Traditional Back Propagation Neural Network Research Articles

Retracted] Supply Capability Evaluation of Intelligent Manufacturing Enterprises Based on Improved BP Neural Network

With the rapid development of economy and information technology, traditional manufacturing industry is facing severe challenges. Enterprises need to rectify the traditional manufacturing industry and realize the transformation from traditional manufacturing industry to intelligent manufacturing industry. In order to adapt to market demand, enterprises need to constantly integrate resources to improve the competitiveness of enterprise supply chain. Based on the background of suppliers in intelligent manufacturing enterprises, the evaluation method of supplier efficiency was studied by using machine learning. In this paper, based on the traditional backpropagation (BP) neural network, combined with the improved particle swarm optimization (PSO) algorithm, and on the basis of the supplier evaluation index system, the supplier efficiency evaluation model of intelligent manufacturing enterprises based on DPMPSO‐BP neural network is constructed. Through the collected sample data, the network is trained and simulated, and the results are analyzed. Finally, the designed model is applied to a large battery manufacturing enterprise, and the supplier efficiency evaluation method based on DPMPSO‐BP neural network is validated and analyzed. Compared with the traditional BP neural network method, the supplier efficiency evaluation method is effective and feasible.

A Study on Carbon Emission Forecasting in China Based on PSO-BP Neural Network

To address the problem that the traditional BP(Back-Propagation) neural network is prone to fall into the local optimal situation for carbon emission prediction, and thus the prediction results have large errors, this paper proposes to improve the BP neural network by using particle swarm optimization algorithm, and conducts simulation experiments with national carbon emission data. The results show that the model has a 12.1% reduction in error compared with the traditional BP neural network model, and has a better prediction effect on carbon emissions.

CROP CLASSIFICATION BASED ON A GAOFEN 1/WIDE-FIELD-VIEW TIME SERIES

The accurate acquisition of crop planting area information allows for agricultural management departments to understand crop production information promptly. Concurrently, support vector machine (SVM) algorithms are unable to determine parameter combinations in remote sensing image crop classification to obtain optimal classification results. To solve this issue, this paper proposes an SVM that is optimized by the adaptive mutation particle swarm optimization algorithm. To test the algorithm, we undertook an experiment in Acheng District, Harbin City, Heilongjiang province, China, using Gaofen 1/wide-field-view satellite images to construct a time series for various vegetation indices. The SVM model with optimized parameters was compared with a traditional backpropagation neural network, a decision tree, and an SVM without parameter optimization. The experimental results verified that the improved SVM model obtained the highest classification accuracy.

Study on Early Warning on the Financial Risk of Project Venture Capital through a Neural Network Model

This paper aims to effectively reduce the financial loss of enterprises by accurately and reasonably making early warning of investment project risks. This paper briefly introduced the index system used for investment project risk early warning. It constructed a project investment risk early-warning model with a back-propagation neural network (BPNN) algorithm, and improved it with a genetic algorithm (GA) to solve the defect that the traditional BPNN is easy to fall into, over-fitting when reversely adjust parameters. An analysis was conducted on an electric power company in Hunan Province. Orthogonal experiments are performed to determine the population size and the number of hidden layers in the improved BPNN algorithm. The results showed that the improved BPNN algorithm had the best performance when the population size was set as 25 and the number of hidden layers was four; compared with support vector machine (SVM) and traditional BPNN algorithms, the GA-improved BPNN algorithm had better performance for early risk warning of investment projects. In conclusion, adjusting the parameters of a BPNN with a GA in the training stage can effectively avoid falling into over-fitting, thus improving the early warning performance of the algorithm; in addition, the improved BPNN has better early warning performance.

Internet Financial Risk Monitoring and Evaluation Based on GABP Algorithm

Due to the generality and particularity of Internet financial risks, it is imperative for the institutions involved to establish a sound risk prevention, control, monitoring, and management system and timely identify and alert potential risks. Firstly, the importance of Internet financial risk monitoring and evaluation is expounded. Secondly, the basic principles of backpropagation (BP) neural network, genetic algorithm (GA), and GABP algorithms are discussed. Thirdly, the weight and threshold of the BP algorithm are optimized by using the GA, and the GABP model is established. The financial risks are monitored and evaluated by the Internet financial system as the research object. Finally, GABP is further optimized by the simulated annealing (SA) algorithm. The results show that, compared with the calculation results of the BP model, the GABP algorithm can reduce the number of BP training, has high prediction accuracy, and realizes the complementary advantages of GA and BP neural network. The GABP network optimized by simulated annealing method has better global convergence, higher learning efficiency, and prediction accuracy than the traditional BP and GABP neural network, achieves better prediction effect, effectively solves the problem that the enterprise financial risk cannot be quantitatively evaluated, more accurately assesses the size of Internet financial risk, and has certain popularization value in the application of Internet financial risk prediction.

