Improved BP Neural Network Research Articles

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.

Intelligent STEP-NC-compliant setup planning method

This paper proposes an intelligent STEP-NC-compliant setup planning method with the combination of rules, analytic hierarchy process (AHP) and improved back propagation neural network (IBPNN). In the method, the STEP-NC-compliant manufacturing features are firstly automatically and reasonably grouped based on tool approach direction (TAD) selection rules. Then, the AHP is used to intelligently and accurately determine the sequence between STEP-NC-compliant manufacturing feature groups in consideration of machining stability. Next, with the combination of firefly algorithm, chaos algorithm and back propagation neural network, an IBPNN algorithm for the locating datum decision-making of STEP-NC-compliant manufacturing feature groups is presented. Further, the locating datum decision-making of STEP-NC-compliant manufacturing feature groups is intelligently and efficiently made after the machining information of a part is normalized and input into the IBPNN. At the end, the effectivity and feasibility of the proposed method are verified by case study.

Prediction of Ship Heave Motion Using Regularized BP Neural Network with Cross Entropy Error Function

Accurate prediction of ship’s heave motion can greatly enhance the safety of offshore operation. Due to its complexity and nonlinearity, however, ship’s heave motion prediction is a difficult task to be solved. In this paper, a new method for predicting ship’s heave motion is proposed based on an improved back propagation neural network (IBPNN). To overcome the gradient saturation phenomenon of traditional BPNN, the mean square error (MSE) loss function is replaced with a cross entropy (CE) loss function in IBPNN. Meanwhile, the weights of IBPNN is regularized by L_2 norm to enhance the generalization ability of traditional BPNN. Finally, conjugate gradient method is adopted to train IBPNN. The IBPNN is used to predict ship’s heave motion and the prediction results prove its effectiveness.

Complete characteristic curve processing method based on improved backpropagation neural network and Logarithmic curve projection

The calculation of transient process is the basis of the design and construction of pumped storage power plant, which directly affects operation stability of pumped storage units. However, for satisfying the design of uniform and smooth flow in both directions, the complete characteristic curves of pumped turbine show a significant S feature, which brings difficulties to the interpolation calculation in the transient process as there are crossover and overlap phenomenon in pump and turbine working conditions. In this paper, a transformation method for complete characteristic curves based on logarithmic curve projection and improved backpropagation neural network is presented to solve the above problem. Furthermore, the feasibility and accuracy of the method are verified through the comparison of load rejection condition with on-site measurement, the results show that the proposed method overcomes the multi-value problem that exists in the original curve, in especial makes the small opening region well expressed. The simulation based on the transformed curve reaches a high similarity of transient process with the field test, both in the trend and extreme values, which provides great convenience for the calculation of transient process.

VSG damping adaptive adjustment based on BP neural network

The moment of inertia and damping of virtual synchronous generator (VSG) can be adjusted flexibly, which also has a significant impact on the transient performance of VSG. Constant damping or moment of inertia can not reduce frequency overshoot and fast response performance, so it is necessary to introduce adaptive damping control. Based on universal approximation theorem, BP neural network can fit continuous nonlinear function well. At the same time, it has the advantages of simple algorithm, powerful learning ability and fast learning speed. Based on the characteristics of the control object, the BP neural network is improved and a new adaptive control strategy is designed. The strategy uses improved BP neural network to adjust VSG virtual damping D online. Python-MATLAB-Simulink was used for co-simulation, BP neural network algorithm was integrated into the control object to establish an adaptive simulation model, and the proposed control strategy was simulated and verified. Simulation results show that the adaptive control strategy can eliminate overshoot and respond quickly when the frequency and active power of virtual synchronous generator change.

