Back Propagation Neuron Network Research Articles

Application of Poisson’s ratio structures and decoupling algorithm for 3D force sensing

Flexible tactile electronic devices are extensively used in the fields of robotics, medical detection, and human-computer interaction. Monitoring contact parameters, including force magnitude, direction, and contact location, is particularly vital for skin-like tactile sensing devices. Herein, a 3D force sensor is designed based on porous structure with deliberately designed Poisson’s ratios. A genetic algorithm (GA) optimized back propagation neuronal network (BPNN) model is proposed to support the 3D force decoupling, which can greatly improve the decoupling accuracy. The introduction of the GA-BPNN significantly enhances decoupling accuracy compared to the initial neural network. Micro-porous structures with varied Poisson’s ratios are embedded into the sensing unit to achieve better sensibility. Significantly, this study underscores that the decoupling accuracy of the force components along the Z-axis can be further improved by substituting the solid unit with a designed porous structure unit featuring a specific Poisson’s ratio in an arrayed 3D force sensor.

Inversion of soil heavy metals in metal tailings area based on different spectral transformation and modeling methods

The exploitation of mineral resources has seriously polluted the environment around mines, notably in terms of heavy metal contamination of tailings pond soil. Hyperspectral remote sensing, as opposed to conventional on-site sampling and laboratory analysis, offers a potent tool for effective monitoring the content of soil heavy metals. Therefore, we investigated the inversion models of heavy metal content in metal tailings area based on measured hyperspectral and multispectral data. Hyperspectral and its transformation, as well as the simulated Landsat8-OLI multispectral were used for model inversion respectively. Stepwise Multiple Linear Regression (SMLR), Partial Least Squares Regression (PLSR) and Back Propagation Neuron Network (BPNN) were established to study the spectral inversion of eight heavy metals (Cu, Cd, Cr, Ni, Pb, Zn, As, and Hg). The direct inversion models were established on the basis of correlation analysis and the adjust coefficient of determination (Adjust_R2) and Root Mean Square Error (RMSE) were used for model evaluation. Then the best combination of spectral transformation and inversion model were explored. The model inversion results suggested that: (1) Hyperspectral transformation can generally improve the model accuracy, especially the second derivative spectral, based on which the training Adjust_R2 of Hg SMLR and PLSR models are as high as 0.795 and 0.802. (2) The BP neural network inversion based on the denoised hyperspectrum demonstrate that both the training and testing Adjust_R2 of Cd, Ni and Hg models are all greater than 0.5, indicating good applicability in practical extrapolation. (3) Both the training and testing Adjust_R2 of Cu and Hg PLSR models based on simulated R_Landsat8-OLI multispectral are greater than 0.5, and Hg has lower RMSE and lager Adjust_R2 with training and testing Adjust_R2 values of 0.833 and 0.553 respectively. (4) Multispectral remote sensing detection and mapping of Hg contamination were realized by the optimal simulation model of Hg. Hence, it is feasible to simulate the multispectral with hyperspectral data for investigating heavy metal contamination.

Micro-expression recognition based on EEG signals

Research into micro-expression recognition is increasingly motivated by its potential applications to mental-health diagnosis and policing. Thus far, micro-expression recognition research and applications have been based on computer vision processing technology, with notable limitations. For instance, changes in ambient lighting or in the head posture and facial occlusion can all affect micro-expression recognition. The high temporal resolution of EEG technology captures brain activity associated with micro-expressions, and hence objectively identifies micro-expressions from a neurophysiological perspective. Herein, we report a real-time supervision and emotional expression suppression (SEES) experimental paradigm. We collected micro-expression video and EEG data from 68 subjects under positive emotion induction. Thereafter, the EEG frequency-band power, statistical features, autoregressive (AR) model parameters, and wavelet entropy were extracted as features. We ranked all 128 channels based on the micro-expression recognition task classification accuracy using all the features extracted for each channel. Based on the channel ranking, different sizes of channel combinations were chosen. Subsequently, the optimal subset of these features from different combinations was selected using a sequential backward selection (SBS) algorithm and fed into different classifiers, including a support vector machine (SVM), K-nearest neighbor (KNN), gradient-boosted decision tree (GBDT), and back-propagation (BP) neuron network. The different classifiers achieve promising accuracy for micro-expression recognition, with GBDT exhibiting the highest accuracy of 85.96 % when 11 channels were selected. This demonstrates the feasibility of recognizing micro-expressions based on EEG signals.

