Back Propagation Neural Network Model Research Articles

The reliability analysis and experiment verification of pressure spherical model for deep sea submersible based on data BP and machine learning technology

Spherical pressure-resistant shells, as a universal structural component of deep-sea submersibles, provide a safe and normal operating environment for personnel and internal equipment. In the paper it presented and optimized the BP neural network model based on a genetic algorithm (GA) accordingly, and the method and accuracy are validated through by a beam model. Simultaneously focusing on steel spherical shells, the study proposed a dataset that captures the influence of the primary dimension of the shell (radius-to-thickness ratio, R/t) on the critical pressure response. The genetic algorithm is employed to optimize the back propagation (BP) neural network model for predicting critical pressure. The structural reliability is adopted as a design criterion to determinate and optimize the geometric parameters and critical pressure of the spherical shell structure. Finally, an ultra-high-strength steel spherical model is designed, constructed and meanwhile collapse pressure tests are accomplished to verify the accuracy of the presented improved BP neural network model based on the computational reliability method. The results reveal that the machine learning optimization design method proposed in this paper can effectively enhance the accuracy of critical pressure predictions and the precision of reliability assessments for deep-sea spherical shells.

Quality Analysis and Shelf-Life Prediction of Antarctic Krill (Euphausia superba) Sauce Based on Kinetic Model and Back Propagation Neural Network Model

The study is aimed at determining how the quality of Antarctic krill (Euphausia superba) sauce (AkS) changed over time, including changes in color, moisture content, acid value (AV), peroxide value (POV), thiobarbituric acid reactive substance (TBARS), aerobic plate count, and sensory score. Quality variations and shelf life of AkS were estimated using kinetic model and back propagation (BP) neural network model. The results showed that sensory score, moisture content, and a∗ values of AkS declined as storage temperature increased at 4, 25, and 37°C. In addition, the L∗ values, b∗ values, AV, POV, and TBARS of AkS increased as storage duration increased, indicating that high storage temperature of the samples accelerated quality degradation. The primary reason for AkS degradation was the oxidation of proteins and lipids. The POV, TBARS, and total sensory evaluation rating exhibited a highly significant correlation, and therefore, POV and TBARS were selected as the indicators for the two models. The BP neural network outperformed the kinetic model in predicting quality changes over the whole storage period, with relative errors of less than 10%. In terms of shelf-life prediction, the BP neural network’s relative errors were 11.76% and 13.39% in POV and TBARS, respectively. POV and TBARS had experimental shelf lengths of 119 and 142 d, respectively. Compared with the kinetic model, the BP neural network model predicted the quality changes and shelf life of AkS with greater accuracy and stability. The findings offer fundamental insights and innovative concepts for the production of high-value Antarctic krill products, as well as the exploitation of Antarctic krill resources.

Real-time prediction method of carbon concentration in carburized steel based on a BP neural network

ABSTRACT Gears are key components of mechanical transmission systems. They need to be carburized and quenched to meet the requirements of internal toughness and external hardness, to obtain higher hardness and wear resistance. Heat treatment engineers can calculate carbon concentration distribution through the finite element analysis method, facing the challenge of high computational cost and high memory requirements. To solve this problem, a real-time prediction method of 1D and 2D carburizing concentration based on a Back Propagation (BP) neural network (BPNN) is proposed. First, carburizing experiments were conducted to verify the accuracy of the carburizing numerical model. Then, by establishing an accurate carburizing model, 37,800 and 177,147 1D and 2D carburizing training samples were generated using the finite element model (FEM) method, respectively. Finally, for 1D carburizing, the average relative error of the training model on the test set is 0.2911%. For 2D carburizing, the average relative error of the reconstructed BPNN model on the test set was 0.4381%, and a real-time performance evaluation was performed. The carbon concentration prediction time for each set of process parameters is only 0.12 s, nearly 175 times faster than FEM calculation, which meets the accuracy and real-time requirements for real-time prediction of carburizing carbon concentration.