An Algorithm for Construction Project Cost Forecast Based on Particle Swarm Optimization-Guided BP Neural Network

Construction project cost prediction is an important function in construction-related fields; it can provide an important basis for project feasibility study and design scheme comparison and selection, and its accuracy will directly affect the investment decision of the project. The successful realization of construction cost prediction can bring great convenience to the control and management of construction cost. The purpose of this paper is to study a fast, accurate, convenient, deducible, and rational construction project cost prediction method, to provide a basis for the cost management of the whole life cycle of the project. Therefore, this paper uses particle swarm optimization algorithm to improve BP neural network and proposes a novel construction project cost prediction algorithm based on particle swarm optimization-guided BP neural network. Aiming at the defects of BP neural network updating weights and thresholds with the gradient descent method, this paper uses the advantages of particle swarm optimization in the field of parameter optimization to optimize BP neural network with PSO algorithm. The structure of BP neural network weights and the threshold of each neuron in the coding, through intelligent search for each particle, find the most suitable weights and thresholds, so that the BP neural network has faster convergence speed, better generalization ability, and higher prediction precision. Simulation results also show that the proposed algorithm is competitive enough.

Probabilistic Prediction of Unsafe Event in Air Traffic Control Department Based on the Improved Backpropagation Neural Network

Air traffic control is an important tool to ensure the safety of civil aviation. For the departments that do the work of air traffic control, reducing the percentage of unsafe event is the core task of safety management. If the relationship between the percentage of unsafe event and their influencing factors can be effectively clarified, then the probability of unsafe event in some control department can be predicted. So, it is of great importance to improve the level of safety management. To quantitatively estimate the probability of unsafe event, a three-layer BP neural network model is introduced in this paper. First, a probabilistic representation of unsafe event related to air traffic control department is made, and then, the probability of different classes of unsafe events and safe events is taken as the outputs of the BP neural network, the factors influencing occurrence of unsafe event connected with air traffic control is taken as inputs, and the sigmoid function is chosen as activation function for the hidden layer. Based on the error function of neural network, it is proved that the general BP neural network has two drawbacks when used for the training of small probability events, which are as follows: the pattern does not ensure that the sum of probability of all events is equal to one and the relative error between the actual outputs and desired outputs is very large after the training of neural network. The reason proved in this paper is that the occurrence rate of the unsafe event is much smaller than that of the safe event, resulting in each weight in the hide layer being subjected to the desired outputs of the safe event when using the gradient descent method for network training. To address this issue, a new mapping method is put forward to reduce the large difference of the desired outputs between the safe event and unsafe event. It is theoretically proved that the mapping method proposed in this paper can not only improve the training accuracy but also ensure that the sum of probability is equal to one. Finally, a numeric example is given to demonstrate that the method proposed in this paper is effective and feasible.

Short-term wind speed forecasting based on long short-term memory and improved BP neural network

Accurate and reasonable wind speed prediction system has a significant impact on the utilization of wind energy. A novel combination forecasting model based on Long Short-Term Memory (LSTM) network and BP neural network is designed in this paper. This model combines the principle of deep learning algorithm and the improved BP neural network to deal with nonlinear wind speed prediction. Before the prediction, singular spectrum analysis (ssa) and complete ensemble empirical model decomposition adaptive noise (CEEMDAN) are selected as the data pretreatment part to de-noise the original wind speed data and decompose it into multiple components. This part is conducive to improving the signal-to-noise ratio (SNR) of wind speed data and simplifying the characteristics of wind speed data. Then, in order to reduce the error accumulation and computation redundancy, fuzzy entropy (FE) is used to calculate the time complexity of each component, according to the Spearman correlation, the inherent mode function (IMF) components are recombined to form a new subsequence. Experimental results show that the error accumulation can be reduced by 48.65% for dataset 1 and 29.53% for dataset 2, and the operation time can be reduced by about 50% for two datasets. To avoid the limitation of a single model, introducing the LSTM and improved BPNN which improved by sparrow search algorithm (SSA) two different prediction models are used to predict the sub sequences with high complexity and the low complexity subsequences, respectively. Finally, the predicted values of the models are superimposed to get the final values. In order to verify the validity of the proposed model, the final predictions, compared with six different prediction models, show that this model can achieve the best performance and obtain higher prediction accuracy. Such as the performance evaluation indexes (RMSE = 0.051, MAPE = 0.929%) are smallest obtained from dataset1 by one-step prediction, and (RMSE = 0.086, MAPE = 0.966%) are smallest obtained from dataset2 by one-step prediction. In addition, the Pearson correlation between the predicted value and the true wind speed value obtained by the prediction model applied to the two data sets is the highest 99.17% and 98.73%, respectively.

