BP Network Research Articles

Aircraft engine fault diagnosis based on fused convolutional Transformer

Aircraft engines will face corrosion and erosion problems when working in harsh gas path environments for a long time, and the fault parameter characteristics are not obvious. Therefore, accurate aircraft engine fault diagnosis methods are of great significance to ensure the safe operation of aircraft. In order to improve the prediction accuracy, an aircraft engine fault diagnosis method based on fused convolutional Transformer is proposed. The self-attention mechanism is used to extract useful features and suppress redundant information. The maximum pooling layer (MaxPool) is introduced into the Transformer model to further reduce the model memory consumption and parameter quantity, and alleviate the overfitting phenomenon. The simulation data set of turbofan engines based on GasTurb modeling is used for verification. The results show that compared with the Transformer network and other traditional deep learning models Back Propagation Neural Networks (BP network), Convolutional Neural Networks (CNN) and Recurrent Neural Network (RNN), the accuracy rates are improved by 6.552%、28.117%、13.189%and respectively 10.29%, which proves the effectiveness of the proposed method and can provide a certain reference for aircraft engine fault diagnosis.

Linear Model Predictive Control and Back-Propagation Controller for Single-Point Magnetic Levitation with Different Gap Levitation and Back-Propagation Offline Iteration

Magnetic suspension balance systems (MSBSs) need to allow vehicle models to levitate stably in different attitudes, so it is difficult to ensure the stable performance of the system under different levitation gaps using a controller designed with single balance point linearization. In this paper, a levitation controller based on linear model predictive control and a back-propagation neural network (LMPC-BP) is proposed and simulated for single-point magnetic levitation. The deviation of the BP network is observed and compensated by an expansion state observer (ESO). The iterative BP neural network model is further updated using current data and feedback data from the ESO, and then the performance of the LMPC-BP controller is evaluated before and after the update. The simulation results show that the LMPC-BP controller can achieve stable levitation at different gaps of the single-point magnetic levitation system. With further updating and iteration of the BP network, the controller anti-jamming performance is improved.

Prediction of Optimal Design Parameters for Reinforced Soil Embankments with Wrapped Faces Using a GA–BP Neural Network

Under the same geological conditions, the thickness and length of the reinforced strip, the slope ratio of the reinforced embankment, the modulus of elasticity of the fill and the reinforced strip, and the friction angle at the interface between the reinforcement and the soil, are the main design parameters that have an important influence on the stress, deformation, and stability of the encompassing reinforced soil embankment. To quickly and accurately determine the optimal design parameters for reinforced soil embankments with wrapped faces, ensuring minimal cost, while maintaining structural safety, we propose a design parameter prediction model based on a GA–BP neural network. This model evaluates parameters within their specified ranges, using maximum lateral displacement, maximum vertical displacement, maximum stress in the XZ direction, the maximum shear strain increment, and the safety factor, as assessment criteria. The primary objective is to minimize the overall cost of the embankment. A comparison with five machine learning algorithms shows that the model has high prediction accuracy, and the optimal design parameter combinations obtained from the optimization search can significantly reduce the cost of the embankment, while controlling the displacement and stability of the embankment. Therefore, the GA–BP network is suitable for predicting the optimal design parameters of reinforced soil embankments with wrapped faces.

Prediction of residual aluminum concentration and size in water plants of South-to-North Water Transfer Project through machine learning

Effluent water in South-to-North Water Transfer Project is complicated, leading to possibility of excessive residual aluminum if the coagulant dosage is improper. There exists certain risk for controlling residual aluminum based on experiences, while using machine learning can offer appropriate dosage based on the effluent water parameters. Much attention has been paid to the aluminum below 0.45 μm while for water plant with sand filter only, aluminum over 0.45 μm can still be remnant in the effluent water. In this study, machine learning algorithms were applied for the prediction of pH, aluminum concentration and size after coagulation based on the water parameters and ESI-TOF-MS results of the coagulant. The results showed that conventional neural network (CNN) showed best prediction ability compared with BP network, random tree and support vector machine, whose R2 was 0.936. The predicted results showed that the aluminum size decreased as the basicity increased at the same dosage, the concentration of aluminum between 20 and 30 μm dropped from 65 % to 24 %.