Prediction of Low-Energy Building Energy Consumption Based on Genetic BP Algorithm

Combined with the energy consumption data of individual buildings in the logistics group of Yangtze University, the analysis model scheme of energy consumption of individual buildings in the university is studied by using Back Propagation (BP) neural network to solve nonlinear problems and have the ability of global approximation and generalization. By analyzing the influence of different uses, different building surfaces and different energy-saving schemes on the change of building energy consumption, the grey correlation method is used to determine the main influencing factors affecting each building energy consumption, including uses, building surfaces and energy-saving schemes, which are used as the input of the model and the building energy consumption as the output of the model, so as to establish the building energy consumption analysis model based on BP neural network. However, in practical application, BP neural network has the defects of slow convergence and easy to fall into local minima. In view of this, this paper uses genetic algorithm to optimize the weight and threshold of BP neural network, completes the improvement of various building energy consumption analysis models, and realizes the qualitative analysis of building energy consumption. The model verification results show that the viscosity of the building energy consumption analysis model based on genetic algorithm improved BP neural network algorithm (GABP) in this paper is relatively high, which is more accurate than the results of the traditional BP neural network model, and the relative error of the analysis model is reduced from 11.56% to 8.13%, which proves that the GABP can be better suitable for the study of school building energy consumption analysis model, It is applied to the prediction of building energy consumption, which lays a foundation for the realization of carbon neutralization in the South expansion plan of Yangtze University.

Forecasting the Amount of Recyclables Using an Improved Differential Evolution-based Neural Network

Forecasting the amount of recyclables accurately affects succeeding activities such as the allocation and dispatch of resources and the manufacturing of recycled products, and thus plays an essential role in building a successful municipal recycling system. In this paper, the logistic chaotic map and differential evolution algorithm are used to provide optimal initial weights and thresholds for the back-propagation neural network to improve its prediction accuracy. A case study with real data collected from four communities in Shanghai demonstrates the performance of the proposed model. Results show that the proposed model can reduce the average mean absolute error by 27.17% compared to the traditional back-propagation neural network model.

BP Neural Network Retrieval for Remote Sensing Atmospheric Profile of Ground-Based Microwave Radiometer

Vertical distributions of temperature and humidity are two essential factors for understanding the atmospheric structure, extreme weather events, and regional and global climate. The ground-based microwave radiometer (MWR), which acts as a passive sensor and operates continuously under all weather conditions, has an irreplaceable role in measuring the vertical information of the temperature and water content in the atmosphere. In this letter, we proposed a four-layer back-propagation neural network (BPNN) method to retrieve temperature and relative humidity (RH) profiles from the bright temperature measured by the MWR. In contrast to the traditional BPNN, this method has greater advantages in dealing with the problems of overfitting, gradient disappearance, and gradient explosion in vertical atmospheric retrieval. By adding dropout layers, it can also help to describe the nonlinear relationships for RH profiles. Results showed that the performance of the four-layer BPNN method was better than the quadratic regression (QR, provided by MWR manufacturer) method under both cloud and cloud-free conditions. Compared with measurements of radiosonde data, root-mean-square error of temperature and RH, BPNN achieves 1.88 K and 19.30% under cloud conditions and 2.03 K and 15.10% under cloud-free conditions, respectively, whereas the corresponding values by using the QR method were only 3.07 K and 24.28% under cloud conditions and 4.14 K and 18.96% under cloud-free conditions, respectively. Temperature and RH profiles retrieval with high precision have increased the efficiency of the MWR observations and provided a data foundation for further atmospheric climate research.

