Secondly,the key influencing factors chosen by GRNN were used as the input and output terminals of urban and rural RL for simulating and learning.In addition,the suitable parameters of final model were obtained through applying the evidence theory to combine the optimization results which were calculated with the PSO method and the Bayes theory.Then,the model of PSO-Bayes least squares support vector machine(PSO-Bayes-LS-SVM) was established.A case study was then provided for the learning and testing.The empirical analysis results show that the mean square errors of urban and rural RL forecast are 0.02% and 0.04%,respectively.At last,taking a specific province RL in China as an example,the forecast results of RL from 2011 to 2015 were obtained.

Risk based security assessment of power system using generalized regression neural network with feature extraction

A comprehensive risk based security assessment which includes low voltage, line overload and voltage collapse was presented using a relatively new neural network technique called as the generalized regression neural network (GRNN) with incorporation of feature extraction method using principle component analysis. In the risk based security assessment formulation, the failure rate associated to weather condition of each line was used to compute the probability of line outage for a given weather condition and the extent of security violation was represented by a severity function. For low voltage and line overload, continuous severity function was considered due to its ability to zoom in into the effect of near violating contingency. New severity function for voltage collapse using the voltage collapse prediction index was proposed. To reduce the computational burden, a new contingency screening method was proposed using the risk factor so as to select the critical line outages. The risk based security assessment method using GRNN was implemented on a large scale 87-bus power system and the results show that the risk prediction results obtained using GRNN with the incorporation of principal component analysis give better performance in terms of accuracy.

Introducing atmospheric angular momentum into prediction of length of day change by generalized regression neural network model

The general regression neural network(GRNN) model was proposed to model and predict the length of day(LOD) change, which has very complicated time-varying characteristics. Meanwhile, considering that the axial atmospheric angular momentum(AAM) function is tightly correlated with the LOD changes, it was introduced into the GRNN prediction model to further improve the accuracy of prediction. Experiments with the observational data of LOD changes show that the prediction accuracy of the GRNN model is 6.1% higher than that of BP network, and after introducing AAM function, the improvement of prediction accuracy further increases to 14.7%. The results show that the GRNN with AAM function is an effective prediction method for LOD changes.

Prediction of blast boulders in open pit mines via multiple regression and artificial neural networks

The most important objective of blasting in open pit mines is rock fragmentation.Prediction of produced boulders(oversized crushed rocks) is a key parameter in designing blast patterns.In this study,the amount of boulder produced in blasting operations of Golegohar iron ore open pit mine,Iran was predicted via multiple regression method and artificial neural networks.Results of 33 blasts in the mine were collected for modeling.Input variables were:joints spacing,density and uniaxial compressive strength of the intact rock,burden,spacing,stemming,bench height to burden ratio,and specific charge.The dependent variable was ratio of boulder volume to pattern volume.Both techniques were successful in predicting the ratio.In this study,the multiple regression method was superior with coefficient of determination and root mean squared error values of 0.89 and 0.19,respectively.

Performance Prediction of Switched Reluctance Motor using Improved Generalized Regression Neural Networks for Design Optimization

Since practical mathematical model for the design optimization of switched reluctance motor(SRM)is difficult to derive because of the strong nonlinearity,precise prediction of electromagnetic characteristics is of great importance during the optimization procedure.In this paper,an improved generalized regression neural network(GRNN)optimized by fruit fly optimization algorithm(FOA)is proposed for the modeling of SRM that represent the relationship of torque ripple and efficiency with the optimization variables,stator pole arc,rotor pole arc and rotor yoke height.Finite element parametric analysis technology is used to obtain the sample data for GRNN training and verification.Comprehensive comparisons and analysis among back propagation neural network(BPNN),radial basis function neural network(RBFNN),extreme learning machine(ELM)and GRNN is made to test the effectiveness and superiority of FOA-GRNN.

An Approach to Carbon Emissions Prediction Using Generalized Regression Neural Network Improved by Genetic Algorithm

The study on scientific analysis and prediction of China’s future carbon emissions is conducive to balancing the relationship between economic development and carbon emissions in the new era,and actively responding to climate change policy.Through the analysis of the application of the generalized regression neural network(GRNN)in prediction,this paper improved the prediction method of GRNN.Genetic algorithm(GA)was adopted to search the optimal smooth factor as the only factor of GRNN,which was then used for prediction in GRNN.During the prediction of carbon dioxide emissions using the improved method,the increments of data were taken into account.The target values were obtained after the calculation of the predicted results.Finally,compared with the results of GRNN,the improved method realized higher prediction accuracy.It thus offers a new way of predicting total carbon dioxide emissions,and the prediction results can provide macroscopic guidance and decision-making reference for China’s environmental protection and trading of carbon emissions.