Construction and application of a core competitiveness evaluation index system for public health personnel training based on the CIPP model

BACKGROUND: With the change of human disease spectrum and the increase of public emergencies, the public’s demand for health services and health is gradually increasing and higher expectations are placed on public health personnel. OBJECTIVE: Research needs to establish a comprehensive system of evaluation indices for accurate assessment of the core competencies of public health personnel, to enhance their core competitiveness, and introduce novel approaches to evaluate talent development in the field of public health. METHODS: The study is based on the CIPP (Context Input Process Product) model and uses literature analysis, semi-structured interviews, and Delphi methods to construct an evaluation index system for the core competitiveness of public health talent cultivation. The entropy method is used to determine the weight of the model evaluation index. Finally, the improved Artistic Be Colony algorithm (ABC) is used to optimize the BP network, and apply it to evaluate the core competitiveness of public health talent cultivation. RESULTS: The improved BP network achieved the target accuracy within 11 iterations, with the optimal value observed after 16 iterations, producing an MSE (Mean Square Error) value of 10–13. The evaluation of the index system reported a 97% accuracy, and upon application to the university’s public health training programme, nearly 50% of students and teachers achieved core competitiveness quality scores above 90. CONCLUSIONS: The aforementioned method suggests that it can proficiently assess the fundamental competitiveness of training for public health personnel and offer guidance for future development in the domain of public health.

Optimization of Power Prediction of BP Network with Improved Pelican Algorithm

Abstract In view of the large fluctuation of photovoltaic output power affected by different weather, accurate prediction of photovoltaic output power is particularly important for the safe and stable operation of power system. Firstly, the pelican optimization algorithm ( POA ) is improved in the following three aspects : adding Circle chaotic map to make the population evenly distributed, introducing mutation factor to expand the search range of pelican when approaching prey, adding adaptive weight and firefly disturbance to avoid falling into local optimum in the water surface flight stage ; then, in order to improve the prediction accuracy of BP algorithm, the improved pelican algorithm ( IPOA ) is used to optimize the weights and thresholds of BP neural network, and the IPOA-BP photovoltaic power prediction model is built to improve the accuracy of power prediction. Finally, this paper tests the prediction performance of IPOA-BP, POA-BP and basic BP power prediction models in sunny, cloudy and rainy days through experiments. The experimental results demonstrate that the IPOA-BP prediction model outperforms both the POA-BP and traditional BP neural network models under various weather conditions.

Research on Electricity Load Forecasting Based on LSTM Networks and BP Networks

The aim of this study is to compare the performance of Long Short-Term Memory (LSTM) network with Back Propagation Neural Network (BP network) for electricity load forecasting. Two prediction methods are used in the study: the first one uses a sliding window approach to predict the load of the next hour using the electricity load data of the past 24 hours and external factors such as temperature and humidity; the second one uses the data of the same time point of the past 24 days to predict the load of the corresponding time point of the next day. The results show that in the first method, the BP network outperforms the LSTM in prediction; while in the second method, both networks perform poorly, but the LSTM network outperforms the BP network in prediction. This finding provides new insights into power load forecasting, which is important for power system efficiency and reliability improvement.

Inversion of the Permeability Coefficient of a High Core Wall Dam Based on a BP Neural Network and the Marine Predator Algorithm

The parameters’ inversion of saturated–unsaturated is important in ensuring the safety of earth dams; many scholars have conducted some research regarding the inversion of hydraulic conductivity based on seepage pressure monitoring data. The van Genuchten model is widely used in saturated–unsaturated seepage analysis, which considers the permeability connected to the water content of the soil and the soil’s shape parameters. A BP neural artificial network is a mature prediction technique based on enough data, and the marine predator algorithm is a new nature-inspired metaheuristic inspired by the movement of animals in the ocean. The BP neural artificial network and marine predator algorithm are applied in the permeability coefficient inversion of a core-rock dam in China; the results show that in the normal operation status, the BP network shows better accuracy, and the average of the absolute error and variance of the absolute error are both minimum values, which are 2.21 m and 1.43 m, respectively. While the water storage speed changes, the marine predator algorithm shows better accuracy; the objective function is calculated to be 0.253. So, the marine predator algorithm is able to accurately reverse the desired results in some situations. According to the actual condition, employing suitable methods for the inverse permeability coefficient of a dam can effectively ensure the safe operation of dams.

