Life Prediction Method Research Articles

Path Graph Attention Network-based Bearing Remaining Useful Life Prediction Method

Path Graph Attention Network-based Bearing Remaining Useful Life Prediction Method

Fatigue damage calculation of cold recycled asphalt pavement considering measured temperature field and traffic volume distribution

<abstract> <p>A finite element simulation method for fatigue damage calculation and life prediction of pavement structures under the influence of temperature and traffic distribution factors was proposed in this study. Relying on the test cold recycled asphalt pavement structure, the existing distress, pavement structure forms, field monitored temperature and strain were first introduced and analyzed. Then, in constructing the numerical model, the viscoelastic constitutive model was introduced to characterize the effect of temperature and loading conditions on the mechanical response of the hot mix asphalt (HMA) layers and the emulsified asphalt cold recycled (EACR) layer. The damage variables are defined by fatigue equations, and the damage accumulation can be determined by Miner's linear fatigue accumulation theory. To reflect the distribution of traffic volume, the total traffic volume of a year was divided into 144 axle load groups according to the monthly and hourly distribution conditions. Accordingly, based on the monthly maximum and minimum temperature, 12 representative days were selected to represent the climate characteristics of 12 months, respectively. Then, each representative day's measured structural temperature data were extracted every 2 hours and linearly interpolated to obtain 144 representative temperature fields corresponding to 144 axle load groups. Through the above method, simulation calculations were performed for cold recycled asphalt pavement structures with different cement-stabilized aggregate (CSA) base stiffnesses. The results show that the fatigue damage accumulation of the EACR layers reaches its highest value in winter and midday hours, owing to the temperature variation and traffic distribution. Due to the weak fatigue resistance of EACR mixtures, it is not recommended to be paved EACR layers directly on top of the CSA base with poor bearing capacity. Otherwise, fatigue cracking is likely to occur first. For this reason, recommendations for ensuring the durability of the cold recycled pavement structure were also proposed in the study.</p> </abstract>

Remaining Useful Life Prediction With Partial Sensor Malfunctions Using Deep Adversarial Networks

In recent years, intelligent data-driven prognostic methods have been successfully developed, and good machinery health assessment performance has been achieved through explorations of data from multiple sensors. However, existing data-fusion prognostic approaches generally rely on the data availability of all sensors, and are vulnerable to potential sensor malfunctions, which are likely to occur in real industries especially for machines in harsh operating environments. In this paper, a deep learning-based remaining useful life (RUL) prediction method is proposed to address the sensor malfunction problem. A global feature extraction scheme is adopted to fully exploit information of different sensors. Adversarial learning is further introduced to extract generalized sensor-invariant features. Through explorations of both global and shared features, promising and robust RUL prediction performance can be achieved by the proposed method in the testing scenarios with sensor malfunctions. The experimental results suggest the proposed approach is well suited for real industrial applications.

Remaining Useful Life Prediction for Circuit Breaker Based on SM-CFE and SA-BiLSTM

Targeting at the uncertainty of the degradation of conventional circuit breakers (CCBs) and the perfect mechanical degradation characterization by vibration signals, a remaining useful life (RUL) prediction method based on similarity measure (SM), comprehensive features extraction (CFE) and self-attention bidirectional long short-term memory network (SA-BiLSTM) is proposed in this paper. Firstly, the degradation-sensitive intrinsic mode functions (IMFs) of the vibration signals are separated by the SM method, and the reconstructed signals are divided into intervals based on the working process of CCBs to obtain the vibration signals for life prediction. Secondly, the explicit parametric features (EPFs) are obtained based on the vibration events and waveform features, and the implicit parametric features (IPFs) are extracted using the multi-scale convolutional auto-encoder (MSCAE). Next, the comprehensive features are obtained by combination of the EPFs and IPFs. On this basis, the degradation mode judgment method is formed by using the least square method (LSM) linear fitting of the EPFs. Finally, a quantitative life prediction model based on SA-BiLSTM is constructed, and greater weights are assigned to the important time steps. The proposed method is proved to show high prediction accuracy and good stability, which is more advantageous compared with other hybrid prediction models.

Graph Structure and Temporal Data Driven Remaining Useful Life Prediction Method for Machinery

Graph Structure and Temporal Data Driven Remaining Useful Life Prediction Method for Machinery

Research on Crack Propagation and Remaining Life Prediction Method of Surface Strengthened Axle

Research on Crack Propagation and Remaining Life Prediction Method of Surface Strengthened Axle

Research on Tool Remaining Life Prediction Method Based on CNN-LSTM-PSO

Research on Tool Remaining Life Prediction Method Based on CNN-LSTM-PSO

Bayesian-based durability life prediction method of nano-modified concrete

Bayesian-based durability life prediction method of nano-modified concrete

A Remaining Useful Life Prediction Method for Lithium-ion Battery Based on Temporal Transformer Network

The remaining useful life prediction is significant for Lithium-ion batteries to ensure safety and reliability. Due to the advantages of handling time sequence data, recurrent neural network based methods have achieved impressive performance on RUL prediction. However, most of these methods develop the RUL prediction model without considering the operating time of the battery, which is an important factor on capacity degradation. Therefore, this paper proposed a Temporal Transformer Network (TTN) for RUL of Lithium-ion batteries. The proposed method combines the self-attention mechanism of the Transformer Network with Denoising Autoencoder to implement the noise of raw data. More importantly, the proposed method designs a temporal encoding layer to introduce the operating time to the input of RUL prediction model. The performance of the proposed method is evaluated on two frequently used battery datasets. The proposed method achieves the best result compared with other frequently used RUL prediction methods.