Numerical Simulation and Deformation Prediction of Deep Pit Based on PSO-BP Neural Network Inversion of Soil Parameters

The finite element numerical simulation results of deep pit deformation are greatly influenced by soil layer parameters, which are crucial in determining the accuracy of deformation prediction results. This study employs the orthogonal experimental design to determine the combinations of various soil layer parameters in deep pits. Displacement values at specific measurement points were calculated using PLAXIS 3D under these varying parameter combinations to generate training samples. The nonlinear mapping ability of the Back Propagation (BP) neural network and Particle Swarm Optimization (PSO) were used for sample global optimization. Combining these with actual onsite measurements, we inversely calculate soil layer parameter values to update the input parameters for PLAXIS 3D. This allows us to conduct dynamic deformation prediction studies throughout the entire excavation process of deep pits. The results indicate that the use of the PSO-BP neural network for inverting soil layer parameters effectively enhances the convergence speed of the BP neural network model and avoids the issue of easily falling into local optimal solutions. The use of PLAXIS 3D to simulate the excavation process of the pit accurately reflects the dynamic changes in the displacement of the retaining structure, and the numerical simulation results show good agreement with the measured values. By updating the model parameters in real-time and calculating the pile displacement under different working conditions, the absolute errors between the measured and simulated values of pile top vertical displacement and pile body maximum horizontal displacement can be effectively reduced. This suggests that inverting soil layer parameters using measured values from working conditions is a feasible method for dynamically predicting the excavation process of the pit. The research results have some reference value for the selection of soil layer parameters in similar areas.

An intelligent model to decode students’ behavioral states in physical education using back propagation neural network and Hidden Markov Model

This paper highlights the need for intelligent analysis of students’ behavioral states in physical education tasks. The hand-ring inertial data is used to identify students’ motion sequence states. First, statistical feature extraction is performed based on the acceleration and angular velocity data collected from the bracelet. After completing the filtering and noise reduction of the data, we perform feature extraction by Back Propagation Neural Network (BPNN) and use the sliding window method for analysis. Finally, the classification capability of the model sequence is enhanced by the Hidden Markov Model (HMM). The experimental results indicate that the classification accuracy of student action sequences in physical education exceeds 96% after optimization by the HMM method. This provides intelligent means and new ideas for future student state recognition in physical education and teaching reform.

A Clinical-Information-Free Method for Early Diagnosis of Lung Cancer from the Patients with Pulmonary Nodules Based on Backpropagation Neural Network Model

Lung cancer is the main cause of cancer-related deaths worldwide. Due to lack of obvious clinical symptoms in the early stage of the lung cancer, it is hard to distinguish between malignancy and pulmonary nodules. Understanding the immune responses in the early stage of malignant lung cancer patients may provide new insights for diagnosis. Here, using high-through-put sequencing, we obtained the TCRβ repertoires in the peripheral blood of 100 patients with Stage I lung cancer and 99 patients with benign pulmonary nodules. Our analysis revealed that the usage frequencies of TRBV, TRBJ genes, and V-J pairs and TCR diversities indicated by D50s, Shannon indexes, Simpson indexes, and the frequencies of the largest TCR clone in the malignant samples were significantly different from those in the benign samples. Furthermore, reduced TCR diversities were correlated with the size of pulmonary nodules. Moreover, we built a backpropagation neural network model with no clinical information to identify lung cancer cases from patients with pulmonary nodules using 15 characteristic TCR clones. Based on the model, we have created a web server named “Lung Cancer Prediction” (LCP), which can be accessed at http://i.uestc.edu.cn/LCP/index.html.