Electrochemical Exfoliation and Growth of Nickel–Cobalt Layered Double Hydroxides@Black Phosphorus Hetero‐Nanostructure Textiles for Robust Foldable Supercapacitors

AbstractAdvanced innovation of flexible electrode with adequate redox activity and stably mechanical endurance that promotes charges kinetic migration and faradaic storage is pivotal for textile‐based supercapacitors (T‐SCs). Herein, this study reports a high‐performance T‐SCs electrode based on nickel–cobalt layered double hydroxides@black phosphorus (NiCo‐LDHs@BP) on a conductive silk (c‐silk) textile (NiCo‐LDHs@BP/c‐silk). Under negative voltage‐induced electrochemical exfoliation, the Ni2+ and Co2+ are embedded into bulk BP framework to form exfoliated BP nanosheets, and the NiCo‐LDHs are in situ grown within the BP networks, generating 3D NiCo‐LDHs@BP hetero‐nanostructure. Significantly, the NiCo‐LDHs@BP exhibits a large space‐charge area, enhanced adsorption energy for OH− and accelerated charges transfer/storage as confirmed using density functional theory calculations. Additionally, the T‐SCs electrode is fabricated by loading the blended NiCo‐LDHs@BP, sericin, and carbon nanotubes on fibroin textile via a silk reconstruction strategy, producing large area production, superior mechanical flexibility, and impressive electrochemical performance. The resultant NiCo‐LDHs@BP/c‐silk electrode exhibits large specific capacitance of 1291.3 F g−1 and considerable rate capacity in 1 M KOH electrolyte. Furthermore, the flexible solid‐state asymmetric T‐SCs deliver high specific areal energy density of 279.6 µWh cm−2 and robust folding capability (85.6% capacitance retention after 5000 folding cycles), which successfully power wearable watch and heart rate meter devices.

Study on Soil Total Nitrogen Content Prediction Method Based on Synthetic Neural Network Model

Rational utilization of soil total nitrogen is one of the keys to achieving sustainable agricultural development. By accurately measuring the content of total nitrogen in the soil, the utilization efficiency of nitrogen in the soil can be improved, and the scientific use of chemical fertilizers can reduce the pressure of agriculture on natural resources and realize the sustainable development of agriculture. In order to measure soil total nitrogen content simply and accurately, combined with the method of artificial olfactory systems, a new method of soil total nitrogen content detection based on convolutional noise reduction autoencoder (CDAE)–whale optimization algorithm (WOA)–deep residual shrinkage network (DSRN) is proposed. In order to obtain more salient features for fusion, the channel mechanism of the DSRN is improved by adding global Max pooling. The model uses a CDAE for the first filtering stage to automatically obtain data that filters simple noise and uses the WOA to automatically optimize hyperparameters. Finally, the optimized hyperparameters were used to train the DRSN for secondary filtering and predict the soil total nitrogen content. Experimental results show that the R2 of CAE-WOA-DSRN test set is 0.968, which is significantly better than the R2 of a traditional algorithm (0.873) and a simple BP network (0.877), and it can more accurately measure soil total nitrogen content.

Enhancing urban blue-green landscape quality assessment through hybrid genetic algorithm-back propagation (GA-BP) neural network approach: a case study in Fucheng, China

This study employs an artificial neural network optimization algorithm, enhanced with a Genetic Algorithm-Back Propagation (GA-BP) network, to assess the service quality of urban water bodies and green spaces, aiming to promote healthy urban environments. From an initial set of 95 variables, 29 key variables were selected, including 17 input variables, such as water and green space area, population size, and urbanization rate, six hidden layer neurons, such as patch number, patch density, and average patch size, and one output variable for the comprehensive value of blue-green landscape quality. The results indicate that the GA-BP network achieves an average relative error of 0.94772%, which is superior to the 1.5988% of the traditional BP network. Moreover, it boasts a prediction accuracy of 90% for the comprehensive value of landscape quality from 2015 to 2022, significantly outperforming the BP network’s approximate 70% accuracy. This method enhances the accuracy of landscape quality assessment but also aids in identifying crucial factors influencing quality. It provides scientific and objective guidance for future urban landscape structure and layout, contributing to high-quality urban development and the creation of exemplary living areas.