Fault Diagnosis of Synchronous Belt of Machine Tool Based on Improved Back Propagation Neural Network

Aiming at the problem that the machine tool synchronous belt failure during the transmission process will affect the machine tool transmission, a machine tool synchronous belt fault diagnosis method based on genetic algorithm (GA) optimized back propagation (BP) neural network is proposed. First, utilize wavelet decomposition to extract the energy characteristics of the synchronization belt fault; construct a BP neural network, and use genetic algorithms to optimize the BP neural network; finally, the energy characteristic of the vibration signal of the synchronous belt is used as the input of the neural network, and the fault simulation test is carried out. The results show that the genetic algorithm GA-optimized BP neural network has higher accuracy than the traditional BP neural network for fault diagnosis of machine tool synchronous belt.

Using a Hybrid Neural Network Model DCNN–LSTM for Image-Based Nitrogen Nutrition Diagnosis in Muskmelon

In precision agriculture, the nitrogen level is significantly important for establishing phenotype, quality and yield of crops. It cannot be achieved in the future without appropriate nitrogen fertilizer application. Moreover, a convenient and real-time advance technology for nitrogen nutrition diagnosis of crops is a prerequisite for an efficient and reasonable nitrogen-fertilizer management system. With the development of research on plant phenotype and artificial intelligence technology in agriculture, deep learning has demonstrated a great potential in agriculture for recognizing nondestructive nitrogen nutrition diagnosis in plants by automation and high throughput at a low cost. To build a nitrogen nutrient-diagnosis model, muskmelons were cultivated under different nitrogen levels in a greenhouse. The digital images of canopy leaves and the environmental factors (light and temperature) during the growth period of muskmelons were tracked and analyzed. The nitrogen concentrations of the plants were measured, we successfully constructed and trained machine-learning- and deep-learning models based on the traditional backpropagation neural network (BPNN), the emerging convolution neural network (CNN), the deep convolution neural network (DCNN) and the long short-term memory (LSTM) for the nitrogen nutrition diagnosis of muskmelon. The adjusted determination coefficient (R2) and mean square error (MSE) between the predicted values and measured values of nitrogen concentration were adopted to evaluate the models’ accuracy. The values were R2 = 0.567 and MSE = 0.429 for BPNN model; R2 = 0.376 and MSE = 0.628 for CNN model; R2 = 0.686 and MSE = 0.355 for deep convolution neural network (DCNN) model; and R2 = 0.904 and MSE = 0.123 for the hybrid model DCNN–LSTM. Therefore, DCNN–LSTM shows the highest accuracy in predicting the nitrogen content of muskmelon. Our findings highlight a base for achieving a convenient, precise and intelligent diagnosis of nitrogen nutrition in muskmelon.

Lithological information extraction and classification in hyperspectral remote sensing data using Backpropagation Neural Network.

The purposes are to solve the isomorphism encountered while processing hyperspectral remote sensing data and improve the accuracy of hyperspectral remote sensing data in extracting and classifying lithological information. Taking rocks as the research object, Backpropagation Neural Network (BPNN) is introduced. After the hyperspectral image data are normalized, the lithological spectrum and spatial information are the feature extraction targets to construct a deep learning-based lithological information extraction model. The performance of the model is analyzed using specific instance data. Results demonstrate that the overall accuracy and the Kappa coefficient of the lithological information extraction and classification model based on deep learning were 90.58% and 0.8676, respectively. This model can precisely distinguish the properties of rock masses and provide better performance compared with the state of other analysis models. After introducing deep learning, the recognition accuracy and the Kappa coefficient of the proposed BPNN model increased by 8.5% and 0.12, respectively, compared with the traditional BPNN. The proposed extraction and classification model can provide some research values and practical significances for the hyperspectral rock and mineral classification.

Short-Term Electricity Price Forecasting Based on BP Neural Network Optimized by SAPSO

In the electricity market environment, the market clearing price has strong volatility, periodicity and randomness, which makes it more difficult to select the input features of artificial neural network forecasting. Although the traditional back propagation (BP) neural network has been applied early in electricity price forecasting, it has the problem of low forecasting accuracy. For this reason, this paper uses the maximum information coefficient and Pearson correlation analysis to determine the main factors affecting electricity price fluctuation as the input factors of the forecasting model. The improved particle swarm optimization algorithm, called simulated annealing particle swarm optimization (SAPSO), is used to optimize the BP neural network to establish the SAPSO-BP short-term electricity price forecasting model and the actual sample data are used to simulate and calculate. The results show that the SAPSO-BP price forecasting model has a high degree of fit and the average relative error and mean square error of the forecasting model are lower than those of the BP network model and PSO-BP model, as well as better than the PSO-BP model in terms of convergence speed and accuracy, which provides an effective method for improving the accuracy of short-term electricity price forecasting.