Optimal design of noise reduction and shape modification for traction gears of EMU based on improved BP neural network

Under high-speed operating conditions, the noise caused by the vibration of the traction gear transmission system of the Electric Multiple Units (EMU) will distinctly reduce the comfort of passengers. Therefore, analyzing the dynamic characteristics of traction gears and reducing noise from the root cause through comprehensive modification of gear pairs have become a hot research topic. Taking the G301 traction gear transmission system of the CRH380A high-speed EMU as the research object and then using Romax software to establish a parametric modification model of the gear transmission system, through dynamics, modal and Noise Vibration Harshness (NVH) simulation analysis, the law of howling noise of gear pair changes with modification parameters is studied. In the small sample training environment, the noise prediction model is constructed based on the priority weighted Back Propagation (BP) neural network of small noise samples. Taking the minimum noise of high-speed EMU traction gear transmission as the optimization goal, the simulated annealing (SA) algorithm is introduced to solve the model, and the optimal combination of modification parameters and noise data is obtained. The results show that the prediction accuracy of the prediction model is as high as 98.9%, and it can realize noise prediction under any combination of modification parameters. The optimal modification parameter combination obtained by solving the model through the SA algorithm is imported into the traction gear transmission system model. The vibration acceleration level obtained by the simulation is 89.647 dB, and the amplitude of the vibration acceleration level is reduced by 25%. It is verified that this modification optimization design can effectively reduce the gear transmission.

Research on emotion recognition algorithm based on improved BP neural network

The application of small robots in the production of freeze frame animation can speed up the production progress and improve the quality of production. Small robots will play the role of intelligent agents. It is necessary to recognize their emotions for the promotion of animation films. With the development of intelligent technology, BP neural network has been applied to the field of emotion recognition because of its advantages of self-learning, self-adaptive and self-organization. In this paper, the parameters of BP neural network are adjusted and improved to avoid the defects of BP neural network in emotion recognition. Then, the parameters and structure of the network are designed according to the emotion category. Then a three-layer BP neural network with 4 input nodes, 13 hidden layer nodes and 4 output nodes is constructed. Finally, it is applied to emotion recognition, and 10 groups of data are selected from the training samples for detection, and the diagnostic accuracy is 80%.

A particle swarm optimization improved BP neural network intelligent model for electrocardiogram classification

BackgroundAs proven to reflect the work state of heart and physiological situation objectively, electrocardiogram (ECG) is widely used in the assessment of human health, especially the diagnosis of heart disease. The accuracy and reliability of abnormal ECG (AECG) decision depend to a large extent on the feature extraction. However, it is often uneasy or even impossible to obtain accurate features, as the detection process of ECG is easily disturbed by the external environment. And AECG got many species and great variation. What’s more, the ECG result obtained after a long time past, which can not reach the purpose of early warning or real-time disease diagnosis. Therefore, developing an intelligent classification model with an accurate feature extraction method to identify AECG is of quite significance. This study aimed to explore an accurate feature extraction method of ECG and establish a suitable model for identifying AECG and the diagnosis of heart disease.MethodsIn this research, the wavelet combined with four operations and adaptive threshold methods were applied to filter the ECG and extract its feature waves first. Then, a BP neural network (BPNN) intelligent model and a particle swarm optimization (PSO) improved BPNN (PSO-BPNN) intelligent model based on MIT-BIH open database was established to identify ECG. To reduce the complexity of the model, the principal component analysis (PCA) was used to minimize the feature dimension.ResultsWavelet transforms combined four operations and adaptive threshold methods were capable of ECG filtering and feature extraction. PCA can significantly deduce the modeling feature dimension to minimize the complexity and save classification time. The PSO-BPNN intelligent model was suitable for identifying five types of ECG and showed better effects while comparing it with the BPNN model.ConclusionIn summary, it was further concluded that the PSO-BPNN intelligent model would be a suitable way to identify AECG and provide a tool for the diagnosis of heart disease.

