In order to solve the problems in fatigue reliability analysis of mechanical components, a new learning function based on the Kriging model is proposed. In the process of fatigue reliability analysis, surrogate model is often used to fit the implicit performance function to avoid large numbers of calculations of fatigue crack samples. The existing learning functions ignore the information of probability density function (PDF). To overcome this defect and avoid unnecessary sampling in low PDF regions, a novel learning function takes into account the PDF and the local accuracy of Kriging model. The accuracy of the Kriging model is improved by adding samples step by step, and the new training samples are determined by the proposed learning function. The proposed method is verified by two examples from literatures. The results show that, compared with other surrogate models and learning functions, the proposed method has advantages in efficiency, convergence and accuracy. Finally, the proposed method is employed to calculate the fracture failure probability of cracked structures.
Efficient and accurate prediction of tool Remaining Useful Life (RUL) is the key to improve product accuracy, improve work efficiency and reduce machining costs. Aiming at the problems of weak tool wear state features, difficult extraction, and low prediction precision and accuracy, this study proposes a CNN-LSTM-PSO tool remaining life prediction method based on multi-channel feature fusion.Firstly, based on computer vision, feature extraction, information fusion technology, the multi-source sensor signals collected during the tool life cycle are effectively processed and analyzed, and a sample data set of spatio-temporal correlation of traffic flow is constructed. Secondly, the sample data set was input into the CNN-LSTM-PSO model, the CNN network obtained the sequence feature vector by extracting the spatial characteristics of traffic flow data, and the feature vector was input into the multi-layer LSTM network to extract the time-dependent features, and the PSO algorithm optimized the hyperparameters in the CNN-LSTM model. The accuracy of tool RUL prediction model and the efficiency of model fitting are further improved. The results show that the CNN-LSTM-PSO model can effectively predict tool wear, with the mean absolute error (MAE) value of 1.0892, the root mean square error (RMSE) value of 1.3520, and the determination coefficient R 2 value of 0.9961; Through the comparative analysis of ablation experiments, it is found that the method proposed in this study has the highest efficiency in fitting the tool RUL prediction model, the lowest values of MAE value and root mean square error RMSE, and the value of determination coefficient R 2 is closest to 1, which has certain advantages.The proposed method has reference value and engineering practical significance for the related research of tool wear residual life prediction.
As the core component of the automobile braking system, brake pads have a complex structure and high failure rate. Their accurate and effective state monitoring can help to evaluate the safety performance of brake pads and avoid accidents caused by brake failure. The wear process of automobile brake pads is a gradual, nonlinear, and non-stationary time-varying system, and it is difficult to extract its features. Therefore, this paper proposes a CNN-LSTM brake pad wear state monitoring method. This method uses a Convolutional Neural Network (CNN) to complete the deep mining of brake pad wear characteristics and realize data dimensionality reduction, and a Long Short-Term Memory (LSTM) network to capture the time dependence of the brake pad wear sequence, so as to construct the nonlinear mapping relationship between brake pad wear characteristics and brake pad wear values. At the same time, the artificial Gorilla Troops Optimization (GTO) algorithm is used to perform multi-objective optimization of the network structure parameters in the CNN-LSTM model, and its powerful global search ability improves the monitoring effect of the brake pad wear status. The results show that the CNN-LSTM model based on GTO multi-objective optimization can effectively monitor the wear state of brake pads, and its coefficient of determination R2 value is 0.9944, the root mean square error RMSE value is 0.0023, and the mean absolute error MAE value is 0.0017. Compared with the BP model, CNN model, LSTM model, and CNN-LSTM model, the value of the coefficient of determination R2 is the closest to 1, which is increased by 8.29%, 5.52%, 4.47%, 3.30%, respectively, which can more effectively realize the monitoring and intelligent early warning of the brake pad wear state.
29 March 2018 in Advances in Mechanical Engineering
As the numerical model of engineering structure becomes more and more complicated and time consuming, efficient structural reliability analysis is badly in need. To reduce the number of calls to the performance function and iterative times during structural reliability analysis, an innovative strategy of the design of experiments (DoE) called Isomap-Clustering strategy is proposed. According to the statistical information provided by Kriging, points with the worst uncertainty for reliability analysis are on the estimated limit state. Therefore, by combining Isomap and k-means clustering algorithm, Isomap-Clustering strategy refreshes the DoE of the Kriging model with a few representative points in the vicinity of the estimated limit state each iteration and iteratively “pushes” the estimated limit state to the real one until a stopping condition is satisfied. By employing the proposed DoE strategy and sparse polynomial-Kriging model, a structural reliability analysis method is constructed, whose stopping criterion is defined by derivation. Three examples are studied. Results show that the proposed method can lower the number of calls to the performance function and remarkably reduces the iterations of structural reliability analysis.
31 October 2002 in Fatigue & Fracture of Engineering Materials & Structures
The resonance of the gear-rotor system will produce a large number of responses that do not exceed the threshold value, resulting in structural fatigue failure and transmission failure, affecting its life and reliability. It is particularly critical to consider the temperature rise under high-speed and heavy-load conditions. Therefore, the research will take the main drive gear-rotor system of a certain type of aeroengine accessory gearbox as the research object, consider the influence of the temperature field on the natural frequency of the gear-rotor system, and take the difference between the natural frequency of the gear-rotor system and the excitation frequency (gear meshing frequency) as the performance function. The PC-Kriging and adaptive design of experimental strategies are applied to the thermal-structural coupling parametric model to analyze the resonance reliability and sensitivity of the gear-rotor system. For complex mechanical mechanisms, the method has better accuracy than other surrogate models and greatly saves the time of finite element simulation in reliability analysis. The results show that the natural frequency of a gear rotor decreases with an increase in temperature, and the natural frequency of different orders varies with the change in temperature. The influence of the sensitivity of different random parameters on the resonance reliability of the gear-rotor system is obtained. Reliability research on resonance failure of high-speed and heavy-load aviation gear-rotor systems considering random parameters under a temperature rise field has important practical engineering application value and scientific research significance.