Back Propagation Neural Network Algorithm Research Articles

Simulation of Optimization Decision Model for Low-carbon Economy Schemes Based on Machine Learning Algorithms

Low-carbon economy (LCE) is a shift from high-carbon intensive industries and activities to low-emission or zero-emission industries and activities, which requires a lot of investment in new technologies and infrastructure, as well as changes in policies and regulations to promote sustainability and reduce emissions. In this article, the back propagation neural network (BPNN) algorithm in machine learning (ML) is applied to the construction of low-carbon economic scheme optimization decision model, and the performance of the model is simulated and tested. The results show that BPNN algorithm has higher accuracy and reliability in predicting the development level of LCE, which is 23.8% higher than Support Vector Machine (SVM) algorithm. Through the comparative study of BPNN algorithm and SVM algorithm in forecasting the development level of LCE, the advantages of BPNN algorithm in forecasting accuracy and application value are verified. BPNN algorithm can better adapt to and deal with the complex data patterns in the field of LCE, improve the accuracy and reliability of prediction, and provide strong technical support and guarantee for the growth of LCE.

Laser-induced breakdown spectroscopy as a method for millimeter-scale inspection of surface flatness

A non-contact method for millimeter-scale inspection of material surface flatness via Laser-Induced Breakdown Spectroscopy (LIBS) is investigated experimentally. The experiment is performed using a planished surface of an alloy steel sample to simulate its various flatness, ranging from 0 to 4.4 mm, by adjusting the laser focal plane to the surface distance with a step length of 0.2 mm. It is found that LIBS measurements are successful in inspecting the flatness differences among these simulated cases, implying that the method investigated here is feasible. It is also found that, for achieving the inspection of surface flatness within such a wide range, when univariate analysis is applied, a piecewise calibration model must be constructed. This is due to the complex dependence of plasma formation conditions on the surface flatness, which inevitably complicates the inspection procedure. To solve the problem, a multivariate analysis with the help of Back-Propagation Neural Network (BPNN) algorithms is applied to further construct the calibration model. By detailed analysis of the model performance, we demonstrate that a unified calibration model can be well established based on BPNN algorithms for unambiguous millimeter-scale range inspection of surface flatness with a resolution of about 0.2 mm.

Improved BP Neural Network Algorithm for Predicting Structural Parameters of Mirrors

Addressing the nonlinear correlations between input variables and output responses, in addition to the time-consuming nature of finite element analysis in mirror design, this study introduces an enhanced back-propagation (BP) neural network (BR-TLDBO-BPNN) employing Bayesian regularization and an optimized dung beetle algorithm. This novel approach facilitates rapid and efficient parameter estimations, significantly reducing the computational overhead. Utilizing an integrated analysis platform, the study obtained training and test samples, and the BR-TLDBO-BPNN model is used to predict the reflector’s mass and root mean square (RMS). The optimization mathematical model is built, and the nonlinear planning function (fmincon) is utilized to solve the problem and find an ideal set of structural parameters. The outcomes demonstrate that the prediction model is accurate enough to predict the mirror characteristics to optimize the mirror structural parameters. Empirical validation demonstrates that the proposed model achieves an over 99% accuracy in predicting mirror characteristics against finite element simulations. As a result, the BR-TLDBO-BPNN algorithm studied in this article not only broadens the application scope of neural networks, but also provides a new practical technique for engineering design.

Identification of key biomarkers for early warning of diabetic retinopathy using BP neural network algorithm and hierarchical clustering analysis

Diabetic retinopathy is one of the most common microangiopathy in diabetes, essentially caused by abnormal blood glucose metabolism resulting from insufficient insulin secretion or reduced insulin activity. Epidemiological survey results show that about one third of diabetes patients have signs of diabetic retinopathy, and another third may suffer from serious retinopathy that threatens vision. However, the pathogenesis of diabetic retinopathy is still unclear, and there is no systematic method to detect the onset of the disease and effectively predict its occurrence. In this study, we used medical detection data from diabetic retinopathy patients to determine key biomarkers that induce disease onset through back propagation neural network algorithm and hierarchical clustering analysis, ultimately obtaining early warning signals of the disease. The key markers that induce diabetic retinopathy have been detected, which can also be used to explore the induction mechanism of disease occurrence and deliver strong warning signal before disease occurrence. We found that multiple clinical indicators that form key markers, such as glycated hemoglobin, serum uric acid, alanine aminotransferase are closely related to the occurrence of the disease. They respectively induced disease from the aspects of the individual lipid metabolism, cell oxidation reduction, bone metabolism and bone resorption and cell function of blood coagulation. The key markers that induce diabetic retinopathy complications do not act independently, but form a complete module to coordinate and work together before the onset of the disease, and transmit a strong warning signal. The key markers detected by this algorithm are more sensitive and effective in the early warning of disease. Hence, a new method related to key markers is proposed for the study of diabetic microvascular lesions. In clinical prediction and diagnosis, doctors can use key markers to give early warning of individual diseases and make early intervention.

