Jianping Yan

Neural networks have been widely applied in system dynamics modeling. However, traditional neural networks still encounter limitations in capturing long-term dynamics, nonlinear modeling, and interpretability. To address these challenges, this study proposes a novel neural network architecture, Deep Stacked State-observer based Neural Network (DSSO-NN). Firstly, the state-space representation is introduced, integrating discretized state-space equations into the neural network design to leverage both system state information and deep learning capabilities. Subsequently, two optimization measures are employed to enhance the network's nonlinear modeling ability with activation functions and the state observer, respectively. Finally, DSSO-NN is validated using the Case Western Reserve University bearing dataset. Experimental results demonstrate that activation functions provide minimal improvement to model performance. In contrast, the incorporation of the state observer significantly enhances the DSSO-NN's ability to capture system dynamics behaviors and improves modeling accuracy. DSSO-NN exhibits higher precision and greater stability, offering a novel perspective on using the state observer as an alternative to traditional activation functions.

Inspired by the architecture of natural spiderwebs, this paper introduces the three‐dimensional helical‐radial synergistic load‐bearing mechanism of spiderwebs into the design of thin‐walled multicell tubes for the first time to enhance structural crashworthiness. A theoretical prediction model for energy absorption (EA) of the spiderweb‐inspired multicell tubes (SMCT‐n) is developed, and its accuracy is validated through quasi‐static compression experiments and finite element simulations. Compared with conventional multicellular circular tubes (TMCTs), the proposed SMCTs exhibit significantly improved crashworthiness. The peak crushing force increases with the number of helical layers, with SMCT‐1, SMCT‐2, and SMCT‐3 showing increases of 17.1%, 19.8%, and 2.2%, respectively. The specific EA (SEA) also increases progressively, reaching a 35.0% improvement over the TMCT for SMCT‐1 and a further enhancement with additional helical layers. Parametric analysis further reveals that radial thickness strongly influences crashworthiness, with optimal performance achieved at a thickness of approximately 1.6 mm.

This paper introduces a novel mathematical model designed to determine combustion characteristics across the length of the combustion zone, underpinned by logical theoretical foundations and demonstrating notable alignment with experimental findings. The model facilitates the calculation of combustion efficiency based on inlet flow parameters (temperature, pressure, velocity, length, porosity, and air excess ratio), achieving efficiencies beyond conventional limits (η = 0.98–0.99) and reaching high values (η = 0.999–0.9999), which is crucial for optimizing low-emission combustors. Verification of the mathematical model was conducted through a dual approach: experimental validation and computational fluid dynamics (CFD) simulations. Comparative analysis revealed a high degree of consistency among theoretical predictions, CFD results, and experimental data, particularly for temperature and combustion efficiency distributions along the combustor length. This robust agreement affirms the model’s accuracy, reliability, and utility in advancing combustion research and developing efficient, low-emission combustors.

This study investigates a novel approach for assessing the health status of rotating machinery transmission systems by analyzing the dynamic degradation of bearings. The proposed method generates multi-dimensional data by creating virtual states and constructs a multi-dimensional model using virtual state-space in conjunction with mechanism model analysis. Innovatively, the Hammerstein–Wiener (HW) modeling technique from control theory is applied to identify these dynamic multi-dimensional models. The modeling experiments are performed, focusing on the model’s input and output types, the selection of nonlinear module estimators, the configuration of linear module transfer functions, and condition transfer. Dynamic degradation response signals are generated, and the method is validated using four widely recognized databases consisting of accurate measurement signals collected by vibration sensors. Experimental results demonstrated that the model achieved a modeling accuracy of 99% for multiple bearings under various conditions. The effectiveness of this dynamic modeling method is further confirmed through comparative experimental data and signal images. This approach offers a novel reference for evaluating the health status of transmission systems.

In recent years, research on bearing fault modeling has witnessed significant advancements. However, the modeling of bearing faults using digital twins (DTs) remains an emerging area of exploration. This paper introduces a bearing digital twin developed by integrating a signal-based response model with reinforcement learning techniques. Initially, a signal-based model is constructed, comprising a unit fault impulse function and a decay oscillation function. This model illustrates the bearing’s acceleration response under fault conditions and acts as the environmental component within the bearing digital twin. Subsequently, a parameter estimation process identifies two critical parameters from the signal-based model: the load proportional factor and the decaying constant. The Deep Deterministic Policy Gradient (DDPG) algorithm is employed as the agent for online learning of these parameters. The cosine similarity metric is employed to define the state and reward by comparing the real acceleration measurements with the simulation data generated by the digital twin. To validate the effectiveness of the digital twin, experimental data sourced from the three datasets are utilized. The results underscore the digital twin’s capacity to faithfully replicate the bearing’s acceleration response under diverse conditions, demonstrating a high degree of similarity in both the time and frequency domains.

In the machine learning and deep learning field, there are two main kinds of tasks: classification and regression. The label for the former is discrete, while for the latter is continuous. Due to the big gaps in labels, these two tasks are generally re solved separately, bringing low training efficiency and waste of computing resources. To this end, this paper proposes a new labeling method based on fuzzy membership. The main idea is to build an intermediate variable, which behaves between continuous and discrete variables. Then, the relation between the intermediate variable and the discrete label can be identified with fuzzy membership. Finally, the fuzzy membership is adopted for building labels to train the source model. After training, the source model can be transferred to achieve both classification and regression tasks. To validate the new labeling method, two typical tasks in the PHM field, aging stage classification and RUL prediction, are selected as the representative for classification and regression tasks, respectively. Furthermore, LSTM with two dense layers is chosen as the benchmark source model. With the C-MAPSS dataset, the superiority of the proposed fuzzy-membership based labeling to improve the network’s task transfer learning performance has been verified.

Convolutional Neural Network (CNN) has been extensively used in bearing fault diagnosis and Remaining Useful Life (RUL) prediction. However, accompanied by CNN’s increasing performance is a deeper network structure and growing parameter size. This prevents it from being deployed in industrial applications with limited computation resources. To this end, this paper proposed a two-step method to build a cell-based light CNN by Neural Architecture Search (NAS) and weights-ranking-based model pruning. In the first step, a cell-based CNN was constructed with searched optimal cells and the number of stacking cells was limited to reduce the network size after influence analysis. To search for the optimal cells, a base CNN model with stacking cells was initially built, and Differentiable Architecture Search was adopted after continuous relaxation. In the second step, the connections in the built cell-based CNN were further reduced by weights-ranking-based pruning. Experiment data from the Case Western Reserve University was used for validation under the task of fault classification. Results showed that the CNN with only two cells achieved a test accuracy of 99.969% and kept at 99.968% even if 50% connections were removed. Furthermore, compared with base CNN, the parameter size of the 2-cells CNN was reduced from 9.677MB to 0.197MB. Finally, after minor revision, the network structure was adapted to achieve bearing RUL prediction and validated with the PRONOSTIA test data. Both tasks confirmed the feasibility and superiority of constructing a light cell-based CNN with NAS and pruning, which laid the potential to realize a light CNN in embedded systems.