Back Propagation Neural Network Analysis Research Articles

Studies on Submersible Short-Circuit Blowing Based on Orthogonal Experiments, Back Propagation Neural Network Prediction, and Pearson Correlation Analysis

Short-circuit blowing is a crucial technical approach for ensuring the rapid surfacing of submersibles. In order to investigate the law, L18(37) orthogonal experiments based on a proportional short-circuit blowing model test bench were conducted. Subsequently, a Back Propagation Neural Network (BPNN) and Pearson correlation analysis were employed to train the experimental data; further examination of the correlation between individual factors and blowing served as an enhancement to the orthogonal experiments. It has been proved that both multi-factor combinations and personal factors, including blowing duration, sea tank back pressure, gas blowing pressure from the cylinder group, and sea valve flowing area, exert significant influence with Pearson correlation coefficients of 0.6535, 0.8105, 0.5569, and 0.5373, respectively. Notably, the F-ratio for blowing duration exceeds the critical value of 3.24. The statistical evaluation metrics for the BPNN ranged from 10−1 to 10−12, with relative errors below 3%, and achieving a prediction accuracy rate of 100%. Based on these findings, a robust predictive methodology for submersible short-circuit blowing has been established along with recommendations for engineering design and operational strategies that highlight its advantages as well as its initial condition settings.

Classification of Pesticide Residues in Sorghum Based on Hyperspectral and Gradient Boosting Decision Trees

ABSTRACTTo address the challenges posed by chemical methods for detecting pesticide residues in sorghum, such as complicated sample preparation and prolonged detection periods, this study presents a rapid and nondestructive detection approach based on hyperspectral imaging (HSI) technology. A group of sorghum without pesticide residues and three groups uniformly sprayed with pesticides were used in this study. Firstly, support vector machine (SVM) classification models were built using spectral data preprocessed with Savitzky–Golay (SG), discrete wavelet transform (DWT), and standard normal variate (SNV) methods, respectively, and SNV was determined to be the best preprocessing method. Secondly, the gradient boosting decision tree (GBDT) algorithm, principal component analysis (PCA), and the successive projections algorithm (SPA) were respectively used to extract feature wavelengths. Pesticide residue identification models based on full and feature wavelengths were then respectively established using backpropagation neural network (BPNN), SVM, and partial least squares discriminant analysis (PLS‐DA). The results show that the BPNN model developed using the feature wavelengths obtained from GBDT was the best for identification of pesticide residues, with an accuracy of 97.8% for both the training and testing sets. Finally, visualization of pesticide residue species in sorghum was achieved using the optimal model. This study demonstrates that utilizing HSI in conjunction with the GBDT‐BPNN model is an effective, rapid, and nondestructive method for identifying pesticide residues in sorghum.

A parametric optimization framework for fin-and-tube heat exchangers based on response surface methodology and artificial intelligence

This study presents a numerical investigation and provides a novel framework for the rapid parametric optimization of attack angles (α) and aspect ratios (e) in plate fin-and-tube heat exchangers (PFTHEs) featuring elliptical tubes. The methodology incorporates three-dimensional numerical simulations, response surface methodology (RSM), backpropagation neural network (BPNN), and Entropy-VIKOR. The study focuses on a PFTHE with a single elliptical tube, varying the attack angles (α) from −30° to 0° in 5° increments and the aspect ratios (e) from 0.4 to 1.0 in a step of 0.1. Air velocities (u) range from 0.4 to 3.2 m/s with an interval of 0.2 m/s, corresponding Reynolds number from 200 to 2200. The results indicate that using elliptical tubes reduces Δp and enhances heat transfer capacity, with the most significant improvements at an e of 0.4. Exponential relationships between dimensionless parameters Nu or f and Re are established for different range of α at specific e values. The synergistic effect of e, α, and Re or u on PFTHE performance is confirmed through RSM and BPNN analysis. Their R2 values are no less than 0.896 and reach a maximum of 0.998, indicating excellent predictive accuracy. Utilizing Entropy-VIKOR method, the framework identifies a suitable design solution with e = 0.6, α = -20°, and u = 3.2 m/s. This solution exhibits a 5 % increase in h and a 58 % reduction in Δp relative to the corresponding circular tube. The study provides specific expressions relating design variables to thermal–hydraulic behavior and indicates carefully selecting the α and e can improve the overall performance. In summary, these findings have significant implications for reducing optimization design period for PFTHEs with small tube diameters.