Mid-to-long Term Wind Power Forecasting Based on ARIMA-BP Combined Model

Background: Increasing the accuracy of the output power forecasting for wind power is helpful to the improvement of the reliability of power dispatching. Objective: This study aimed to improve the forecasting accuracy of mid-to-long term wind power. Methods: A mid-to-long term wind power forecasting based on ARIMA-BP combined model was proposed. The Empirical Mode Decomposition (EMD) was used to decompose the historical wind power series and obtain the Intrinsic Mode Function (IMF) and residual components, thereby obtaining more regular components. Then, the optimum feature set was obtained based on the minimum Redundancy Maximum Relevance (mRAR) to improve the prediction accuracy for feature extraction. After that, the high-frequency components were predicted using the Back Propagation (BP) neural network model, while the low-frequency components were predicted using the Autoregressive Integrated Moving Average model (ARIMA). Finally, the predicted components obtained were superimposed to deduce the final mid- and long-term wind power prediction results. Results: An analysis was conducted according to the actual data from a typical wind farm. After comparison, it was found that, after empirical mode decomposition and feature extraction analysis, the error of the intelligent combination algorithm based on the ARIMA-BP combined model was smaller than that using only the BP neural network or only the ARIMA. Conclusion: By means of actual data analysis, the effectiveness of the method proposed by the study for mid- and long-term wind power prediction was verified.

Parametric study and neural network-based prediction for stress concentration factor of concrete-filled steel tubular T-joint

In this study, an extensive parametric investigation for the stress concentration factor (SCF) of concrete-filled steel tubular (CFST) T-joints was conducted based on a validated finite element (FE) model in which the weld profile, mesh, and contact simulation were specifically and precisely dealt with. Distribution of the SCF on the intersecting line was depicted. Effect of non-dimensional geometric parameters and the elastic modulus of concrete was discussed in detail. Results show that the SCF is monotonically distributed on the intersecting line in the chord side, and the maximum SCF probably located at the crown position. There is the possibility that the maximum SCF of the brace occurs at a position neither the crown nor the saddle. Reducing the length of chord or promoting the elastic modulus of concrete lowers the SCF at most positions, but has no effect on the SCF at chord saddle. A back-propagation neural network (BPNN) model was established, and was trained based on 724 FE results of the SCF from the parametric study. The trained BPNN model predicts the SCF of CFST T-joints with a low error of 14.5%, and it improves the accuracy by 62.0% compared to the current formulae.

Tight carbonate reservoir evaluation case study based on neural network assisted fracture identification and analytic hierarchy process

AbstractComprehensive evaluation of reservoirs is an important link in gas reservoir exploration and development. The evaluation of tight carbonate reservoirs often focuses on the characteristics of porosity and permeability, ignoring the important factor of fractures, also the quantitative evaluation of reservoirs is relatively few. It is difficult to identify fractures and evaluate the reservoir factors qualitatively and quantitatively. Herein, the sedimentary microfacies and microporosity of the tight carbonate reservoir of the Ma55 submember in the eastern Sulige area are comprehensively studied by casting thin section, rock physical property, and capillary pressure test data. The backpropagation (BP) neural network algorithm is used to identify and predict fractures. Finally, through the analytic hierarchy process, the above reservoir influencing factors are modeled and quantitatively analyzed for reservoir evaluation. The results show that the highest probability of fracture development in the central and northwest areas of the study area can reach 0.92. The accuracy of the BP neural network model in identifying cracks can reach 80%, which is reliable and effective compared with the conventional logging identification method. Reservoirs can be classified into four types according to their quality. The synthetic weights of porosity, permeability and fracture development probability are 0.2, 0.2 and 0.216 respectively, which are the three most important evaluation parameters. This study improves the accuracy of fracture identification and prediction of tight reservoirs in comprehensive reservoir evaluation, which provides guidance and scheme for more detailed exploration and development of tight carbonate reservoirs.

Performance degradation trend prediction of proton exchange membrane fuel cell based on GA-TCN

To improve the prediction accuracy of the performance degradation trend of proton exchange membrane fuel cell (PEMFC), this paper proposes a temporal convolutional network (TCN) model based on genetic algorithm (GA) optimization to predict the performance degradation trend of PEMFC. Firstly, variational mode decomposition and wavelet threshold denoising algorithms are used to denoise the original data. Then the hyperparameters of the TCN model are optimized by GA, and the GA-TCN model for predicting the performance degradation trend of PEMFC is constructed. Finally, this paper uses the PEMFC stack degradation experimental dataset disclosed in the IEEE PHM 2014 Data Challenge to verify, and compares the proposed model with the backpropagation neural networks model, the long short-term memory model and the classical TCN model. The results show that the proposed method has the highest performance degradation trend prediction accuracy. In particular, when the training dataset accounts for 30%, i.e. the training samples are small, the root mean square error, mean absolute error and mean absolute percentage error of the GA-TCN model are 0.004 726, 0.003 119 and 9.62%, respectively, which are 14.48%, 20.05% and 2.42% lower than that of the classical TCN model. Consequently, this methodology can forecast the degradation trend of PEMFC with high accuracy.