Compressive strength prediction of hydrothermally solidified clay with different machine learning techniques

Hydrothermal solidification technology can transform waste clay into high-strength building materials with lower energy consumption. Few machine learning methods have been applied to strength prediction of hydrothermally solidified materials, whose mechanical properties differ from those of conventional cement-based materials by virtue of the production method. In this study, six machine learning methods, random forest (RF), gated recurrent unit (GRU), k-nearest neighbors (KNN), back-propagation neuron network (BPNN), extreme gradient boosting (XGBoost) and gaussian process regression (GPR), were used to develop the compressive strength prediction models of hydrothermally solidified clay. These methods can efficiently predict the compressive strength without time-consuming and costly tests and guide the optimization of the mix proportions, thus facilitating the application of hydrothermally solidified clay. For this purpose, 140 experimental data obtained from hydrothermally solidified clay with different Ca(OH)2 content, dry density, moisture content, curing time and curing temperature were used to proposed the prediction models. In the modeling process, these five effective variable parameters were considered as the modeling input parameters. The performance of these models was compared by correlation coefficient (R2), mean absolute error (MAE) and root mean square error (RMSE). The validity of these models was confirmed by the K-fold cross-validation method, and the influence weight of each input parameter on the prediction results was characterized by sensitivity analysis. The results demonstrated that the GPR model provided the best prediction with an R2 value of 0.989. In the sensitivity analysis, curing time had the greatest influence on the predicted results, followed by dry density, Ca(OH)2 content, curing temperature and water content. Comprehensively considering the cost, energy consumption and compressive strength, the applicable mix proportion of hydrothermally solidified clay includes Ca(OH)2 content (not less than 16.64%), moisture content (10%), curing temperature (not less than 87.35 °C), curing time (18 h) and dry density (1.9 g/cm3).

Multidisciplinary Structural Optimization of Low-Pyroshock Separation Nuts for Aerospace

This paper presents a multidisciplinary structural optimization approach to design separation nuts with low-level pyroshock responses for aerospace. A parametric modeling method is developed to automatically construct a multibody finite element model for separation nuts according to the predefined structural parameters. A new explicit dynamics analysis workflow tailored to separation nuts is proposed to simulate the pyroshock environment efficiently. Based on a backpropagation neuron network technique, certain surrogate models are established to describe the equivalence relationship between the structural parameters and the multidisciplinary structural responses including the pyroshock, the bending stiffness, and the structural weight. Finally, a pyroshock minimization problem with the constraints of structural weight and bending stiffness is solved by using a genetic algorithm, resulting in an optimized low-pyroshock separation nut design. The comparison between the optimized structure and a reference structure shows the effectiveness of the proposed approach.

Linear Active Disturbance Rejection Control-Based Diagonal Recurrent Neural Network for Radar Position Servo Systems with Dead Zone and Friction

This paper proposes a control scheme for the radar position servo system facing dead zone and friction nonlinearities. The controller consists of the linear active disturbance rejection controller (LADRC) and diagonal recurrent neural network (DRNN). The LADRC is designed to estimate in real time and compensate for the disturbance with vast matched and mismatched uncertainties, including the internal dead zone and friction nonlinearities and external noise disturbance. The DRNN is introduced to optimize the parameters in the linear state error feedback (LSEF) of the LADRC in real time and estimate the model information, namely Jacobian information, of the plant on-line. In addition, considering the Cauchy distribution, an adaptive tracking differentiator (ATD) is designed in order to manage the contradiction between filtering performance and tracking speed, which is introduced to the LADRC. Another novel idea is that the back propagation neuron network (BPNN) is also introduced to tune the parameters of the LADRC, just as in the DRNN, and the comparison results show that the DRNN is more suitable for high precision control due to its feedback structure compared with the static BPNN. Moreover, the regular controller performances and robust performance of the proposed control approach are verified based on the radar position servo system by MATLAB simulations.