Global Navigation Satellite System Receiver Positioning in Harsh Environments via Clock Bias Prediction by Empirical Mode Decomposition and Back Propagation Neural Network Method.

This paper proposes a novel method to improve the clock bias short-term prediction accuracy of navigation receivers then solve the problem of low positioning accuracy when the satellite signal quality deteriorates. Considering that the clock bias of a navigation receiver is equivalent to a virtual satellite, the predicted value of clock bias is used to assist navigation receivers in positioning. Consequently, a combined prediction method for navigation receiver clock bias based on Empirical Mode Decomposition (EMD) and Back Propagation Neural Network (BPNN) analysis theory is demonstrated. In view of systematic errors and random errors in the clock bias data from navigation receivers, the EMD method is used to decompose the clock bias data; then, the BPNN prediction method is used to establish a high-precision clock bias prediction model; finally, based on the clock bias prediction value, the three-dimensional positioning of the navigation receiver is realized by expanding the observation equation. The experimental results show that the proposed model is suitable for clock bias time series prediction and providing three-dimensional positioning information meets the requirements of navigation application in the harsh environment of only three satellites.

The Flavor Characteristics, Antioxidant Capability, and Storage Year Discrimination Based on Backpropagation Neural Network of Organic Green Tea (Camellia sinensis) during Long-Term Storage.

The storage period of tea is a major factor affecting tea quality. However, the effect of storage years on the non-volatile major functional components and quality of green tea remains largely unknown. In this study, a comparative analysis of organic green teas with varying storage years (1-16 years) was conducted by quantifying 47 functional components, using electronic tongue and chromatic aberration technology, alongside an evaluation of antioxidative capacity. The results indicated a significant negative correlation between the storage years and levels of tea polyphenols, total amino acids, soluble sugars, two phenolic acids, four flavonols, three tea pigments, umami amino acids, and sweet amino acids. The multivariate statistical analysis revealed that 10 functional components were identified as effective in distinguishing organic green teas with different storage years. Electronic tongue technology categorized organic green teas with different storage years into three classes. The backpropagation neural network (BPNN) analysis demonstrated that the classification predictive ability of the model based on the electronic tongue was superior to the one based on color difference values and 10 functional components. The combined analysis of antioxidative activity and functional components suggested that organic green teas with shorter storage periods exhibited stronger abilities to suppress superoxide anion radicals and hydroxyl radicals and reduce iron ions due to the higher content of eight components. Long-term-stored organic green teas, with a higher content of substances like L-serine and theabrownins, demonstrated stronger antioxidative capabilities in clearing both lipid-soluble and water-soluble free radicals. Therefore, this study provided a theoretical basis for the quality assessment of green tea and prediction of green tea storage periods.

Thermal-vibration correlation study for high-temperature superconducting maglev intelligent monitoring based on back propagation neural network analysis

The internal temperature rise inside the high-temperature superconducting (HTS) superconductor (SC) resulting from irregular magnetic field (MF) above the permanent magnet guideway is a major factor contributing to the decline of levitation performance. Real-time monitoring of the temperature rise inside YBCO SC is an important issue for the safe operation of the maglev train systems. However, the existing temperature rise testing method involves destructive intrusion less or more, easily affected by strong MF, occupying limited space and sensors prone to detachment. This paper innovatively proposes a non-contact internal temperature rise testing method combining artificial intelligence (AI) methods. Vibration is common signal of a maglev train system, which inspires to establish a fundamental thermal-dynamic levitation force synchronous testing device for YBCO SC. Then, a set of temperature rise-vibration dataset exposed to different alternating MF frequencies is created. The wavelet transform is chosen to extract the frequency band energy of vibration, and the backpropagation neural network is used to identify the corresponding temperature rise. The recognition accuracy can reach over 99.9%, which firstly proves the effectiveness of AI algorithms in the thermal-vibration correlation analysis for the HTS maglev system. The results can provide the foundation reference for the intelligent monitoring and fault diagnosis of thermal-dynamics stabilities of HTS maglev train in the future.