Application of Artificial Neural Network Methods to Anatolian Plate Earthquake Magnitude and Location Prediction

Same-region earthquakes usually have a pattern that is difficult to identify clearly. Therefore, time series analysis methods have been proposed for earthquake prediction. Our work attempts to predict three earthquake parameters in the Anatolian Peninsula using pure artificial neural network methods. An optimized BP-NN model and optimally hyper-parameterized LSTM Model have been designed to predict earthquake magnitude, latitude, and longitude. The two models are compared among each other and with previous works for their prediction performances, using four well-accepted metrics: Mean Squared Error, Mean Absolute Error, Median Absolute Error, and Standard Deviation. The time, depth, sun, and moon distances to Earth were identified as the most contributing factors in earthquake occurrence through analysis by five different feature extraction algorithms. The date harmed the prediction accuracy. The LSTM model outperformed the BP-NN Model in magnitude prediction with 0.062 MSE. Latitude predictions of both methods were satisfactory and close. However, BP-NN had lower error rates in latitude prediction. However, longitude prediction errors were large in both models. Therefore, our designs were not successful in predicting the exact location of the earthquakes. However, multi-variate, stacked LSTM models are promising in predicting Anatolian Peninsula earthquake magnitudes, but future work is necessary for location and timing predictions.

Multi-Model Comprehensive Inversion of Surface Soil Moisture from Landsat Images Based on Machine Learning Algorithms

Soil moisture plays an important role in maintaining ecosystem stability and sustainable development, especially for the upper reaches of the Yellow River region. Therefore, accurately and conveniently monitoring soil moisture has become the focus of scholars. This study combines three machine learning algorithms: random forest (RF), support vector machine (SVM), and back propagation neural network (BPNN)—with the traditional monitoring of soil moisture using remote sensing indices to construct a more accurate soil moisture inversion model. To enhance the accuracy of the soil moisture inversion model, 27 environmental variables were screened and grouped, including vegetation index, salinity index, and surface temperature, to determine the optimal combination of variables. The results show that screening the optimal independent variables in the Xijitan landslide distribution area lowered the root mean square error (RMSE) of the RF model by 16.95%. Of the constructed models, the combined model shows the best applicability, with the highest R2 of 0.916 and the lowest RMSE of 0.877% with the test dataset; the further research shows that the BPNN model achieved higher overall accuracy than the other two individual models, with the test set R2 being 0.809 and the RMSE 0.875%. The results of this study can provide a theoretical reference for the effective use of Landsat satellite data to monitor the spatial and temporal distribution of and change in soil water content on the two sides of the upper Yellow River basin under vegetation cover.

Visualisation of moisture content distribution maps and classification of freshness level of loquats

Moisture content (MC) and freshness undergo evident changes during the storage of loquat fruits and thus are important indices of loquat fruit quality. Conventional methods for assessing MC and freshness are typically destructive and time-consuming, and cannot provide distribution maps of MC in the whole loquats. In this study, the use of hyperspectral imaging (HSI) was investigated for rapid and non-destructive prediction of MC in loquats, and for the visualisation of their MC distribution maps. Furthermore, freshness level discrimination models for loquats at different storage times were established using partial least squares discriminant analysis (PLS-DA), multinomial logistic regression (MLogR), and back-propagation neural network (BPNN) classification models. MC was well-predicted, with the best residual predictive deviation (RPD) of 2.40. Using the best prediction model, the distribution maps of MC in loquats were visualised. Additionally, the competitive adaptive reweighted sampling (CARS)-BPNN model obtained superior discrimination performance, with 95.56 % and 93.33 % accuracy for the calibration and prediction sets. Our results show that HSI is a promising fast and non-destructive method for determining MC and freshness of loquats.