Coupling Backpropagation Neural Network and AdaBoost Algorithm for Quantitative Analysis of Nickel via Laser-Induced Breakdown Spectroscopy

A wide area of cropland or soil might well be contaminated with heavy metals, contaminating agricultural goods and posing a risk to human health. As a result, it is required to evaluate the concentration of heavy metals in soil. The combination between laser-induced breakdown spectroscopy (LIBS) and multivariate chemometrics methods was employed to determine heavy metal Ni concentration in twelve soil samples. The comparison between univariate calibration curve, traditional backpropagation neural network (BPNN), and hybrid BPNN-AdaBoost was presented. The result revealed that BPNN-AdaBoost outperformed other models with the coefficient determination calibration , coefficient determination prediction , root mean square error calibration (RMSEC), root mean square error prediction (RMSEP) are 0.985, 0.977, 2.04, 3.18, respectively. This study indicates that BPNN-AdaBoost can be adopted as a reliable chemometric technique to enhance the quantitative analysis of heavy metals based on LIBS.

Projectile drag coefficient identification based on extreme learning

Aerodynamic parameters play a decisive role in the ballistic characteristics of the projectile. How to accurately obtain the aerodynamic parameters of the projectile is an important task in the development process of the projectile. In order to further improve the identification accuracy of the projectile drag coefficient, this paper generates huge ballistic data through numerical simulation and uses the extreme learning method to identify the ballistic drag coefficient under three kinds of noise conditions. The method avoids the iterative updating process of weights and thresholds by randomly generating the input weights and threshold values of hidden layer neurons and overcomes the problem of long identification time of the traditional back propagation (BP) neural network algorithm. Based on the least squares principle, the Moore–Penrose generalized inverse matrix of the hidden layer output matrix was solved to determine the optimal output weight of the network, and then, the projectile drag coefficient was accurately identified. Comparing the extreme learning method with the traditional BP neural network method, the results show that the proposed method has higher identification accuracy and faster convergence speed and can effectively identify the projectile drag coefficient, which can meet the practical needs of engineering.

Application Of Particle Swarm Optimization and Improved PSO-BP Algorithm in Computer Forecasting Model

Given the shortcomings of the traditional BP (Back Propagation) neural network in dealing with medium and long-term stock forecasting, this paper presents a combination forecasting model of BP network optimized by PSO (Particle Swarm Optimization). Firstly, the ARIMA (Autoregressive Integrated Moving Average) model was used to predict the stock's closing price, which was used as part of the BP neural network input. On this basis, the BP network was optimized by particle swarm optimization algorithm, and the optimal threshold and weight were given to the network for further prediction, which improved the generalization ability and prediction accuracy of the whole network. The results show that, compared with BP neural network, LSTM (Long Short-Term Memory) and ARIMA model, the PSO-BP network model has a significant improvement in forecasting accuracy, which verifies the effectiveness of the combination forecasting method in dealing with the problem of medium and long-term stock price forecasting.

Ship Short-Term Trajectory Prediction Based on RNN

In the application of ship supervision, ship collision avoidance, maritime search and rescue, the trajectory prediction of the target ship is a key issue. Given the ship navigation trajectory is easily affected by wind and waves, in order to improve the accuracy and efficiency of prediction, a ship short-term trajectory prediction method combining with Automatic Identification System (AIS) data and deep learning is proposed. Based on the preprocessing of AIS data, a recurrent neural network (RNN) model is constructed to achieve the accurate prediction of ship position information is realized. Through the real ship AIS trajectory data experiment, the results show that the method is practical and effective. Compared with the traditional backpropagation (BP) neural network processing method, it has certain advantages in prediction accuracy.