A hybrid model for predicting spatial distribution of soil organic matter in a bamboo forest based on general regression neural network and interative algorithm

A general regression neural network model,combined with an interative algorithm(GRNNI)using sparsely distributed samples and auxiliary environmental variables was proposed to predict both spatial distribution and variability of soil organic matter(SOM)in a bamboo forest.The auxiliary environmental variables were:elevation,slope,mean annual temperature,mean annual precipitation,and normalized difference vegetation index.The prediction accuracy of this model was assessed via three accuracy indices,mean error(ME),mean absolute error(MAE),and root mean squared error(RMSE)for validation in sampling sites.Both the prediction accuracy and reliability of this model were compared to those of regression kriging(RK)and ordinary kriging(OK).The results show that the prediction accuracy of the GRNNI model was higher than that of both RK and OK.The three accuracy indices(ME,MAE,and RMSE)of the GRNNI model were lower than those of RK and OK.Relative improvements of RMSE of the GRNNI model compared with RK and OK were 13.6%and 17.5%,respectively.In addition,a more realistic spatial pattern of SOM was produced by the model because the GRNNI model was more suitable than multiple linear regression to capture the nonlinear relationship between SOM and the auxiliary environmental variables.Therefore,the GRNNI model can improve both prediction accuracy and reliability for determining spatial distribution and variability of SOM.

Application of generalized regression neural network on fast 3D reconstruction

In robot-assisted surgery projects,researchers should be able to make fast 3D reconstruction. Usually 2D images acquired with common diagnostic equipments such as UT, CT and MRI are not enough and complete for an accurate 3D reconstruction. There are some interpolation methods for approximating non value voxels which consume large execution time. A novel algorithm is introduced based on generalized regression neural network (GRNN) which can interpolate unknown voxles fast and reliable. The GRNN interpolation is used to produce new 2D images between each two succeeding ultrasonic images. It is shown that the composition of GRNN with image distance transformation can produce higher quality 3D shapes. The results of this method are compared with other interpolation methods practically. It shows this method can decrease overall time consumption on online 3D reconstruction.

Parameters optimization for exponentially weighted moving average control chart using generalized regression neural network

As a useful alternative of Shewhart control chart, exponentially weighted moving average （EWMA） control chat has been applied widely to quality control, process monitoring, forecast, etc. In this paper, a method was introduced for optimal design of EWMA and multivariate EWMA （MEWMA） control charts, in which the optimal parameter pair （ λ, k） or （ λ, h ） was searched by using the generalized regression neural network （GRNN）. The results indicate that the optimal parameter pair can be obtained effectively by the proposed strategy for a given in-control average running length （ARLo） and shift to detect under any conditions, removing the drawback of incompleteness existing in the tables that had been reported.

Modelling the impact of climate change on rangeland forage production using a generalized regression neural network:a case study in Isfahan Province,Central Iran

Monitoring of rangeland forage production at specified spatial and temporal scales is necessary for grazing management and also for implementation of rehabilitation projects in rangelands. This study focused on the capability of a generalized regression neural network（GRNN） model combined with GIS techniques to explore the impact of climate change on rangeland forage production. Specifically, a dataset of 115 monitored records of forage production were collected from 16 rangeland sites during the period 1998–2007 in Isfahan Province, Central Iran. Neural network models were designed using the monitored forage production values and available environmental data（including climate and topography data）, and the performance of each network model was assessed using the mean estimation error（MEE）, model efficiency factor（MEF）, and correlation coefficient（r）. The best neural network model was then selected and further applied to predict the forage production of rangelands in the future（in 2030 and 2080） under A1 B climate change scenario using Hadley Centre coupled model. The present and future forage production maps were also produced. Rangeland forage production exhibited strong correlations with environmental factors, such as slope, elevation, aspect and annual temperature. The present forage production in the study area varied from 25.6 to 574.1 kg/hm^2. Under climate change scenario, the annual temperature was predicted to increase and the annual precipitation was predicted to decrease. The prediction maps of forage production in the future indicated that the area with low level of forage production（0–100 kg/hm^2） will increase while the areas with moderate, moderately high and high levels of forage production（≥100 kg/hm^2） will decrease both in 2030 and in 2080, which may be attributable to the increasing annual temperature and decreasing annual precipitation. It was predicted that forage production of rangelands will decrease in the next couple of decades, especially in the western and southern parts of Isfahan Province. These changes are more pronounced in elevations between 2200 and 2900 m. Therefore, rangeland managers have to cope with these changes by holistic management approaches through mitigation and human adaptations.

Term Structure of Interest Rates Based on Artificial Neural Network

In light of the nonlinear approaching capability of artificial neural networks （ ANN）, the term structure of interest rates is predicted using The generalized regression neural network （GRNN） and back propagation （BP） neural networks models. The prediction performance is measured with US interest rate data. Then, RBF and BP models are compared with Vasicek＇s model and Cox-Ingersoll-Ross （CIR） model. The comparison reveals that neural network models outperform Vasicek＇s model and CIR model, which are more precise and closer to the real market situation.