Estimating vehicle braking distance over wet and rutted pavement surface through back-propagation neural network

Due to repeated load cycles and climate impacts, road pavement deteriorates. One of the primary causes of pavement degradation is the formation of rutting under the wheel path on the road surface. Rutting's impact on vehicle performance, especially on rainy days (since the rain would fill the rutting depression), leads to longer braking distances compared to dry conditions. This study developed a MATLAB-based model for calculating vehicle braking distance on wet asphalt pavement affected by rutting, using dynamic skid resistances generated through Back-Propagation Neural Network (BPNN) analysis. This study addresses the worst-case scenario in which rutting is filled with water, then calculates the required vehicle braking distance under various Water Film Thickness (WFT) conditions. The developed model can perform these evaluations for different operational conditions across various input ranges, such as precipitation intensity, number of lanes, lane width, cross slope, average texture depth, rutting depths, and accumulated WFT. As an outcome, the vehicle braking distance can be estimated as a function of Rutting Depth (RD) within a known vehicle speed interval. After validating the proposed model against existing approaches from the literature, several sensitivity analyses are conducted to assess the impact of influencing parameters on the results. Moreover, the study examines the relationship between AASHTO braking distance requirements and the RD threshold levels adopted by several highway agencies. Furthermore, this model is also applicable to real-world case studies, enabling the calculation of vehicle braking distances with varying RDs in the presence of various WFTs on the pavement surface.

Cost Estimation of Manufacturing Enterprises based on BP Neural Network and Big Data Analysis

The manufacturing industry is the pillar industry of modern industry, and the cost estimation of manufacturing enterprises is an important management means of the manufacturing industry. Aiming at the cost estimation problem of manufacturing enterprises, this research proposes a cost estimation method based on Back Propagation (Back Propagation) neural network and big data analysis. In the process, the Lambda architecture was used to construct the big data analysis architecture of manufacturing enterprises, the K-means clustering algorithm was introduced for data clustering, and then the genetic algorithm was combined with the Back Propagation neural network to estimate the cost. In the estimation accuracy test, the accuracy of the research method can reach 94.7% after 240 iterations; in the calculation time test, the calculation time of the research method is 403 Ks when the data size is 500 Gb in a large-scale data set; in the call data volume test, the call data volume of the research method is 164 Kb when the research method is carried out to the seventh step in the small-scale data set; when the application analysis is carried out, the research method completes accurate cost estimation for 9 target parts. This research method has good model performance and calculation accuracy, and can effectively estimate manufacturing enterprises’ costs.

Harnessing climate variables for predicting PV power output: A backpropagation neural network analysis in a subtropical climate region

This study examines the influence of weather conditions on the electricity production of photovoltaic (PV) panels in the Conghua district, a region in Southern China distinguished by its subtropical mountainous climate. The area's high humidity levels, high temperatures, and substantial rainfall often lead to predominantly cloudy skies, which alter solar irradiation patterns as a result. Utilizing the EnergyPlus simulation tool, the power generation of PV panels on rooftop application was simulated. A structured classification method is adopted to sort out weather conditions into categories of sunny, cloudy, and rainy. Through examining the influence of these varied weather scenarios on power generation, unique factors affecting electricity outputs are identified. As a result, this study illuminates the relationships between potential weather variables and PV power generation across each weather category. Subsequently, a back propagation neural network (BPNN) model is utilized to explore the relationship between weather categories and PV generation. The high coefficient of determination value (R2 > 0.95) from the comparison of simulated and predicted PV generation validates the model accuracy. This research enables the development of updated dataset and enriches the prediction of rooftop PV power generation. It equips designers and owners with a reliable, user-friendly, and real-time tool for PV power generation forecasting, and provides a working platform for PV performance evaluation during the design phase of PV projects. The insights derived from the BPNN model could guide the optimal design and implementation of PV systems in similar climates. Furthermore, these findings advance sustainable energy solutions in regions affected by complex weather patterns.