A method for predicting hobbing tool wear based on CNC real-time monitoring data and deep learning

Intelligent monitoring and diagnosis of tool status are of great significance for improving the manufacturing efficiency and accuracy of the workpiece. It is difficult to quickly and accurately predict the wear state of worm gear hob under different working conditions. This paper proposes a novel approach to predict hob wear status based on CNC real-time monitoring data. Based on the open platform communication unified architecture (OPC UA) technology and orthogonal test, the machine data of motor power, current, etc. related to tool wear are collected online in the worm gear machining process. And then, an improved deep belief network (DBN) is used to generate a tool wear model by training data. A growing DBN with transfer learning is introduced to automatically decide its best model structure, which can accelerate its learning process, improve training efficiency and model performance. The experiment results show that the proposed method can effectively predict hob wear status under multi-cutting conditions. To show the advantages of the proposed approach, the performance of the DBN is compared with the traditional back propagation neural network (BP) method in terms of the mean-squared error (MSE). The compared results show that this tool wear prediction method has better prediction accuracy than the traditional BP method during worm gear hobbing.

Hourly water demand forecasting using a hybrid model based on mind evolutionary algorithm

Abstract A hybrid model based on the mind evolutionary algorithm is proposed to predict hourly water demand. In the hybrid model, hourly water demand data are first reconstructed to generate appropriate samples so as to represent the characteristics of time series effectively. Then, the mind evolutionary algorithm is integrated into a back propagation neural network (BPNN) to improve prediction performance. To investigate the application potential of the proposed model in hourly water demand forecasting, real hourly water demand data were applied to evaluate its prediction performance. In addition, the performance of the proposed model was compared with a traditional BPNN model and another hybrid model where the genetic algorithm (GA) is used as an optimization algorithm for BPNN. The results show that the proposed model has a satisfactory prediction performance in hourly water demand forecasting. On the whole, the proposed model outperforms all other models involved in the comparisons in both prediction accuracy and stability. These findings suggest that the proposed model can be a novel and effective tool for hourly water demand forecasting.

Prediction of the compressive strength of high-performance self-compacting concrete by an ultrasonic-rebound method based on a GA-BP neural network.

To address the problem of low accuracy and poor robustness of in situ testing of the compressive strength of high-performance self-compacting concrete (SCC), a genetic algorithm (GA)-optimized backpropagation neural network (BPNN) model was established to predict the compressive strength of SCC. Experiments based on two concrete nondestructive testing methods, i.e., ultrasonic pulse velocity and Schmidt rebound hammer, were designed and test sample data were obtained. A neural network topology with two input nodes, 19 hidden nodes, and one output node was constructed, and the initial weights and thresholds of the resulting traditional BPNN model were optimized using GA. The results showed a correlation coefficient of 0.967 between the values predicted by the established BPNN model and the test values, with an RMSE of 3.703, compared to a correlation coefficient of 0.979 between the values predicted by the GA-optimized BPNN model and the test values, with an RMSE of 2.972. The excellent agreement between the predicted and test values demonstrates the model can accurately predict the compressive strength of SCC and hence reduce the cost and time for SCC compressive strength testing.

Identification of Pilots’ Fatigue Status Based on Electrocardiogram Signals

Fatigue is an important factor affecting modern flight safety. It can easily lead to a decline in pilots’ operational ability, misjudgments, and flight illusions. Moreover, it can even trigger serious flight accidents. In this paper, a wearable wireless physiological device was used to obtain pilots’ electrocardiogram (ECG) data in a simulated flight experiment, and 1440 effective samples were determined. The Friedman test was adopted to select the characteristic indexes that reflect the fatigue state of the pilot from the time domain, frequency domain, and non-linear characteristics of the effective samples. Furthermore, the variation rules of the characteristic indexes were analyzed. Principal component analysis (PCA) was utilized to extract the features of the selected feature indexes, and the feature parameter set representing the fatigue state of the pilot was established. For the study on pilots’ fatigue state identification, the feature parameter set was used as the input of the learning vector quantization (LVQ) algorithm to train the pilots’ fatigue state identification model. Results show that the recognition accuracy of the LVQ model reached 81.94%, which is 12.84% and 9.02% higher than that of traditional back propagation neural network (BPNN) and support vector machine (SVM) model, respectively. The identification model based on the LVQ established in this paper is suitable for identifying pilots’ fatigue states. This is of great practical significance to reduce flight accidents caused by pilot fatigue, thus providing a theoretical foundation for pilot fatigue risk management and the development of intelligent aircraft autopilot systems.