Determinants influencing ecological footprints of China; An empirical evidence from factor analysis

China has advanced significantly in the last decades, but ecological concerns raise questions about the nation's long-term affluence. The hunt for possible determinants in explaining the quality of the environment (ecological footprint) is still in progress. Therefore, the current study attempts to investigate the influencing factors from various explanatory variables (eco-innovation, economic complexity, renewable energy consumption, economic growth, natural resource rent, foreign direct investment, the energy intensity level of primary energy, economic development, Public-private partnerships investment in energy, electric power consumption, and population growth) through factor analysis augmented with Back Propagation Neural Network (BPNN) approach. The annual data from 1995 to 2021 for China is retrieved for empirical analysis. The data set comprises the factors classified into training, validation, and testing divisions with a ratio of 2:1:1. The outcomes revealed that extracted factors are robust in explaining China's ecological footprint variation. Outcomes of the BPNN algorithm suggest that renewable energy consumption, eco-innovation, economic complexity, natural resource rent, and electric power consumption are the most useful determinants in explaining the ecological footprints of China. The study furnishes the possible policy implications of the research.

Freeway merging trajectory prediction for automated vehicles using naturalistic driving data

Due to differences in speed and the complex interaction between merging and through-lane vehicles at freeway merging sections, crashes involving both human drivers and automated vehicles (AVs) persist. To assist AVs to predict the intentions of surrounding vehicles for further dynamic motion planning, researchers have focused on developing trajectory prediction algorithms. Few studies, however, have developed merging trajectory prediction models using naturalistic driving data in China, making it urgent to put it on the agenda for AVs’ safety and efficiency at freeway merging sections. Based on merging periods extracted from the Shanghai Naturalistic Driving Study (SH-NDS), this study compares merging behavior on freeways with through-lane speed limits of 80 km/h, 100 km/h, and 120 km/h using analysis of variance (ANOVA). Merging trajectory prediction algorithms for these three speed limit cases are trained and tested using backpropagation neural network (BPNN) and long short-term memory neural network (LSTM NN) approaches. Results show 1) significant differences among the three cases in all merging behavior variables except for longitudinal gap; and that 2) the BPNN algorithm for merging trajectory prediction demonstrates superior performance compared to the LSTM NN. Two major contributions to the safe operation of AVs are provided: 1) the developed algorithms can be integrated into AV systems to accurately predict real-time desired trajectories of nearby merging vehicles in uncongested traffic conditions, and assist ongoing motion planning strategies for AVs; and 2) the algorithms can be incorporated in simulation tests for AV safety evaluation involving freeway merging sections.

Seafloor topography refinement from multisource data using genetic algorithm—backpropagation neural network

SUMMARY During the inversion of seafloor topography (ST) using the backpropagation neural network (BPNN), the random selection of parameters may decrease the accuracy. To address this issue and achieve a more efficient global search, this paper introduces a genetic algorithm-backpropagation (GA-BP) neural network. Benefiting from the global search and parallel computing capabilities of the GA, this study refines the ST of the South China Sea using multisource gravity data. The results indicate that the GA-BP model, with a root mean square (RMS) value of 126.0 m concerning ship-measured water depths. It is noteworthy that when dealing with regions characterized by sparse survey line distributions, the GA-BP neural network stronger robustness compared to BPNN, showing less sensitivity to the distribution of survey data. Furthermore, the paper explores the influence of different data pre-processing methods on the neural network inversion of sea depths. This research introduces an optimization algorithm that reduces instability during BPNN initialization, resulting in a more accurate prediction of ST.