Study on the Prediction of Low-Index Coal and Gas Outburst Based on PSO-SVM

Low-index coal and gas outburst (LI-CGO) is difficult to predict, which seriously threatens the efficient mining of coal. To predict the LI-CGO, the Support Vector Machine (SVM) algorithm was used in this study. The Particle Swarm Optimization (PSO) algorithm was used to optimize the parameters of the SVM algorithm. The results show that based on the training sets and test set in this study, the prediction accuracy of SVM is higher than that of Back Propagation Neural Network and Distance Discriminant Analysis. The prediction accuracy of the SVM model trained by the training set T2 with LI-CGO cases is higher than that of the SVM model trained by the training set T1 without LI-CGO cases. The prediction accuracy gets better when the SVM model is trained by the training set T3, made by adding the data of the other two coal mines (EH and SH) to the training set T2, that only contains the data of XP and PJ. Furthermore, the PSO-SVM model achieves a better predictive effect than the SVM model, with an accuracy rate of 90%. The research results can provide a method reference for the prediction of LI-CGO.

Determination of monophenolase activity based on backpropagation neural network analysis of three-dimensional fluorescence spectroscopy

Tyrosinase is pivotal for melanin formation. Measuring monophenolase activity is of great importance for both fundamental research and industrial applications. For the first time, a backpropagation (BP) artificial neural network with three-dimensional fluorescence spectroscopy was applied for the real-time determination of tyrosinase monophenolase activity. Principal component analysis (PCA) was utilized for the dimension reduction of three-dimensional fluorescence data. The four principal components served as inputs for the neural network. Network parameters were optimized using a genetic algorithm (GA). BP learning algorithm was applied to train the network model to determine tyrosine levels in a binary mixture containing tyrosine and L-DOPA without any chemical separation. The time course of tyrosine consumption by monophenolase was determined to calculate the initial velocity of the enzymatic reaction. The limit of detection of the monophenolase assay was 0.0615 U·mL−1. This combined strategy of PCA, GAs, and BP artificial neural networks for three-dimensional fluorescence spectroscopy was efficient for the real-time and in-situ determination of monophenolase activity in a cascade reaction.

Evaluation of smart city construction and optimization of city brand model under neural networks

The study aims to avoid the phenomenon that thousand cities seem the same in the construction of smart cities and the efforts of all walks of life are jointed to implement the construction of smart cities and the creation of city brands. First, the basic theory of smart city construction is introduced. Second, the restricting and promoting factors influencing smart city construction and development are analyzes, and the evaluation system of smart city development is established. Then, the model for smart city construction and development based on the neural network is implemented. Finally, some domestic cities are selected as the dataset to build a training model, and the city brand optimization strategy is proposed. On this basis, an evaluation system for smart city development based on the intelligent neural network and Grey relational analysis Back Propagation Neural Network (GRA-BPNN) is obtained. The entropy weight method (EWM), grey relational analysis (GRA) method and the evaluation method based on Technique for Order Preference by Similarity to an Ideal Solutionv (TOPSIS) are regarded as the members of the control group, and the results of different methods for evaluating the development of smart cities are compared. The results show that the modeling of smart city construction based on neural networks can help to implement an evaluation model for smart city construction and development, and it can help evaluate smart city construction and development accurately. And then the corresponding strategies are proposed to speed up the construction of smart cities with its local characteristics, and then city brands are built. Compared with other evaluation algorithms, the performance of the algorithm proposed is better and more stable and the evaluation results are more reasonable than those of the others, which can prove that the evaluation algorithm is feasible and scientific. This study provides a new idea for the application of deep learning to smart city construction and city brand building.

Feature Ranking and Differential Evolution for Feature Selection in Brushless DC Motor Fault Diagnosis

A fault diagnosis system with the ability to recognize many different faults obviously has a certain complexity. Therefore, improving the performance of similar systems has attracted much research interest. This article proposes a system of feature ranking and differential evolution for feature selection in BLDC fault diagnosis. First, this study used the Hilbert–Huang transform (HHT) to extract the features of four different types of brushless DC motor Hall signal. When there is a fault, the symmetry of the Hall signal will be influenced. Second, we used feature selection based on a distance discriminant (FSDD) to calculate the feature factors which base on the category separability of features to select the features which have a positive correlation with the types. The features were entered sequentially into the two supervised classifiers: backpropagation neural network (BPNN) and linear discriminant analysis (LDA), and the identification results were then evaluated. The feature input for the classifier was derived from the FSDD, and then we optimized the feature rank using differential evolution (DE). Finally, the results were verified from the BLDC motor’s operating environment simulation with the same features by adding appropriate signal-to-noise ratio magnitudes. The identification system obtained an accuracy rate of 96% when there were 14 features. Additionally, the experimental results show that the proposed system has a robust anti-noise ability, and the accuracy rate is 92.04%, even when 20 dB of white Gaussian noise is added to the signal. Moreover, compared with the systems established from the discrete wavelet transform (DWT) and a variety of classifiers, our proposed system has a higher accuracy with fewer features.