Remaining Useful Life Prediction Method of Coated Spherical PlainBearing Based on VMD-EEMD-LSTM

Remaining Useful Life Prediction Method of Coated Spherical PlainBearing Based on VMD-EEMD-LSTM

Bayesian based durability life prediction method of nano-modified concrete

Bayesian based durability life prediction method of nano-modified concrete

Residual life prediction of thermal insulation material for cold chain logistics transportation vehicle

Aiming at the problems of low prediction accuracy and long prediction time cost in traditional methods, a residual life prediction method of thermal insulation material for cold chain logistics transportation vehicle is proposed. The distribution position of thermal insulation material in the car body is analysed. According to the distribution position of thermal insulation materials, the life prediction parameters of thermal insulation material are calculated. According to the determined parameters, the key degree of thermal insulation material is calculated by gradient descent method. So as to construct the residual life prediction model of cold chain logistics transport vehicle compartment thermal insulation material, the key parameters of thermal insulation material is taken as the input, and the residual life prediction results of compartment thermal insulation material as the output. Experimental results show that the prediction accuracy of the proposed method is about 97% when 100 times, and the shortest time cost is about 0.5 s.

A multi-source information fusion method for tool life prediction based on CNN-SVM

For milling tool life prediction and health management, accurate extraction and dimensionality reduction of its tool wear features are the key to reduce prediction errors. In this paper, we adopt multi-source information fusion technology to extract and fuse the features of cutting vibration signal, cutting force signal and acoustic emission signal in time domain, frequency domain and time-frequency domain, and downscale the sample features by Pearson correlation coefficient to construct a sample data set; then we propose a tool life prediction model based on CNN-SVM optimized by genetic algorithm (GA), which uses CNN convolutional neural network as the feature learner and SVM support vector machine as the trainer for regression prediction. The results show that the improved model in this paper can effectively predict the tool life with better generalization ability, faster network fitting, and 99.85% prediction accuracy. And compared with the BP model, CNN model, SVM model and CNN-SVM model, the performance of the coefficient of determination R2 metric improved by 4.88%, 2.96%, 2.53% and 1.34%, respectively.

An encoder-decoder fusion battery life prediction method based on Gaussian process regression and improvement

The prediction ability of all traditional machine learning models is limited to a few batteries. When the RUL of more batteries needs to be predicted, the prediction performance of traditional machine learning is not very good. For long-term battery capacity prediction, it is obtained in a recursive way. However, adding the predicted value to the input sequence will cause a large cumulative error after multiple recursive predictions. Based on the codec model, this paper introduces Gaussian process regression, which is used in multi-step prediction to improve the codec fusion method and the Savitzky-Golay method is utilized to smooth the training set. A new kernel function was designed to further improve the accuracy. The method of dynamic weights is adopted to minimize accumulated error. The experimental results show that the fusion prediction method can reduce the cumulative error of recursive prediction without losing the declining trend of the actual capacity of the battery, and can predict the RUL of zinc-ion batteries more precisely than other models.

Remaining Useful Life Prediction for Lithium-Ion Batteries Based on Improved Variational Mode Decomposition and Machine Learning Algorithm

Remaining useful life (RUL) prediction of batteries is important for the health management and safety evaluation of lithium-ion batteries. Because lithium-ion batteries have capacity recovery and noise interference during actual use, direct use of measured capacity data to predict their RUL generalization ability is not efficient. Aimed at the above problems, this paper proposes an integrated life prediction method for lithium-ion batteries by combining improved variational mode decomposition (VMD) with a long short-term memory network (LSTM) and Gaussian process regression algorithm (GPR). First, the VMD algorithm decomposed the measured capacity dataset of the lithium-ion battery into a residual component and capacity regeneration component, in which the penalty factor α and mode number K in the VMD algorithm were optimized by the whale optimization algorithm (WOA). Second, the LSTM and GPR models were established to predict the residual component and capacity regeneration components, respectively. Last, the predicted components are integrated to obtain the final predicted lithium-ion battery capacity. The experimental results show that the mean absolute error (MAE) and root mean square error (RMSE) of the proposed lithium-ion battery capacity prediction model are less than 0.5% and 0.8%, respectively, and the method outperforms the five compared algorithms and several recently proposed hybrid algorithms in terms of prediction accuracy.