Pressure Drop Prediction of Crude Oil Pipeline Based on PSO-BP Neural Network

Pipeline transportation of crude oil has great advantages over traditional oil transmission methods, in terms of economic and environmental protection. The main costs in the oilfield surface system are the fuel costs for heating the crude oil during transportation and the electricity costs for the pumping units. In the northeast of China, where winter temperatures are extremely low and the oil has a high freezing point and high viscosity, higher temperatures, and pressures are required to transport crude oil. With machine learning widely used in many industries and achieving better results, the digital management of oil pipelines has stored a large amount of production and operation data, which has laid the foundation for the research of oil pipeline process calculation using machine learning methods. In this paper, a crude oil pressure drop calculation of an oil pipeline in Northeast China is carried out based on a neural network. For pipeline pressure drop calculation, the back propagation neural network (BP) pressure drop calculation model and particle swarm optimization for back propagation neuron network (PSO-BP) pressure drop calculation model are established. Two models were used to calculate and compare the measured data, and the average absolute error of the PSO-BP model was the smallest, which was 0.015%. Compared with the BP model, the average relative error is reduced by 13.16%. Therefore, The PSO-BP pressure drop calculation model has high accuracy and is of practical significance for predicting pipeline pressure drop.

Experimental and neural networks analysis on elevated-temperature mechanical properties of structural steels

This study presents the possibility of using artificial neural network (ANN) for describing and predicting the high temperature mechanical behavior of structural steel. For this reason, a series of steady-state tensile tests were performed on two representative structural steels Q345B and Q460GJ in temperature range of 20 ℃ to 800 ℃, in the forms of steel bars with diameters of 8, 10 and 12 mm. The high-temperature mechanical properties were evaluated and compared, revealing the necessity of a general and reliable prediction model on the elevated-temperature mechanical properties of different structural steels. Furthermore, the experimental results could be well predicted by using the Ramberg-Osgood model only in a limited temperature range. Therefore, the application of back-propagation neuron network (BPNN) was proposed to predict the yield stress and ultimate stress. In order to model flow property, a long short-term memory recurrent neural network (LSTM-RNN) was first adopted in strength of mechanics. Two preprocessing methodologies including one-hot encoding and polynomial feature function were used in the models. The satisfactory agreements indicate that the trained BPNN and LSTM-RNN models are efficient and accurate in predicting the mechanical properties of structural steels.

Rapid identification of lamb freshness grades using visible and near-infrared spectroscopy (Vis-NIR)

In the meat industry, it is essential to monitor and identify meat freshness grades due to its impact on the safety of human diets. This study aimed to identify premium, sub-fresh, and spoiled lamb samples using visible and near-infrared (Vis-NIR) hyperspectral imaging (HSI) in the range of 400–1000 nm coupled with chemometrics methods. The two-dimensional correlation spectroscopy (2D-CS) was utilized to select effective wavelengths for simplifying the model and increasing the calculation speed. The capabilities of the four models including partial least squares discriminant analysis (PLS-DA), soft independent modelling of class analogy (SIMCA), back propagation neuron network (BP), decision tree (DT) and random forest (RF) were compared to select the best identification model. The results showed that the RF model generated excellent performance, the accuracies of the training and test sets were 93% and 91%, respectively. In summary, this study showed that it was feasible to rapidly and non-destructively identify and evaluate the freshness grades of lamb using Vis-NIR.