Hybrid binary quantization and BP network on action mechanism

BP neural network is a kind of artificial neural network, which is mainly used in function approximation, model recognition, classification and data compression. In the case of limited analysis ability of traditional research methods, BP neural network can overcome the problems of nonlinear relationship. In this paper, the improved BP neural network was used to predict the properties. The data processing of the properties, tastes and was carried out by binary quantization. The prediction results’ correct rate was 87.5%.

Martial Arts Competitive Decision-Making Algorithm Based on Improved BP Neural Network.

The main body of modern Chinese martial arts competition is the strategy, and fighting has just started in sports competitions. Strategy and action correspond to each other and practice as a set. Therefore, constructing the Chinese martial arts competition decision-making algorithm and perfecting the martial arts competition are intuitive and essential. The formulation of martial arts competition strategies requires scientific analysis of athletic data and more accurate predictions. Based on this observation, this paper combines the popular neural network technology to propose a novel additional momentum-elastic gradient descent. The BP neural network adapts to the learning rate. The algorithm is improved for the traditional BP neural network, such as selecting learning step length, the difficulty of determining the size, and direction of the weight, and the learning rate is not easy to control. The experimental results show that this paper's algorithm has improved both network scale and running time and can predict martial arts competition routines and formulate scientific strategies.

Error analysis method of intelligent electricity meter based on improved BP neural network

Many smart electricity meters work in different nonideal environments. The complexity of the working environment often leads to the difference between the measurement error of electricity meters and the ideal conditions. In this paper, BP neural network is improved to make it suitable for error analysis of intelligent electricity meter. By identifying, processing, analyzing and integrating a large number of experimental data, the error model of smart electricity meter is established. Then the internal relationship between various stress factors and the error of electricity meter is revealed from the data level. This method can accurately predict the error of electricity meters under certain conditions and lay a foundation for error correction of smart electricity meters in the future. In this paper, the realization process of the method is briefly described through simulation experiments, and the feasibility, rationality and reliability of the method are verified.

An Improved BP Neural Network Algorithm for Prediction of Roadway Support

Based on the engineering practice and the research and analysis of the knowledge in the field of roadway support, the paper puts forward to use an improved BP neural network to study the supporting types by the investigation, and obtained the related factors of the supporting types of the mining roadway and the successful reinforcement cases of the roadway. The proposed algorithm is applied to the prediction of coal roadway support parameters, and the main influencing factors of coal roadway support design are determined. From the typical engineering cases of roadway support collected on site as neural network training samples, the forecasting model of support parameters is established. Through the experimental data and simulation results, it can be seen that both the error convergence process and results of convergence speed, convergence accuracy and support types are ideal, the prediction error is within the allowable range, and the prediction accuracy is high, which verifies the reliability of this method and provides a new research idea and good application value for the study of support types of mining roadway.

Image Recognition and Simulation Based on Distributed Artificial Intelligence

This paper studies the traditional target classification and recognition algorithm based on Histogram of Oriented Gradients (HOG) feature extraction and Support Vector Machine (SVM) classification and applies this algorithm to distributed artificial intelligence image recognition. Due to the huge number of images, the general detection speed cannot meet the requirements. We have improved the HOG feature extraction algorithm. Using principal component analysis (PCA) to perform dimensionality reduction operations on HOG features and doing distributed artificial intelligence image recognition experiments, the results show that the image detection efficiency is slightly improved, and the detection speed is also improved. This article analyzes the reason for these changes because PCA mainly uses the useful feature information in HOG features. The parallelization processing of HOG features on graphics processing unit (GPU) is studied. GPU is used for high parallel and high-density calculations, and the calculation of HOG features is very complicated. Using GPU for parallelization of HOG features can make the calculation speed of HOG features improved. We use image experiments for the parallelized HOG feature algorithm. Experimental simulations show that the speed of distributed artificial intelligence image recognition is greatly improved. By analyzing the existing digital image recognition methods, an improved BP neural network algorithm is proposed. Under the premise of ensuring accuracy, the recognition speed of digital images is accelerated, the time required for recognition is reduced, real-time performance is guaranteed, and the effectiveness of the algorithm is verified.