Lithium-Ion Battery Capacity Prediction with GA-Optimized CNN, RNN, and BP

Over the last 20 years, lithium-ion batteries have become widely used in many fields due to their advantages such as ease of use and low cost. However, there are concerns about the lifetime and reliability of these batteries. These concerns can be addressed by obtaining accurate capacity and health information. This paper proposes a method to predict the capacity of lithium-ion batteries with high accuracy. Four key features were extracted from current and voltage data obtained during charge and discharge cycles. To enhance prediction accuracy, the Pearson correlation coefficient between these features and battery capacities was analyzed and eliminations were made for some batteries. Using a genetic algorithm (GA), the parameter optimization of Convolutional Neural Network (CNN), Backpropagation (BP), and Recurrent Neural Network (RNN) algorithms was performed. The parameters that provide the best performance were determined in a shorter time using GA, which includes natural selection and genetic processes instead of a trial-and-error method. The study employed five metrics—Mean Square Error (MSE), Root Mean Square Error (RMSE), Normalized Root Mean Square Error (NRMSE), Mean Absolute Error (MAE), and Squared Correlation (R2)—to evaluate prediction accuracy. Predictions based on NASA experimental data were compared with the existing literature, demonstrating superior accuracy. Using 100 training data, 68 data predictions were made with a Root Mean Square Error (RMSE) of 0.1176%. This error rate represents an accuracy level 2.5 times higher than similarly accurate studies in the literature.

Retraction: The deformation monitoring of foundation pit by back propagation neural network and genetic algorithm and its application in geotechnical engineering

Retraction: The deformation monitoring of foundation pit by back propagation neural network and genetic algorithm and its application in geotechnical engineering

Prediction of Drilling Efficiency for Rotary Drilling Rig Based on an Improved Back Propagation Neural Network Algorithm

Accurately predicting the drilling efficiency of rotary drilling is the key to achieving intelligent construction. The current types of principle analysis (based on traditional interactive experimental methods) and efficiency prediction (based on simulation models) cannot meet the requirements needed for the efficient, real-time, and accurate drilling efficiency predictions of rotary drilling rigs. Therefore, we adopted a method based on machine learning to predict drilling efficiency. The extremely complex rock fragmentation process in drilling conditions also brings challenges to predicting drilling efficiency. Therefore, this article went through a combination of mechanism and data analysis to conduct correlation analysis and to clarify the drilling characteristic parameters that are highly correlated with drilling efficiency, and it then used them as inputs for machine learning models. We propose a rotary drilling rig drilling efficiency prediction model based on the GA-BP neural network to construct an accurate and efficient drilling efficiency prediction model. Compared with traditional BP neural networks, it utilizes the global optimization ability of a genetic algorithm to obtain the initial weights and thresholds of a BP neural network in order to avoid the defect of ordinary BP neural networks, i.e., that they easily fall into local optimal solutions during the training process. The average prediction accuracy of the GA-BP neural network is 93.6%, which is 3.1% higher than the traditional BP neural network.

Research of electrical impedance tomography based on multilayer artificial neural network optimized by Hadamard product for human-chest models

Electrical impedance tomography (EIT) is a non-radiation, non-invasive visual diagnostic technique. In order to improve the imaging resolution and the removing artifacts capability of the reconstruction algorithms for electrical impedance imaging in human-chest models, the HMANN algorithm was proposed using the Hadamard product to optimize multilayer artificial neural networks (MANN). The reconstructed images of the HMANN algorithm were compared with those of the generalized vector sampled pattern matching (GVSPM) algorithm, truncated singular value decomposition (TSVD) algorithm, backpropagation (BP) neural network algorithm, and traditional MANN algorithm. The simulation results showed that the correlation coefficient of the reconstructed images obtained by the HMANN algorithm was increased by 17.30% in the circular cross-section models compared with the MANN algorithm. It was increased by 13.98% in the lung cross-section models. In the lung cross-section models, some of the correlation coefficients obtained by the HMANN algorithm would decrease. Nevertheless, the HMANN algorithm retained the image information of the MANN algorithm in all models, and the HMANN algorithm had fewer artifacts in the reconstructed images. The distinguishability between the objects and the background was better compared with the traditional MANN algorithm. The algorithm could improve the correlation coefficient of the reconstructed images, and effectively remove the artifacts, which provides a new direction to effectively improve the quality of the reconstructed images for EIT.

Alignment Prediction of Air Spring Vibration Isolation System for Marine Shafting Based on Genetic Algorithm-Back Propagation Neural Network.