Pattern Recognition of Quality Control Chart of Multi-variety and Small-batch Production Mode Based on MC-GA Optimized BP

To solve the problem of product quality control in multi-variety and small-batch production mode, a quality control chart recognition model of MC-GA optimized BP was established by using Monte Carlo method to simulate the product quality data characteristics of different varieties and combining the data processing capability of neural network and the adaptive global optimization search capability of genetic algorithm; then the MC(Monte Carlo method)-GA(Genetic Algorithm) optimized BP(Back Propagation Neuron Network) was compared with the traditional MC-BP is compared and analyzed in terms of accuracy, time and error in quality control graph pattern recognition, and it is concluded that the former is better in quality control graph pattern recognition. It can be seen that the MC-GA optimized BP model is more effective in identifying the quality status of products in the multi-variety and small-batch production mode, and can accurately predict the stability of the manufacturing process, so as to enhance the competitiveness of enterprises.

Hybrid fuzzy neural network versus backpropagation neural network: An application to predict the Ibex-35 index stock

The use of neural networks has been extended in all areas of knowledge due to the good results being obtained in the resolution of the different problems posed. The prediction of prices in general, and stock market prices in particular, represents one of the main objectives of the use of neural networks in finance. This paper presents the analysis of the efficiency of the hybrid fuzzy neural network against a backpropagation type neural network in the price prediction of the Spanish stock exchange index (IBEX-35). The paper is divided into two parts. In the first part, the main characteristics of neural networks such as hybrid fuzzy and backpropagation, their structures and learning rules are presented. In the second part, the prediction of the IBEX-35 stock exchange index with these networks is analyzed, measuring the efficiency of both as a function of the prediction errors committed. For this purpose, both networks have been constructed with the same inputs and for the same sample period. The results obtained suggest that the Hybrid fuzzy neuronal network is much more efficient than the widespread backpropagation neuronal network for the sample analysed.

Artificial Neural Network Rainfall-Discharge Model Assessment Under Rating Curve Uncertainty and Monthly Discharge Volume Predictions

The accuracy of rainfall-discharge volume model predictions depends on the model design and uncertainty of the available stage-discharge measurements used to fit the rating curve, which converts a time-series of recorded stage into discharge. In general, the rating curve uncertainty is the product of several combined sources. Over Algerian rivers, the extrapolation of the rating curve beyond the gauging range is the main source of this uncertainty. This study, therefore, represents a quantitative approach to reflect rigorously the impact of the rating curve uncertainty on the improvement of monthly discharge volume prediction quality by the artificial neural network (ANN) rainfall-discharge model. The rating curve uncertainty of the Fer a cheval hydrometric station in the Mazafran watershed is performed within Bayesian analysis for stationary rating curves using the BaRatin method. This allows as to build a new time series of discharge in order to assess an ANN rainfall-discharge model. To do that, Levenberg–Marquardt back propagation neuronal network has been applied over 1972-2012 time-period, for five hydrometric stations in the Algiers Coastal Basin. The model inputs were constructed in different ways, during the algorithm development, such as precipitation, antecedent precipitation with different monthly lag times and antecedent monthly discharge volume. The results indicate that training/validation of ANN rainfall-discharge volume model is widely affected by the streamflow datasets uncertainty. A large proportion of model prediction errors are significantly improved when considering the rating curve uncertainty.

