The stage division and state recognition during AC contactor degradation process is an important prerequisite for realizing its self-perception. In the recognition of its degradation state, the traditional method cannot effectively identify similar and overlapping degradation states, which makes it impossible to make accurate judgments when evaluating the health state of AC contactor. To solve the above problems, a method of ac contactor’s contact degradation stage division and state recognition based on Boundary Detection and Temporal Convolutional Network-Transformer-Bidirectional Gated Recurrent Unit (TCN-Transformer-BiGRU) was proposed in this paper. Firstly, the characteristic parameters related to the degradation of the ac contactor were obtained through the full life test, and the kernel principal component analysis (KPCA) was introduced to fuse the characteristic parameters. Then, the degradation trend of the contact system was characterized, and the Boundary Detection method was used to divide the ac contactor degradation stage. Finally, the TCN-Transformer-BiGRU classification prediction model was used to accurately identify the degradation state of the ac contactor. Taking other samples of the same type of AC contactor as examples, it is verified that the method has good universality and high accuracy.

Aiming at the problem that it is difficult to extract useful information from the contactor characteristic parameter data due to a large amount of noise interference, and it is difficult to determine the weight allocation of different characteristic parameters in the process of contactor degradation state characterisation, this paper proposes a contactor degradation state characterisation method based on the improved variational modal decomposition and the optimisation of similarity metrics with combined weights. Firstly, the CJX2-5011 model contactor is taken as the experimental object, and the full-life test is carried out to extract the contactor characteristic parameter data; Then, the sparrow search algorithm (SSA) is used to optimise the parameter settings of the variational modal decomposition, and the optimal component selection and noise reduction are completed based on the correlation determination and wavelet threshold noise reduction, which effectively extracts the useful information of the feature parameters; Secondly, The combined weights of feature parameters are constructed by Analytic Hierarchy Process (AHP) and CRITIC weighting method, which solves the problem of weight allocation of feature parameters in the process of characterising the degraded state of the contactor; Finally, the combination of weights is used to optimise the similarity measurement process and achieve a more accurate characterisation of the contactor degradation state. In the multi-characteristic parameter noise reduction process, the Noise Rejection Rate ( NRR ) and Correlation (ρ) are at least improved by 0.0592 and 0.0006. In the degraded state characterisation, the Goodness of Fit ( R ) is improved by at least 0.06723, and the Root Mean Square Error ( RMSE ) is reduced by at least 0.34689. Therefore, it can be seen that the method in this paper is able to better complete the noise reduction of the contactor characteristic parameter data, and to achieve a more accurate characterisation of the degraded state of the contactor.

In this paper, we address several issues with existing methods for predicting the residual electrical life of railway relays. These issues include the difficulty of single feature prediction in characterizing the degradation process, the neglect of temporal and backward–forward correlations in the degradation process, and low prediction accuracy. To overcome these challenges, we propose a novel approach that combines convolutional neural networks (CNNs) and bidirectional long short-term memory (BiLSTM) to facilitate the life prediction of railway relays and provide an accurate data basis for the maintenance of railway relays. Firstly, we collected voltage and current signals from railway relay electrical life tests and extracted feature parameters that captured the relay’s operating state. Next, we applied Spearman correlation coefficient analysis combined with random forest importance analysis to perform double-feature selection. This process eliminates redundant feature parameters and identifies the optimal feature subset. Finally, we constructed a convolutional neural network bidirectional long short-term memory (CNN BiLSTM) prediction model to accurately predict the remaining electrical life of the railway relay. Through our analysis of the prediction results, we observed that the CNN BiLSTM model achieves an effective prediction accuracy of 96.3%. This accuracy is significantly higher, more stable, and more practical compared to other prediction models such as recurrent neural networks (RNNs), long short-term memory (LSTM), and BiLSTM models. Overall, our proposed CNN BiLSTM model offers higher accuracy, better stability, and greater practicality in predicting the remaining electrical life of railway relays.

Aiming at the problem that there is a lot of noise in the feature parameter data of low-voltage switchgear, which makes it difficult to extract useful information about the feature parameters and unable to construct the performance degradation index reasonably. This paper proposes a noise reduction method through Relative Entropy (KL) optimized Variational Mode Decomposition (VMD) combined with Non-Local Mean (NLM) and a method of constructing performance degradation indexes for low-voltage switching appliances through Convolutional Autoencoder (CAE). Firstly, the voltage and current signals are collected based on the full-life test platform of low-voltage switchgear, and the subset of feature parameters is extracted. Then, the feature parameters are decomposed into different components by relative entropy optimization VMD and the noise-dominant and effective components are selected based on the correlation. Secondly, NLM denoising is performed on the noise-dominated signal, and the noise-reduced signals are obtained by signal reconstruction. Finally, the construction of performance degradation indexes for multi-type switchgear is completed based on CAE. The final results show that the two noise reduction evaluation indicators including Noise Rejection Ratio (NRR) and trendiness (T), and the three performance degradation evaluation indexes including Goodness of Fit (R), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) all show optimal results, indicating that this paper’s method can better complete the construction of noise reduction and performance degradation indicators for the feature parameters of low-voltage switchgear.

Accurate fault identification of AC contactors is challenging due to low signal utilization, weak feature representation, and the limited robustness of existing diagnostic methods. To address these issues, this study proposes a multimodal fault diagnosis approach that integrates enhanced time-frequency features with temporal vibration characteristics. A full-life experimental platform was established to synchronously acquire vibration, voltage, and current signals, enabling the construction of five typical fault categories based on electrical-parameter thresholds. To improve the measurability of vibration features, a multiscale short-time Fourier transform (STFT) combined with a squeeze-and-excitation block (SE Block) was used to generate noise-suppressed, contrast-enhanced time-frequency maps. A dual-branch network was then designed, consisting of a convolutional block attention module (CBAM)-optimized ResNet-18 for spatial feature extraction and a bidirectional gated recurrent unit (BiGRU) module for temporal modeling. The fused multimodal features were used for comprehensive fault classification. Experimental results show that the proposed method achieves an accuracy of 98.6%, significantly outperforming single-modality models. The results demonstrate that multimodal fusion effectively enhances fault feature discrimination and improves the robustness of AC contactor condition monitoring.