Twin model-based fault detection and tolerance approach for in-core self-powered neutron detectors

The in-core self-powered neutron detector(SPND)acts as a key measuring device for the monitoring of parameters and evaluation of the operating conditions of nuclear reactors.Prompt detection and tolerance of faulty SPNDs are indispensable for reliable reactor management.To completely extract the correlated state information of SPNDs,we constructed a twin model based on a generalized regression neural network(GRNN)that represents the common relationships among overall signals.Faulty SPNDs were determined because of the functional concordance of the twin model and real monitoring sys-tems,which calculated the error probability distribution between the model outputs and real values.Fault detection follows a tolerance phase to reinforce the stability of the twin model in the case of massive failures.A weighted K-nearest neighbor model was employed to reasonably reconstruct the values of the faulty signals and guarantee data purity.The experimental evaluation of the proposed method showed promising results,with excellent output consistency and high detection accuracy for both single-and multiple-point faulty SPNDs.For unexpected excessive failures,the proposed tolerance approach can efficiently repair fault behaviors and enhance the prediction performance of the twin model.

Projectile impact point prediction method based on GRNN

In order to forecast projectile impact points quickly and accurately,aprojectile impact point prediction method based on generalized regression neural network（GRNN）is presented.Firstly,the model of GRNN forecasting impact point is established;secondly,the particle swarm algorithm（PSD）is used to optimize the smooth factor in the prediction model and then the optimal GRNN impact point prediction model is obtained.Finally,the numerical simulation of this prediction model is carried out.Simulation results show that the maximum range error is no more than 40 m,and the lateral deviation error is less than0.2m.The average time of impact point prediction is 6.645 ms,which is 1 300.623 ms less than that of numerical integration method.Therefore,it is feasible and effective for the proposed method to forecast projectile impact points,and thus it can provide a theoretical reference for practical engineering applications.

Short-term Load Forecasting of Regional Distribution Network Based on Generalized Regression Neural Network Optimized by Grey Wolf Optimization Algorithm

Short-term load forecasting of regional distribution network is the key to the economic operation of smart distribution systems,which not only requires high accuracy and fast calculation speed,but also has a diversity of influential factors and strong randomness.This paper proposes a short-term load forecasting model for regional distribution network combining the maximum information coefficient,factor analysis,gray wolf optimization,and generalized regression neural network(MIC-FA-GWO-GRNN).To screen and decrease the dimension of the multiple-input features of the short-term load forecasting model,MIC is first used to quantify the non-linear correlation between the load and input features,and to eliminate the ineffective features,and then FA is used to reduce the dimension of the screened input features on the premise of preserving the main information of input features.After that the high-precision short-term丨oad forecasting based on GWO-GRNN model is realized.GRNN is used to regressively analyze the input features after screening and dimension reduction,and the parameter of GRNN is optimized by using the GWO,which has strong global searching ability and fast convergence.Finally a case study of a regional distribution network in Tianjin,China verifies the accuracy and applicability of the proposed forecasting model.

Underwater Positioning Based on an Artificial Lateral Line and a Generalized Regression Neural Network

Taking advantage of the lateral line organ, fish can navigate, feed, and avoid predators and obstacles by sensing surrounding flow fields. The lateral line organ provides an important reference for the development of new underwater detection technology. Inspired by the lateral line organ, in this paper, for the sake of localizing the target dipole source in three-dimensional underwater space, an artificial lateral line consisting of nine underwater pressure sensors forming a cross-shaped sensor array is applied. Combined with the method of gener- alized regression neural network, which is suitable for solving nonlinear pattern recognition problems, a corresponding experimental platform has been built to sample data for training the neural network from a 12 cm by 12 cm by 24 cm cuboid space. The experimental results indicate that the cross-shaped artificial lateral line can localize the target dipole source two body-lengths away. The well- performing perceptual distance is below 13 cm away from the sensing array. Moreover, decreasing the data sampling interval and in- creasing the number of sensors utilized can help improve the positioning accuracy.

A Fast Charging–Cooling Coupled Scheduling Method for a Liquid Cooling-Based Thermal Management System for Lithium-Ion Batteries

Efficient fast-charging technology is necessary for the extension of the driving range of electric vehicles.However,lithium-ion cells generate immense heat at high-current charging rates.In order to address this problem,an efficient fast charging–cooling scheduling method is urgently needed.In this study,a liquid cooling-based thermal management system equipped with mini-channels was designed for the fastcharging process of a lithium-ion battery module.A neural network-based regression model was proposed based on 81 sets of experimental data,which consisted of three sub-models and considered three outputs:maximum temperature,temperature standard deviation,and energy consumption.Each sub-model had a desirable testing accuracy(99.353%,97.332%,and 98.381%)after training.The regression model was employed to predict all three outputs among a full dataset,which combined different charging current rates(0.5C,1C,1.5C,2C,and 2.5C(1C=5 A))at three different charging stages,and a range of coolant rates(0.0006,0.0012,and 0.0018 kg·s^(-1)).An optimal charging–cooling schedule was selected from the predicted dataset and was validated by the experiments.The results indicated that the battery module’s state of charge value increased by 0.5 after 15 min,with an energy consumption lower than 0.02 J.The maximum temperature and temperature standard deviation could be controlled within 33.35 and 0.8C,respectively.The approach described herein can be used by the electric vehicles industry in real fast-charging conditions.Moreover,optimal fast charging-cooling schedule can be predicted based on the experimental data obtained,that in turn,can significantly improve the efficiency of the charging process design as well as control energy consumption during cooling.