Fatigue life prediction for Al–Zn–Mg alloy butt welded joints based on a fatigue damage model accounting for the influence of interior porosity

AbstractThis paper proposed a high‐cycle fatigue life prediction method for Al–Zn–Mg alloy welded joints considering the thermal process of welding and the effect of welding pores. First, the welding process was simulated by a numerical thermal–mechanical analysis. The residual stresses were then obtained and applied as the initial stress field of the subsequent fatigue life calculation. Second, a fatigue damage model was constructed by introducing a dimensionless quantity to equivalently describe the effect of interior pores. Further, a corresponding damage mechanics–finite element numerical method was numerically implemented by the ABAQUS platform to calculate the high‐cycle fatigue life for Al–Zn–Mg alloy welded joints. The calculated lives have good consistency with test data, which validates the feasibility and applicability of the proposed model and the calculation method. Finally, the curves of the cyclic stress versus fatigue life of Al–Zn–Mg alloy welded joints with different interior pore sizes were calculated to explore the influence law of the welding pore.

Residual Life Prediction of XLPE Distribution Cables Based on Time-Temperature Superposition Principle by Non-Destructive BIS Measuring on Site.

Crosslinked polyethylene (XLPE) distribution cables are prone to segmented thermal aging after long-term operation owing to the large spatial spans and complex operating environments, and accurate residual life prediction of each aging cable segment could provide a theoretical basis and reference for performance monitoring, maintenance and the replacement of cables. Existing studies mainly focus on the residual life prediction methods for uniform aging cables, which are not suitable for segmented-aging cables. In this paper, a residual life prediction method for segmented-aging XLPE distribution cables based on the time-temperature superposition principle (TTSP) by non-destructive BIS measuring on site was proposed. Firstly, the applicability of the TTSP in the transformation of the changing process of elongation at break (EAB) of XLPE at different thermal aging temperatures was verified based on the Arrhenius equation. Secondly, to better simulate the thermal aging process under working conditions, XLPE cables were subjected to accelerated external stress aging at 140 °C for different aging times, and the corresponding changing process of EAB along with aging time was further measured. The relationship between the EAB of XLPE cables and aging time was well fitted by an equation, which could be used as a reference curve to predict the thermal aging trends and residual life of service-aged XLPE cables. After that, a calculation method for the transformation of the changing process of EAB of XLPE at different thermal aging temperatures was proposed, in which the corresponding multiplicative shift factor could be obtained based on the TTSP instead of the Arrhenius equation extrapolation. Moreover, the availability of the above calculation method was further proved by accelerated thermal aging experiments at 154 °C; the results show that the prediction error for the cable's EAB is no more than 3.15% and the prediction error for residual life is within 10% in this case. Finally, the realization of non-destructive residual life prediction combined with BIS measuring on site was explained briefly.

A hierarchical mechanism-informed neural network approach for assessing fretting fatigue of dovetail joints

In the present paper, a hierarchical mechanism-informed neural network (HMNN) life prediction method was proposed. The fretting fatigue was decomposed into different fatigue problems and considered in four neural network layers, which were hierarchically and progressively established for proportional multi-axial fatigue, non-proportional multi-axial fatigue, notch fatigue and fretting fatigue, respectively. Each layer can be used to assess the fatigue life of the previous layer based on the progressive construction of fatigue complexity. The HMNN approach can predict all kinds of fatigue implemented in the method with reasonable accuracy and provides a new approach for complex fatigue assessment.

A new multiaxial fatigue life prediction method based on grey theory under small sample condition

AbstractIt is always a challenge to accurately predict the fatigue life of specimens in engineering, and it is urgent to establish a convenient and accurate approach to estimate the fatigue life of specimens with different materials under small sample conditions. To solve the above problems, a new multiaxial fatigue life prediction method is proposed based on grey theory. First, established a modified MGM (1, N) model by fine‐adjusting the time series based on the GM (1, N) model. Then the MGM (1, N) model is used to compare the results with different types of parameters, and the optimal gray prediction model is established based on the gray correlation analysis. The proposed method is verified by 79 test data of 10 kinds of materials including Q355(D), and the predicted results matched well with the experimental results. It displays that the method is reliable. Finally, the predicted results with an accuracy of more than 90% were fused with the test data to obtain a more accurate S‐N field and ε‐N field of Q355(D) based on small sample by the ProFatigue program.

Thermodynamics-based method considering orientation and notch effect to predict the high cycle fatigue life of a nickel-based single crystal superalloy

The high cycle fatigue (HCF) life of DD6 at 760 ℃ was studied by performing a large number of experiments. A new thermodynamics-based life prediction method was proposed by considering the thermodynamic entropy change during the fatigue process. The model is not only suitable for smooth specimens but also can be applied to the life evaluation of notched specimens and those with different orientations. Finally, the life of an aeroengine blade with different orientations was qualitatively evaluated by the proposed model, which demonstrated that the model can be used to analyze the influence of crystal-axis deflection angle on blade life.