Mechanical Performance and Microstructure Prediction of Hypereutectoid Rail Steels Based on BP Neural Networks

Deng, Yong; Qiao, Lei; Zhu, Jingchuan; Yang, Bin

doi:10.1109/access.2020.2976135

Cited by 10 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[ 37 ] In general, the modeling process can be illustrated as follows. [ 38 ] It assumes n neurons in input layer, l neurons in hidden layer, and m neurons in output layer.

θ_{i j}

and

β_{k}

represent the connecting weight and offset between input layer and hidden layer and, additionally, denote the weight and offset from hidden layer to output layer.…”

Section: Methodsmentioning

confidence: 99%

Phase Prediction of High Carbon Pearlitic Steel: An Improved Model Combining Mind Evolutionary Algorithm and Neural Networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

Accurate phase constitution prediction is crucial for guiding the new steel design with desirable properties. This article uses three machine learning (ML) algorithms, backpropagation neural network (BPNN), radial basis function neural network (RBFNN), and fuzzy neural network (FNN), to compare the accuracy of predictive model. Toward the collected data, statistical measure of correlation is taken in the present work by determining Pearson correlation coefficient (PCC). The results show that the testing accuracy using BPNN is higher than the corresponding testing accuracy using RBFNN and FNN. With these understandings, the well‐known optimization algorithms, namely, mind evolutionary algorithm (MEA), are implemented to find optimal hyperparameters set that is able to induce a higher predictive capability. The resulting MEA‐BPNN further enhances the prediction results in the present dataset and efficiently explores the relationship between alloying elements and phase constitution. Finally, the reliability and practicability of the model are verified by experimental measurement on the ferrite content for prepared steels, and their mechanical properties are tested for engineering applications. As such, the work proposes a useful MEA‐BPNN model for predicting the phases of high carbon pearlite steel and provides an alternative route of designing new steels in a given system.

show abstract

“…[ 37 ] In general, the modeling process can be illustrated as follows. [ 38 ] It assumes n neurons in input layer, l neurons in hidden layer, and m neurons in output layer.

θ_{i j}

and

β_{k}

represent the connecting weight and offset between input layer and hidden layer and, additionally, denote the weight and offset from hidden layer to output layer.…”

Section: Methodsmentioning

confidence: 99%

Phase Prediction of High Carbon Pearlitic Steel: An Improved Model Combining Mind Evolutionary Algorithm and Neural Networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…The improvement of prediction results is attributed to the IPOA and the introduction of SE attention mechanisms. Taking the prediction results of stripping BP [20] 0.53 1.32 0.42 LSTM [21] 0.42 0.96 0.69 GRU [22] 0 BP [20] 0.46 0.86 0.49 LSTM [21] 0.47 0.75 0.61 GRU [22] 0 BIGRU [23] 0.41 0.89 0.74 CNN-BIGRU [24] 0 BIGRU [23] 0.38 0.60 0.75 CNN-BIGRU [24] 0…”

Section: Ablation Experimentsmentioning

confidence: 99%

“…The model structure and parameter settings of the rolling bearing experiment are consistent with section 3.1 of the rail experiment. BP [20] 0.94 1.35 0.25 LSTM [21] 0.84 1.11 0.49 GRU [22] 0 BP [20] 1.31 2.08 0.19 LSTM [21] 1.22 1.77 0.42 GRU [22] 1.15…”

Section: Experiments and Analysis Of Rolling Bearingmentioning

confidence: 99%

Degradation trend prediction of rail stripping for heavy haul railway based on multi-strategy hybrid improved pelican algorithm

Zhang,

Jiang,

Liu

et al. 2023

Intelligence & Robotics

View full text Add to dashboard Cite

As a key component of the heavy-haul railway system, the rail is prone to damages caused by harsh operating conditions. To secure a safe operation, it is of great essence to detect the damage status of the rail. However, current damage detection methods are mainly manual, so problems such as strong subjectivity, lag in providing results, and difficulty in quantifying the degree of damage are easily generated. Therefore, a new prediction method based on the improved pelican algorithm and channel attention mechanism is proposed to evaluate the stripping of heavy-haul railway rails. By processing the rail vibration acceleration, it predicts the stripping damage degree. Specifically, a comprehensive health index measuring the degree of rail stripping is first established by principal component analysis and correlation analysis to avoid the one-sidedness of a single evaluation index. Then, the convolutional bidirectional gated recursive network is trained and generalized, and the pelican algorithm, improved by multiple hybrid strategies, is used to optimize the hyperparameters in the network so as to find the optimal solution by constantly adjusting the search strategy. The squeeze-excitation channel attention module is then incorporated to re-calibrate the weights of valid features and to improve the accuracy of the model. Finally, the proposed method is tested on a specific rail stripping dataset and a public dataset of PHM2012 bearings, and the generalization and effectiveness performance of the proposed method is proved.

show abstract

“…It is also the earliest, most researched, and most widely used type of ANN model in pattern recognition and classification. The multi-layer forward network in which the back propagation algorithm is used for training is called a back propagation neural network, or BP network for short [ 15 , 16 ]. In practical application, the BP network reflects the essential part of ANN at this stage.…”

Section: Research-related Theories and Experimental Designmentioning

confidence: 99%

Monitoring Technology of Abnormal Displacement of BeiDou Power Line Based on Artificial Neural Network

Yang

Chen

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In the practice of power line engineering, navigation and positioning technology is often used in the fields of information collection and analysis, optimized line design, and deformation monitoring. Compared with traditional measurement technology, it has the characteristics of high precision and high reliability. In order to realize the measurement of abnormal displacement of power lines, improve the efficiency and quality of monitoring, and reduce the occurrence of faults, firstly, this study introduces the basic theory of artificial neural network (ANN). The core algorithm of the ANN-BP (back propagation) neural network has been improved. The improved algorithm is used to improve the BeiDou Navigation Satellite System (BDS). The improved and the unimproved BDS are used to solve the collected related data. The results show that the geometric dilution of precision (GDOP) values obtained by conventional BDS are small, all within the range of less than 4. After the introduction of the BP neural network into the system, the geometric space distribution of positioning satellites is improved, the GDOP is reduced, the reliability and availability of satellite positioning are enhanced, and the accuracy requirements are met. The accuracy of the measured data positioning results of the two systems has reached the cm level. There is not much difference between the processing results of the two modes. Among them, the Z direction accuracy has the largest difference, which is 2.5 cm. The introduction of the BP neural network has improved the spatial combination structure, and the positioning results in the three directions of X, Y, and Z are all better. From the perspective of root mean square (RMS), the RMS fluctuation of the simulation results obtained by observing the conventional BDS is large. The RMS value of BDS displacement based on the BP neural network is smaller, and the change is gentle. With the increase in the number of epochs and the increase in the number of simulations, its value is also more convergent. These data show that the quality of BDS observations based on the BP neural network is significantly better. These contents will effectively improve the monitoring accuracy and operational reliability of the system and have important practical significance and application value.

show abstract

Mechanical Performance and Microstructure Prediction of Hypereutectoid Rail Steels Based on BP Neural Networks

Cited by 10 publications

References 17 publications

Phase Prediction of High Carbon Pearlitic Steel: An Improved Model Combining Mind Evolutionary Algorithm and Neural Networks

Phase Prediction of High Carbon Pearlitic Steel: An Improved Model Combining Mind Evolutionary Algorithm and Neural Networks

Degradation trend prediction of rail stripping for heavy haul railway based on multi-strategy hybrid improved pelican algorithm

Monitoring Technology of Abnormal Displacement of BeiDou Power Line Based on Artificial Neural Network

Contact Info

Product

Resources

About