Exploring the relationship between the dielectric properties and viability of human normal hepatic tissues from 10 Hz to 100 MHz based on grey relational analysis and BP neural network

Liver is an important parenchyma organ, and its tissue viability plays an important role in liver transplantation and liver ischemic injury assessment. Dielectric property is a useful biophysical feature that provides insights into the structure and composition of biological tissues. This work aims to establish the relationship between the dielectric properties and viability of human normal hepatic tissues and explore the possibility of evaluating tissue viability by using dielectric properties. First, data on dielectric properties and tissue viability (including cell morphology and enzyme indicators) were collected from human liver tissues at 0.25–24 h after isolation. Grey relational analysis was conducted to select dielectric property and tissue viability indices that were highly correlated with prolonged ex vivo time as the inputs and outputs, respectively, of back-propagation (BP) neural network analysis. Finally, a BP neural network was developed with the Levenberg–Marquardt algorithm to explore the possibility of using dielectric properties as the basis for tissue viability evaluation. Results showed that the mean relative error for prediction was 2.40%, indicating that the model showed potential in forecasting liver tissue viability by applying dielectric properties.

Exploration of Social Benefits for Tourism Performing Arts Industrialization in Culture-Tourism Integration Based on Deep Learning and Artificial Intelligence Technology.

As a product of the tourism performing arts industry in culture–tourism integration development, to develop a featured culture–tourism town is a new trend for tourism development in the new era. To analyze the social benefit of the culture–tourism industry, in this study, an artificial intelligence model for social benefit evaluation is constructed based on backpropagation (BP) neural network and fuzzy comprehensive analysis, with Yiyang Town taken as an example. The criterion layer in the model includes three indexes (life benefit G1, environmental benefit G2, and economic benefit G3), and the index layer contains 11 indexes (H1–H11). The weight values of cultural inheritance and protection, ecological environment improvement, and commercial economy development to the social benefit of the town are 0.522, 0.570, and 0.424, respectively. For G1, 41.20% is excellent; for G2, 39.5% is excellent; and for G3, 40.5% is good. In general, 30.76% of the total social benefit is excellent, with 37.69% being good, 21.48% being qualified, and 10.07% being unqualified. It is inferred that the total social benefit level of Yiyang Town is good according to the constructed model. Therefore, the culture inheritance and protection, the ecological environment improvement, and the commercial economy development are the key evaluation factors of social benefit.

Modeling study of sandstone permeability under true triaxial stress based on backpropagation neural network, genetic programming, and multiple regression analysis

Permeability evolution of sandstone is of great significance in the development of tight sandstone gas reservoirs. Traditional laboratory tests have the disadvantages of high cost and long testing time. Therefore, the present study employed use artificial intelligence systems, i.e., backpropagation neural network (BPNN), genetic programming (GP), and multiple regression analysis to construct prediction models of sandstone permeability based on the coupling effect of true triaxial stress field and pore pressure. The results showed that the permeability prediction obtained from the systems fit well with the experimental data, and evidenced that permeability increased with pore pressure and decreased with increase in principal stress. Sensitivity analysis showed that the pore pressure has the greatest influence on sandstone permeability under different true triaxial stress. The effect of anisotropic principal stress on permeability exhibited σ1 > σ2 > σ3 under fixed pore pressure. Further assessment based on a combination of five evaluation indexes showed that the prediction accuracy of the BPNN model was better.

Rapid identification of cervical adenocarcinoma and cervical squamous cell carcinoma tissue based on Raman spectroscopy combined with multiple machine learning algorithms.