Prediction and simulation of wearable sensor devices for sports injury prevention based on BP neural network

In the research of sports injury prevention, the recognition of sports action plays an important role in the action recognition model and the prediction evaluation model. In view of the above problems, this paper constructs a new mathematical model through the idea of BP neural network. The model combines wearable technology and can improve the recognition accuracy of sports actions. The model uses the BP network classifier, which can be used in data. Processes such as feature extraction improve efficiency and reliability. In this paper, the algorithm simulation is carried out, and the experiments are carried out for three movements: running, running and static. The results show that for the recognition of running and running action, when the hidden layer node is 11, the BP neural network classifier shows the best recognition effect. For static motion, the recognition effect of each classifier is basically the same. This paper analyzes the wearable sports action recognition system, including perception layer, application layer and service layer, to realize the recognition and classification of sports actions, predict actions in advance and prevent sports injuries. Finally, this paper analyzes the causes of sports injury, puts forward specific measures to prevent sports injury, and further reduces sports injury events through wearable devices.

Supplier Evaluation in Supply Chain Environment Based on Radial Basis Function Neural Network

The comprehensive evaluation and selection of suppliers under the environment of supply chain management has become a key factor affecting the success of supply chain. How to select suppliers and the strategic partnership between suppliers under the environment of supply chain management has become an important challenge. To solve this problem, this paper takes the supplier evaluation and selection of Guangzhou Automobile Toyota Company as the research object, constructs the index system of supplier comprehensive evaluation and selection, uses the RBF neural network algorithm to establish the supplier evaluation and selection model, and makes an experimental study. The results show that radial basis function neural network is a local approximation network, which has a unique and definite solution to the problem, and there is no local minimum problem in BP network. It is a method that enables enterprises and suppliers to have a clear understanding and seek further promotion together. The research provides theoretical data support for enterprise managers to make decisions.

Micro drill defect detection with hybrid BP networks, clusters selection and crossover

Micro drill defect detection with hybrid BP networks, clusters selection and crossover

Prediction of State of Charge for Lead-acid Batteries Based on GRU Network and Isolated Forest

Accurate prediction of the state of charge (SOC) of lead-acid batteries is the key to ensuring battery life. In this paper, a new combined SOC prediction model IF-GRU (Isolation Forest, Gated Recurrent Unit) is proposed. The model combines the Isolation Forest anomaly detection algorithm and the Gated Recurrent Network. The Isolation Forest algorithm is used to detect anomalous and missing values in the raw data. Length dependence of the GRU network can be further utilized to perform high-accuracy SOC estimation by implementing a sliding window that takes into account the data's charging and discharging details. In addition, the conventional Adam optimizer is utilized to improve the convergence speed of model training. The experimental data demonstrate that the IF-GRU model proposed in this paper has higher prediction accuracy and convergence speed with a RMSE of 1.59% compared with traditional LSTM network, GRU network, and BP network.

Ultra-conformal epidermal antenna for multifunctional motion artifact-free sensing and point-of-care monitoring

Accurate acquisition of physiological and physical information from human tissue is essential for health monitoring, disease prevention and treatment. The existing antennas with traditional rigid or flexible substrates are susceptible to motion artifacts in wearable applications due to the miniaturization limitation and lack of proper adhesion and conformal interfaces with the skin. Recent advances in wearable radio frequency (RF) bioelectronics directly drawn on the skin are a promising solution for future skin-interfaced devices. Herein, we present a first-of-its kind epidermal antenna architecture with skin as the antenna substrate, which is ultra-low profile, ultra-conformal, ultra-compact, and simple fabrication without specialized equipment. The radiation unit and ground of antenna are drawn directly on the skin with the strong adhesion and ultra conformality. Therefore, this RF device is highly adaptable to motion. As a proof-of- feasibility, epidermal antenna can be freely drawn on demand at different locations on the skin for the development of temperature sensor, skin hydration sensor, strain sensor, glucose sensor and other devices. An epidermal antenna-based temperature sensor can offer accurate and real-time monitoring of human body temperature changes in the ultra-wideband (UWB) range. The results during the monitoring of hydration level with and without stretching show that the epidermal antenna drawn on the skin is motion artifact-free. We also designed an epidermal antenna array employing a horseshoe-shaped configuration for the precise identification of various gestures. In addition, the non-invasive blood glucose level (BGL) monitoring results during the in-vivo experiments report high correlation between the epidermal antenna responses and BGLs, without any time hysteresis. After the prediction of BGL by BP network, all the predicted BGL values are fallen 100% into the clinically acceptable zones. Together, these results show that epidermal antenna offers a promising new approach for biosensing platform.