A novel toxicity prediction model for hydrazine compounds based on 1D–3D molecular descriptors

The present study analyzed the lethal dose (LD50) of 73 hydrazine compounds in rats and established a comprehensive toxicity database of hydrazine compounds. Based on the existing toxicity database, the 1D–3D molecular descriptors of hydrazine compounds were obtained by quantum chemistry calculations and using DRAGON 6.0 software, and subsequently a quantitative structure–activity relationship prediction model was constructed. Aiming at the existing defects of the traditional neural network, such as easy to fall into local optimum and limited by the number of data samples, this paper proposed an improved BP neural network to simulate the toxicity prediction of hydrazine compounds. The determination coefficient and error coefficient of the prediction data set were found to be 0.864 and 0.679, respectively. The simulation results showed that the improved neural network model offered better prediction results for the toxicity of hydrazine compounds, which met the actual accuracy requirements for the toxicity prediction for hydrazine compounds.

Research on timbre classification based on BP neural network and MFCC

Music exists in all aspects of life. With the development of deep learning and neural network technology in recent years, music information retrieval has become an emerging field that has received widespread attention. In the past, musicians often relied on manual labeling to classify music. This method is not only time-consuming and labor-intensive, but also inaccurate. Although some researchers have tried to extract music features for automatic classification, it is difficult to extract appropriate audio features due to the interaction between the fundamental wave and harmonics of the music itself. In this paper, the Mel cepstrum coefficient is selected to extract the timbre characteristics, and the improved BP neural network is used to process the characteristics, which overcomes the previous problems and obtains a higher accuracy rate.

Prediction of Arc Voltage of Electric Arc Furnace Based on Improved Back Propagation Neural Network

The Production of the quality steel is achieved by melting iron and steel scraps by utilizing electric power in an electric arc furnace, it acts as one of the major troublesome load in the electric power system due to high nonlinear and chaotic nature, thereby creating severe power quality disturbances to the interconnected network, therefor the need for model to describe the behavior of electric arc furnace in a simulation of electrical power system become significant, the relation between the arc voltage and arc current is pivotal in defining the characteristics of an electric arc furnace, in this paper, the Voltage and current characteristics based on real time data is discussed, then arc voltage prediction using improved back propagation neural network based on arc current under different zones of operation are simulated using MATLAB/SIMULINK, the results infer the validation of test results with the actual values and effectiveness of prediction by adoption of improved back propagation algorithm and efficient in terms of prediction with reduced errors.

The Analysis of Trajectory Control of Non-holonomic Mobile Robots Based on Internet of Things Target Image Enhancement Technology and Backpropagation Neural Network.

The trajectory tracking and control of incomplete mobile robots are explored to improve the accuracy of the trajectory tracking of the robot controller. First, the mathematical kinematics model of the non-holonomic mobile robot is studied. Then, the improved Backpropagation Neural Network (BPNN) is applied to the robot controller. On this basis, a mobile robot trajectory tracking controller combining the fuzzy algorithm and the neural network is designed to control the linear velocity and angular velocity of the mobile robot. Finally, the robot target image can be analyzed effectively based on the Internet of Things (IoT) image enhancement technology. In the MATLAB environment, the performances of traditional BPNN and improved BPNN in mobile robots' trajectory tracking are compared. The tracking accuracy before and after the improvement shows no apparent differences; however, the training speed of improved BPNN is significantly accelerated. The fuzzy-BPNN controller presents significant improvements in tracking speed and tracking accuracy compared with the improved BPNN. The trajectory tracking controller of the mobile robot is designed and improved based on the fuzzy BPNN. The designed controller combining the fuzzy algorithm and the improved BPNN can provide higher accuracy and tracking efficiency for the trajectory tracking and control of the non-holonomic mobile robots.