Prototipo de emulación de movimiento humano mediante una plataforma bípeda humanoide

En esta investigación se desarrolló una plataforma bípeda humanoide, cuya función es emular la marcha humana partiendo del principio de mantener el mínimo consumo energético minimizando el movimiento vertical de la pelvis. La plataforma se realizó utilizando herramientas de diseño y simulación como AutoCAD para la estructura mecánica, Solid-edge para el modelamiento tridimensional y Phyton para simular el movimiento de la plataforma. Estas herramientas suministraron las características físicas exactas de la plataforma, permitiendo generar el modelo cinemático. El modelo cinemático de la marcha se desarrolló utilizando MATLAB, así: a partir de una versión modificada de los parámetros de Denavit-Hartenberg se establece la cinemática directa, y la cinemática inversa se determina a partir de una red neuronal de tipo feedforward-backpropagation. Se obtuvo una plataforma bípeda humanoide con diez grados de libertad, capaz de realizar flexión - extensión en la cadera y rodilla, abducción - aducción en la cadera, flexión plantar-dorsal y pronación - supinación en el tobillo. La generación de trayectorias se obtiene en MATLAB mediante el ingreso del número de pasos, la longitud del paso, la altura del paso y una variable de tiempo que separará cada una de las secuencias o subpasos de la trayectoria. Como resultado se obtuvieron las trayectorias articulares en cadera, rodilla y tobillo en los tres planos, las cuales son acordes con la marcha humana y se evidenció con el robot bípedo humanoide.Con esta investigación se espera establecer nuevos modelos para emular la marcha humana normal y patológica. Asimismo, esta es desarrollada por el grupo de investigación DIGITI de la Universidad Distrital Francisco José de Caldas.

A New Model of ROP Prediction for Drilling Engineering with Data Mining Technology

In drilling engineering, predicting the rate of penetration (ROP) is a prominent factor affecting economic and engineering decisions during well planning, which is particularly important for deep wells and ultra-deep wells. Different methods can be utilized in the optimization of drilling. However, traditional methods were always proposed with the use of empirical correlations and simply predictive algorithms, which lacked convenience and accuracy. To improve ROP predicting capabilities, this paper proposes one predicting method called the combined model of AHP (Analytic Hierarchy Process) and BPNN (Back Propagation Neuron Network) based on data mining. The model was built using UCS (Uniaxial Compressive Strength), bit size, bit type, drillability coefficient, gross hours drilled, WOB (Weight on Bit), RPM (Revolutions Per Minute), drilling mud density and AV (Apparent Viscosity) as the input parameters, and obtained the weights of these factors using AHP, then used Neural Network to predict drilling rate. This model improves the rate of convergence and the reliability of results. The validity of this method has been demonstrated with data from an existing field in north-west China. Furthermore, the model can be used in post-well analysis to identify areas where potential drilling performance was not achieved, and help in identifying improvements for future projects.

Recognition system of leaf images based on neuronal network

In forest variety registration, visual traits of the plants appearance are widely used to discern different tree species. The new recognition system of leaf image strategy which based on neural network established to administrate a hierarchical list of leaf images, some sorts of edge detection can be performed to identify the individual tokens of every image and the frame of the leaf can be got to differentiate the tree species. An approach based on back-propagation neuronal network is proposed and the programming language for the implementation is also given by using Java. The numerical simulations results have shown that the proposed leaf strategy is effective and feasible.

Study on the Predictive Model for Shear Strength in Laser Welding Stainless Steel

A predictive model is presented for the prediction of shear strength in laser welding AISI304 stainless steel. Welding experiments conducted using a pulsed Nd:YAG laser machine while the laser welding parameters and their levels have been arranged according to design of experiments of Taguchi method. The tensile tests are performed after welding and the measurements of tensile strength are further calculated for shear strength. The data can be analyzed using the principles of Taguchi method for determining the optimal laser welding parameters and for investigating the most significant laser welding parameter on shear strength. Furthermore, the results are treated as the training and recalling patterns for constructing a predictive model using back-propagation neuron network to predict shear strength for the range of laser welding operation tested. It is indicated that welding speed is the most significant affecting parameters on shear strength. In addition, an increase in welding speed causes a decrease in shear strength is found. An average error 5.75％for shear strength can be found by comparing the experimental results obtained from conducting verification tests with the predicting values obtained from the established predictive model. It shows that the predictive model is capable of good predicting behavior of laser welding AISI304 stainless steel.