Deep transfer learning provides a feasible fault diagnosis method for intelligent mechanical systems. However, this method usually assumes that the source domain and the target domain have the same label space, which greatly limits its application in the actual industry. Therefore, this article proposes a multidiscriminator deep weighted adversarial network (MDWAN) method, which is especially suitable for partial transfer learning where the number of target domain categories is less than the source domain. The proposed method is mainly composed of three parts: feature extractor, multidiscriminator, and classifier. In the feature extraction section, a deep separable convolutional neural network model (DSC) is proposed, which can greatly reduce parameter calculation amount under the premise of ensuring the extraction accuracy. In the multidiscriminator section, a multidiscriminator weighted learning strategy is proposed. This strategy comprehensively considers the domain information and source domain label information, and introduces a weight function to quantify the contribution of source domain samples to the domain discriminator and classifier, which can effectively identify and filter outlier source samples to promote the positive transfer of shared samples. In order to verify the effectiveness and feasibility of the proposed method, the method is applied to three types of bearing datasets of different machines. Comparing the classification results of different methods, the conclusion shows that this method is more beneficial for bearing fault classification.

In high-temperature environments, the signal-to-noise ratio (SNR) of the signal measured by electromagnetic acoustic transducers (EMAT) is low, and the signal characteristics are difficult to extract, which greatly affects their application in practical industry. Aiming at this problem, this paper proposes the least mean square adaptive filtering interpolation denoising method based on variational modal decomposition (AFIV). Firstly, the high-temperature EMAT signal was decomposed by variational modal decomposition (VMD). Then the high-frequency and low-frequency noises in the signal were filtered according to the excitation center frequency. Following the wavelet threshold denoising (WTD) for the noise component after VMD decomposition was carried out. Afterward, the noise component and signal component were connected by an adaptive filtering process to achieve further noise reduction. Finally, cubic spline interpolation was used to smooth the noise reduction curve and obtain the time information. To verify the effectiveness of the proposed method, it was applied to two kinds of ultrasonic signals from 25 to 700 °C. Compared with VMD, WTD, and empirical mode decomposition denoising, the SNR was increased by 2 times. The results show that this method can better extract the effective information of echo signals and realize the online thickness measurement at high temperature.

The step-stress accelerated degradation test (ADT) provides a feasible method for assessing the storage life of high-reliability, long-life products. However, this method results in a slower rate of performance degradation at the beginning of the test, significantly reducing the test efficiency. Therefore, this article proposes an assessment method for the step-down stress ADT that considers random effects and detection errors (SDRD). Firstly, a new Inverse Gaussian (IG) model is proposed. The model introduces the Gamma distribution to characterize the randomness of the product degradation path and uses the normal distribution to describe the detection errors of performance parameters. In addition, to solve the problem that the likelihood function of the IG model is complex and has no explicit expression, the Monte Carlo (MC) method is used to estimate unknown parameters of the model. This approach enhances computational accuracy and efficiency. Finally, to verify the effectiveness of the SDRD method, it is applied to the step-down stress ADT data from a specific missile tank to assess its storage life. Comparing the life assessment results of different methods, the conclusion shows that the SDRD method is more effective for assessing the storage life of high-reliability, long-life products.

Accelerated life test provides a viable method for assessing the life of complex electromechanical products. However, it is difficult to directly assess their storage life using the failure count for complex electromechanical products with few or no failures during storage life tests. Therefore, based on the step-stress accelerated degradation test, this article proposes a new storage life assessment method (RDSA), which is particularly suitable for the storage life assessment of complex electromechanical products based on performance degradation. Firstly, this article proposes a performance degradation model that comprehensively considers the heterogeneity between units and detection errors. This model is based on the Inverse Gaussian (IG) model and introduces unit-specific random effects to capture the heterogeneity and monotonicity of product degradation paths. Additionally, it uses a normal distribution to address the challenges posed by detection errors in data analysis, thereby enhancing the accuracy of life assessment for the product. Secondly, a Quasi-Monte Carlo method (QMC) based on distribution transformation is proposed to estimate the model parameters and reliability. This method overcomes the limitations of traditional Monte Carlo methods (MC), and can significantly reduce errors, decrease computation, and accelerate convergence. Finally, the proposed method is applied to performance degradation data of the electromechanical fuzes to verify its robustness and applicability. By comparing the proposed method with other more advanced methods, the results indicate that this method is more suitable for the storage life assessment of complex electromechanical products.

In the complex urban path planning of unmanned ground vehicles (UGVs), the dung beetle optimization (DBO) algorithm is widely used due to its simple structure and fast convergence speed. However, it still has the disadvantages of poor convergence accuracy and is easy to fall into a local optimum. To solve these problems, this paper proposes a multi-strategy enhanced DBO algorithm (ESDBO). Firstly, sine mapping is introduced in the population initialization stage to enhance solution diversity. Secondly, an adaptive information volatilization mutation strategy is proposed, which dynamically balances the convergence and global search ability. Finally, a multi-mechanism co-evolution strategy is designed, which significantly improves the local search ability and stability. Through ablation experiments and CEC2017 benchmark tests, the optimization ability of the proposed strategy and the convergence accuracy and stability of ESDBO are verified. Further path planning experiments are carried out on the public Random MAPF benchmark map. The results show that ESDBO can generate global optimal paths with short path length, few turns, and high safety margin on different obstacle densities and map scales. The algorithm provides an efficient and reliable solution for autonomous navigation in complex urban environments.