Cervical cancer has a long latency, and early screening greatly reduces mortality. In this study, cervical adenocarcinoma and cervical squamous cell carcinoma tissue data were collected by Raman spectroscopy, and then, the adaptive iteratively reweighted penalized least squares (airPLS) algorithm and Vancouver Raman algorithm (VRA) were used to subtract the background of the collected data. The following five feature extraction algorithms were applied: partial least squares (PLS), principal component analysis (PCA), kernel principal component analysis (KPCA), isometric feature mapping (isomap) and locally linear embedding (LLE). The k-nearest neighbour (KNN), extreme learning machine (ELM), decision tree (DT), backpropagation neural network (BP), genetic optimization backpropagation neural network (GA-BP) and linear discriminant analysis (LDA) classification models were then established through the features extracted by different feature extraction algorithms. In total, 30 types of classification models were established in this experiment. This research includes eight good models, airPLS-PLS-KNN, airPLS-PLS-ELM, airPLS-PLS-GA-BP, airPLS-PLS-BP, airPLS-PLS-LDA, airPLS-PCA-KNN, airPLS-PCA-LDA, and VRA-PLS-KNN, whose diagnostic accuracy was 96.3 %, 95.56 %, 95.06 %, 94.07 %, 92.59 %, 85.19 %, 85.19 % and 85.19 %, respectively. The experimental results showed that the model established in this article is simple to operate and highly accurate and has a good reference value for the rapid screening of cervical cancer.

Back propagation neural network analysis for the detection of explosives based on tagged neutron

A system for detecting explosives in walls based on the tagged neutron method was established. An associated particle neutron generator ING-27 developed by the All-Russian Research Institute of Automatics (VNIIA) is used as the neutron source, two yttrium lutetium silicate (LYSO) detectors are used for the γ detector, and a silicon detector is used as the α detector. The α-γ coincidence spectra of 300 g ammonium nitrate and TNT samples placed behind 10 cm concrete wall were measured. A neural network algorithm was applied to analyze the data. Through fine selection of time windows of α-γ coincidence spectra, the counts of full-energy peak of N, C and O elements and their proportions are selected as the input eigenvectors. The neural network is trained by the experimental data obtained by this system. Through training, eight groups of test data consisting of 32 γ-ray spectra were identified, and a 98.7% of correct detection rate was achieved.

Application of terahertz spectrum and interval partial least squares method in the identification of genetically modified soybeans.

Genetically modified soybeans are the world's most important genetically modified agricultural product. At present, the traditional methods for identifying genetically modified and non-transgenic soybeans are time-consuming, costly, and complicated to operate, which cannot meet the needs of practical applications. Therefore, it is necessary to discover a fast and accurate method for identifying transgenic soybeans. Terahertz (THz) time domain spectra were collected in sequence from 225 transgenic and non-transgenic soybean samples. Fourier transform was used to convert the terahertz time domain spectrum into a THz frequency domain spectrum with a frequency range of 0.1-2.5THz. Firstly, the interval partial least squares (iPLS) method was used to remove interference spectral bands and select appropriate spectral intervals. Secondly, 168 samples were selected as the calibration set. Discriminant partial least squares (DPLS), Grid Search support vector machine (Grid Search-SVM) and principal component analysis back propagation neural network (PCA-BPNN) were used to establish a qualitative identification model. Afterwards, 57 test set samples were predicted. By comparing the experimental results, it was found that iPLS could effectively screen and remove the interference THz band, which was more helpful to improve the efficiency and accuracy of the identification model. After the iPLS and mean center pre-treatment technology, the Grid Search-SVM identification model had the best identification effect, with a total accuracy rate of 98.25% (transgenic identification rate was 96.15%, non-transgenic identification rate was 100%). This study shows that after selecting spectra from iPLS, THz spectroscopy combined with chemometrics can more accurately, quickly, and efficiently identify transgenic and non-transgenic soybeans.

Multiple Kinds of Pesticides Detection Based on Back-Propagation Neural Network Analysis of Fluorescence Spectra

Fluorescence spectroscopy attracted more and more attention in pesticide residue detection field because of its advantages of non-destructive, non-contact, high speed and no requirement of complex pre-process procedure. However, given that the concentration of the pesticide detected via fluorescence spectroscopy is calculated in accordance with the Beer-Lambert law, this method can only be used to detect samples containing a single kind of pesticide or several kinds of pesticides with completely different fluorescence which is not in accordance with practical cases. In this article, to overcome this disadvantage, back-propagation (BP) neural network algorithm was introduced to detect multiple kinds of pesticides via fluorescence spectroscopy. The results from four kinds of pesticides which are usually used for fruits and vegetables indicated the effectiveness of BP neural network algorithm.