Shafting alignment plays an important role in the marine propulsion system, which affects the safety and stability of ship operation. Air spring vibration isolation systems (ASVISs) for marine shafting can not only reduce mechanical noise but also help control alignment state by actively adjusting air spring pressures. Alignment prediction is the first and a key step in the alignment control of ASVISs. However, in large-scale ASVISs, due to factors such as strong interference and raft deformation, alignment prediction faces problems such as alignment measurement sensors failure and difficulty in establishing a mathematical model. To address this problem, a data model for predicting alignment state is developed based on a back propagation (BP) neural network, fully taking advantage of its self-learning and self-adaption abilities. The proposed model exploits the collected data in the ASVIS instead of the alignment measurement data to calculate the alignment state, providing another alignment prediction approach. Then, in order to solve the local optimum issue of BP neural network, we introduce the genetic algorithm (GA) to optimize the weights and thresholds of the BP neural network, and an improved GA-BP model is designed. The GA-BP model can leverage the advantages of the global search capability of GA as well as the BP neural network's fast convergence in local search. Finally, we conduct experiments on a real ASVIS and evaluate the prediction models using different criteria. The experimental results show that the proposed prediction model with the GA-BP neural network can accurately predict the alignment state, with a mean-square error (MSE) of 0.0114. And compared to the BP neural network, the GA-BP neural network reduces the MSE by approximately 74%.

A method for calculating the time‐varying equivalent circuit parameters of large‐capacity synchronous condenser considering field mutual leakage reactance

AbstractAccurate calculation of equivalent circuit parameters is a prerequisite for accurately calculating the large‐capacity synchronous condenser parameter model. Due to its special transient operating conditions, the high transient magnetic saturation effect during operation causes the non‐linearity and time‐varying of the equivalent circuit parameters. The field mutual leakage reactance is the critical factor affecting the field current, and the time‐varying of the equivalent circuit parameters is closely related to the field current. However, the existing equivalent circuit parameter calculation methods considering field leakage reactance cannot achieve time‐varying parameters. A calculation method of time‐varying equivalent circuit parameters based on a back propagation neural network algorithm is proposed, which solves the calculation problem of time‐varying equivalent circuit parameters considering field mutual leakage reactance. Then, a 300MVar condenser is taken as the research object, and the proposed method is used to simulate the different operating conditions of the condenser and verified by the finite element method and experiment. The results show that the method improves the calculation accuracy of the equivalent circuit parameter model, reduces the calculation time, and applies to different operating conditions.

Structural optimization of PVDF cellular resonators for energy-harvesting enhancement based on backpropagation neural network and NSGA-II algorithm

Metamaterials with honeycomb structures have garnered significant attention in energy-harvesting applications. In this study, we present three conventional cantilever resonators with honeycomb substrates featuring positive, zero, and negative Poisson’s ratios (PPR, ZPR, and NPR). Polyvinylidene fluoride consisting electrode layers is attached to the substrate surfaces. A process parameter optimization method of honeycomb structure design was proposed, in which finite element simulation for data generation, backpropagation neural network for data prediction, and nondominated sorting genetic algorithm II (NSGA-II) for data optimization. This method aims to determine the optimum geometric parameters of the honeycomb structures, which addresses general cantilever resonators, where the first structural vibration modes are typically of interest and have to match specific target eigenfrequencies for engineering applications. Finite element simulations show that the peak voltage of PPR, ZPR, and NPR resonators at 75 Hz is increased by 12 %, 22 %, and 32 %, respectively, due to optimization. The optimized structures are fabricated and measured to validate the numerical model. The performance of the resonators with cellular materials featuring a full characteristic range of Poisson’s ratios is then compared systematically to explore their energy-harvesting potential. The results revealed that the NPR structure exhibits superior performance for energy-harvesting. Moreover, stress distribution and displacement have also been highlighted in this paper.

Wearable glove gesture recognition based on fiber Bragg grating sensing using genetic algorithm-back propagation neural network

The recognition of gestures through intelligent wearable devices has a broad array of applications, including low-visibility industrial sites and robotics. However, accurately and swiftly recognizing gestures poses a significant challenge. This study addresses the complex interplay between multimodal information sensing and signal processing with flexible sensors in wearable data gloves by conducting theoretical and experimental investigations into gesture recognition using fiber Bragg grating (FBG) flexible sensors. First, the degree of bending and bending angle of the FBG sensor were calibrated, and the impact of temperature was investigated. Subsequently, a human hand wearable data glove gesture recognition system based on fiber grating sensing was constructed, and gesture recognition experiments were conducted. Finally, an enhanced error backpropagation (BP) neural network model for gesture recognition based on genetic algorithms (GA) was established and tested and the results were compared with those achieved using a conventional error BP neural network. The comparison revealed that under the same network structure, the recognition accuracy of the GA-BP neural network of approximately 100% was higher than the conventional BP neural network’s accuracy of 95.77%. Thus, the development of gesture recognition using FBG sensors provides a theoretical foundation and technical support for multiple applications of wearable data gloves, driving their adoption in areas such as human–computer interactions, intelligent robotics, and complex high-risk environments.