Analysis and Prediction of Urban Surface Transformation Based on Small Baseline Subset Interferometric Synthetic Aperture Radar and Sparrow Search Algorithm-Convolutional Neural Network-Long Short-Term Memory Model.

With the acceleration of urbanisation, urban areas are subject to the combined effects of the accumulation of various natural factors, such as changes in temperature leading to the thermal expansion or contraction of surface materials (rock, soil, etc.) and changes in precipitation and humidity leading to an increase in the self-weight of soil due to the infiltration of water along the cracks or pores in the ground. Therefore, the subsidence of urban areas has now become a serious geological disaster phenomenon. However, the use of traditional neural network prediction models has limitations when examining the causal relationships between time series surface deformation data and multiple influencing factors and when applying multiple influencing factors for predictive analyses. To this end, Sentinel-1A data from March 2017 to February 2023 were used as the data source in this paper, based on time series deformation data acquired using the small baseline subset interferometric synthetic aperture radar (SBAS-InSAR) technique. A sparrow search algorithm-convolutional neural network-long short-term memory (SSA-CNN-LSTM) neural network prediction model was built. The six factors of temperature, humidity, precipitation, and ground temperature at three different depths below the surface (5 cm, 10 cm, and 15 cm) were taken as the input of the model, and the surface deformation data were taken as the output of the neural network model. The correlation between the spatial and temporal evolution characteristics of the ground subsidence in urban areas and various influencing factors was analysed using grey correlation analysis, which proved that these six factors contribute to some extent to the deformation of the urban surface. The main urban area of Hohhot City, Inner Mongolia Autonomous Region, was used as the study area. In order to verify the efficacy of this neural network prediction model, the prediction effects of the multilayer perceptron (MLP), backpropagation (BP), and SSA-CNN-LSTM models were compared and analysed, with the values of the correlation coefficients of the feature points of A1, B1, and C1 being in the range of 0.92, 0.83, and 0.93, respectively. The results show that compared with the traditional MLP and BP neural network models, the SSA-CNN-LSTM model achieves a higher performance in predicting time series surface deformation data in urban areas, which provides new ideas and methods for this area of research.

LEVERAGING DEEP LEARNING FOR ACCURATE RICE LEAF DISEASE RECOGNITION

: Imagine a Rice Field, seemingly healthy, yet harboring invisible foes. This paper delves into the world of crop disease detection for the Tan spot, Leaf blight, Sheath decay, Bacterial leaf rot, and False smut Dataset consists of 3546 images. The image segmentation is done by BIRCH clustering followed by GMM clustering; various texture features have been extracted through this. The classification is done using existing models such as VGG16, VGG19, RESNET50, and INCEPTION V3. After comparing various models and validation, the observation was made that VGG19 performs well compared to other models. The overall accuracy obtained for the VGG19 is 95.88.%. The achieved accuracy surpasses that of conventional backpropagation neural network models, indicating significant advancements in crop disease diagnosis. This study introduces a novel approach that opens avenues for future research in the field of deep learning for crop disease diagnosis.

Modeling of the shape of Taylor cone in EHD printing based on backpropagation neural network and genetic algorithm

Electrohydrodynamic (EHD) printing has been recognized as a promising additive manufacturing technology with superior pattern resolution and economic viability. The shape of Taylor cone is deemed a pivotal element for the amplification of deposition efficiency, and the maintenance of consistent operational steadiness in EHD printing. The correlations between diverse operating parameters and the shape of Taylor cone are presently not well investigated. In this paper, modeling of relationships between operating parameters and the shape of Taylor cone was conducted with a backpropagation neural network (BPNN) and a genetic algorithm optimized backpropagation (GA-BP) neural network. Taylor cone semi-vertical angle and the meniscus height were employed as two indexes to characterize the shape of Taylor cone. The prediction accuracies of BPNN model and GA-BP model were 92.79 % and 95.46 %, respectively. The GA-BP model demonstrated higher precision in forecasting the shape of Taylor cone. A predictive framework for the shape of Taylor cone was proposed, which provided a practical tool for process optimization in EHD printing.