Study of BP and RBF Neural Networks Applied to the Prediction of Vibration Characteristics in Static Blasting of Dry Ice Powder

The novel dry ice powder static blasting rock breaking lacked working standards, especially for the vibration safety assessment of the construction site for the protection of the building structure, in comparison with the traditional drill and blast method which had a proven operational process and safety specifications. The Sadovsky vibration velocity prediction formula could only predict the vibration velocity and was project specific. Oscillation parameters that needed to be considered in the vibration safety assessment, such as the dominant frequency of vibration, could not be obtained through empirical formulas. Using the five parameters of hole depth, blast center distance, dry ice powder mass and rock classification as the main influencing factors, BP and RBF neural network models were constructed by Matlab software to predict the peak vibration velocity, main frequency and maximum displacement of dry ice powder blasting. Projection results revealed that it is structurally simpler than the BP neural network and that the RBF was more accurate in predicting the target than the BP network. The results of the study had significant implications for the safe application of the new technology, and more samples of field data need to be obtained in the future, along with the use of more advanced predictive modelling.

Prediction Model for Safe Operation of Pumping Stations Optimized by the Sparrow Search Algorithm and BP Neural Network

The pumping station is one of the critical parts of the hydraulic structure in China. Traditional forecasting methods are limited in accuracy, time-consuming, and high cost, resulting in limited data availability. Therefore, simulation model analysis based on soft computation is a realistic and valuable alternative. This article intends to use the BP neural network to predict the safe operation status of pump stations and optimize the initial threshold and weight information of the BP network using the sparrow search algorithm (SSA) to improve the accuracy and generalization ability of the model. In addition, to more accurately reflect the correlation between various influencing factors and the safe operation status of the pumping station, the entropy weight method and the analytic hierarchy process were used to obtain the comprehensive weights of each main influencing factor. The experimental results show that the SSA-BP model can accurately predict the safe operation status of pumping stations, and compared with other traditional models, the SSA-BP model has better convergence and higher accuracy. This model provides a new approach for predicting the safe operation of pumping stations and has particular reference significance for predicting the safe operation of other pumping stations.

Student sports data analysis and physical fitness evaluation based on convolutional neural networks

<p>In order to achieve the analysis of student sports data and physical fitness evaluation, the author proposes a method based on convolutional neural networks. A hybrid algorithm combining genetic algorithm and error backpropagation algorithm (BP) is used to train convolutional neural networks. The algorithm first uses genetic algorithm for global training, and then uses BP algorithm for local precise training. This overcomes the drawbacks of traditional BP networks such as long training time and frequent local atmospheric drift, and improves global circulation performance. A neural network model was established to display the relationship between the total physical activity score and multiple test scores of high school students by utilizing electrical networks to demonstrate the connectivity of the neural network. This model aims to evaluate the athletic performance of college students and compare the results with other experimental models. The results indicate that the neural network-based model for evaluating college student physical activity can reflect the differences in physical activity and scores among all students, making it a suitable standard for evaluating high school student physical activity. The fitting accuracy of deterministic neural network models is higher than that of multiple linear regression models, which means that neural network models better reflect the performance of the network. The accuracy of various indicators of student physical fitness and total score makes the model easy to operate, accurate to predict, and effective analysis is scientifically reasonable.</p>

Research on the application of neural networks in macroeconomic forecasting and strategic management decision-making

Abstract Traditional macroeconomic forecasting models have some limitations in utilizing large-scale variables, screening appropriate variables, and forecasting accuracy. Based on the theories of economic fluctuation and New Keynes, this paper uses monetary policy, investment behavior, and consumption behavior as predictors of macroeconomic fluctuation. Then, it explores the principles and steps of the BP network applied to forecasting, the design of BP neural network structure, the selection of training algorithm and training parameters, and finally establishes a reasonable BP network structure model. Finally, using the prediction of GDP growth rate as an example, an empirical comparative analysis of model prediction accuracy is carried out. The empirical results of the model in the training set and the test set show that the structure of the BP neural network model is 3-48-1, and the model obtains the optimal learning error (6.207×10−5) at 500 iterations. In the experimental set, the root-mean-square prediction error between the output predicted value and the actual value is less than ±0.2 . The model accurately predicts the GDP growth rate, which can provide a theoretical basis for proposing macroeconomic strategic management decisions.