Application of Predictions for the Total Productivity of Competency Certificates Electrical Engineering Staff at PT. Eleska Iatki Using Artificial Neural Network Backpropagation Algorithm

Application of Predictions for the Total Productivity of Competency Certificates Electrical Engineering Staff at PT. Eleska Iatki Using Artificial Neural Network Backpropagation Algorithm

Integrated measurement of public safety risks in international construction projects in the belt and road initiative

PurposeThe purpose of this paper is to develop a comprehensive understanding of the public safety risks of international construction projects (ICPs) from the perspective of threat and vulnerability. A novel and comprehensive risk assessment approach is developed from a systemic perspective and applied to the Belt and Road Initiative (BRI) to improve the public safety risk management strategy for ICPs in BRI.Design/methodology/approachFirst, a public safety risk indicator system was constructed from the two dimensions, namely threat and vulnerability. Next, an integrated measurement model was constructed by combining the Genetic Algorithm-Backpropagation (GA-BP) neural network, fuzzy comprehensive evaluation method and matter-element extension (MME) method. Data from 49 countries involved in the BRI, as well as five typical projects, were used to validate the model. Finally, targeted risk prevention measures were identified for use at the national, enterprise and project levels.FindingsThe findings indicate that while the vulnerability risks of typical projects in each region of the BRI were generally low, threat risks were high in West Asia and North Africa, Commonwealth of Independent States (CIS) countries and South Asia.Originality/valueFirst, the structure of the public safety risk system of ICPs was analyzed using vulnerability and system theories. The connotation of public safety risk was defined based on two dimensions, namely threat and vulnerability. The idea of measuring threat risk with public data and measuring vulnerability risk with project data was clarified, and the risk measurement was integrated into the measurement results to help researchers and managers understand and systematically consider the public safety risks of ICPs. Second, a public safety risk indicator system was constructed, including 18 threat risk indicators and 14 vulnerability risk indicators to address the gaps in the existing research. The MEE model was employed to overcome the problem of incompatible indicator systems and provide stable and credible integrated measurement results. Finally, the whole-process public safety risk management scheme designed in this study can help to both provide a reference point for the Chinese enterprises and oversea contractors in market selection as well as improve ICP public safety risk management.

Research on driving condition recognition in hybrid electric vehicle energy management strategy

Abstract The driving conditions encountered by vehicles during operation are complex and varied, and the ECMS (Equivalent fuel Consumption Minimization Strategy) based on minimum equivalent fuel consumption has poor adaptability to different driving conditions. To optimize the adaptability of the ECMS and improve the overall vehicle economy, this paper proposes an operating condition recognition model based on the Back Propagation neural network algorithm to optimize the ECMS energy management control strategy. Firstly, three types of driving conditions were identified: urban, suburban, and high-speed conditions. A complete vehicle model was constructed on the Matlab/Simulink platform. The efficacy of the strategy, before and after optimization, is verified under the WLTP (Worldwide Harmonized Light Vehicles Test Procedure) driving cycle. Simulation results show that the optimized ECMS can better adjust the power apportionment among the engine and motor, reducing the fuel consumption per hundred kilometers by 4.01% and thereby enhancing the overall vehicle economy.

Optimization of the Welding Parameters of High-Quality Aluminum/Copper FSSW Joints Using Taguchi Method Combined with Back Propagation Neural Network and Genetic Algorithm

Optimization of the Welding Parameters of High-Quality Aluminum/Copper FSSW Joints Using Taguchi Method Combined with Back Propagation Neural Network and Genetic Algorithm

Prediction of Physical Ability of College Students Based on the BP Neural Network

The prediction of physical ability is a key point to understand the physical training effect of college students. This paper uses the error Back Propagation neural network algorithms to investigate the college students’ physical test results, and predicts the future trends of the results. The findings indicate that, in future ten years, the physical test grades of both men and women will slip to a certain extent. It implies that the indicators of speed, power, endurance level and physical strength of college students keep going down. This phenomenon should be paid attention to. It is necessary to strength the exercise consciousness, and stimulates the enthusiasm of students’ exercise.