Positioning error estimation models for horizontal-distributed PUU parallel mechanism

Abstract In this paper, two models were proposed to estimate the positioning error of a 3-translational prismatic-universal-universal parallel kinematic mechanism. The two models were a kinematic error model (KEM) and a backpropagation neural network (BPNN) model, respectively. The KEM was constructed by incorporating three translational joint errors into the ideal kinematic model to describe the errors that occur during machining or assembly. Additionally, a sensitivity analysis was presented for each error parameter. The BPNN model was constructed to establish the relationship between the position of the end effector, the posture of each link, and the positioning error of the end effector using a neural network approach. Moreover, a hybrid method was proposed to decrease the final estimated residual error. The average errors of the KEM and BPNN model were 35% and 15% of the original error, respectively. The hybrid model reduced the final average error to less than 10% of its original value.

Sensitivity Analysis of Influencing Factors and Two-Stage Prediction of Frost Resistance of Active-Admixture Recycled Concrete Based on Grey Theory-BPNN.

Sensitivity analysis of influencing factors on frost resistance is carried out in this paper, and a two-stage neural network model based on grey theory and Back Propagation Neural Networks (BPNNs) is established for the sake of predicting the frost resistance of active-admixture recycled concrete quickly and accurately. Firstly, the influence degree of cement, water, sand, natural aggregate, recycled aggregate, mineral powder, fly ash, fiber and air-entraining agent on the frost resistance of active-admixture recycled-aggregate concrete was analyzed based on the grey system theory, and the primary and secondary relationships of various factors were effectively distinguished. Then, the input layer of the model was determined as cement, water, sand, recycled aggregate and air-entraining agent, and the output layer was the relative dynamic elastic modulus. A total of 120 datasets were collected from the experimental data of another author, and the relative dynamic elastic modulus was predicted using the two-stage BPNN prediction model proposed in this paper and compared with the BPNN prediction results. The results show that the proposed two-stage BPNN model, after removing less-sensitive parameters from the input layer, has better prediction accuracy and shorter run time than the BPNN model.

Quality detection and variety classification of pecan seeds using hyperspectral imaging technology combined with machine learning

Pecan (Carya illinoinensis K.), a well-known dried seed and woody oil tree, faces challenges in its industry due to complex quality assessment methods and confusing varieties. These challenges have seriously hampered the development of a large-scale pecan deep processing industry. This work aimed to apply hyperspectral imaging technology (HSI) combined with machine learning to evaluate the quality of pecan seeds and perform variety classification. The samples of this work were composed of 19 varieties of pecan seeds, with 30 seeds per variety. After spectral preprocessing, spectral features were extracted from the spectral profiles using feature extraction methods. Back-propagation neural network models and partial least squares models were established to predict the contents of crude fat and moisture in pecan seeds. Predictions of the best models gave good results with R2-score of 0.887 for the crude fat model and 0.950 for the moisture model. Additionally, support vector machine models were developed to identify pecan varieties. The model achieved good results in 19 pecan varieties identification with accuracy of 0.965. In conclusion, the combination of HSI and machine learning could be an effective tool in improving the pecan industry and providing sustainable and efficient methods in the production of pecan seeds.

Prediction of Tensile Properties in Inconel 625 Superalloy Fabricated by Wire Arc Additive Manufacturing Using Improved Artificial Neural Network

This study exhibits the possibility of using an artificial neural network (ANN) to model the mechanical behavior of wire arc additive manufacturing (WAAM) for Inconel 625. For this reason, tensile tests of Inconel 625 superalloy as-built (AB) samples and samples after heat treatment at 1200 °C (HT-1200) by WAAM were performed. For the HT-1200 samples, the yield stress decreased, and the elongation increased significantly due to grain refinement and the formation of annealed twins. A new hybrid model combining a swarm intelligence optimization algorithm with a back propagation neural network (BPNN) was developed to simulate the flow behavior of the superalloy. Compared with other hybrid BPNN models that have been reported, the proposed BPNN model is in better agreement with the experimental data and provides a better description of the flow stress of the Inconel 625 superalloy. The excellent predictive ability of the model may be attributed to the optimization of the weights and thresholds of the BPNN, which obtains the optimal global solution in the search space more efficiently.