With the ongoing advancements in material technology, the domain of soft robotics has garnered increasing attention. Soft robots, in contrast to their rigid counterparts, offer superior adaptability to the environment, enhanced flexibility, and improved safety, rendering them highly suitable for complex application scenarios such as rescue operations and medical interventions. In this paper, a new type of pneumatic software actuator is proposed. The actuator adopts a combination of a soft structure and pneumatic control, which is highly flexible and versatile. By using the flow of gas inside the soft structure, high-precision and flexible motion control is realized. In the design process, the extensibility and adaptability of the structure are considered, so that the actuator can adapt to different working environments and task requirements. The experimental results show that the pneumatic soft actuator exhibits excellent performance in terms of accuracy, response speed, and controllability. This research provides new ideas and methods for the development of the field of pneumatic actuators and has wide application prospects. The main research content of this paper is as follows: first, the soft pneumatic actuator is modeled and simulated, the structure is optimized on the basis of simulation, and finally, the performance of the actuator is tested.
Electric motors are the principal power units of industry, and their failures cause cascading stoppages of process equipment. The vibration signal carries the most complete information about the condition of rotating assemblies, yet machine-learning diagnostics is constrained by the shortage of labelled fault data. This work proposes and verifies a methodology for generating a physically grounded synthetic dataset of vibration signals: analytical models of imbalance, misalignment, rolling-bearing defects (BPFO, BPFI, BSF, FTF) and mechanical looseness produced 8000 one-second fragments of vibration acceleration covering eight classes of technical state, with variation of the rotational frequency (±5%), signal-to-noise ratio (5–30 dB) and defect amplitude. The synthetic signals were verified against reference Case Western Reserve University (CWRU) records: the outer- and inner-ring defect frequencies agree within 0.22%, and the envelope spectra are class-specific for these defects. Six classifiers were compared in three stages: Random Forest, XGBoost, a 1D-CNN (WDCNN), two 2D-CNNs on STFT and CWT representations, and a PatchTST transformer. Under the full set of variations, models reached a Macro F1 of 99.6–99.8%. On the real CWRU records, separate models were trained on the dataset based on real data from other operating loads. On the unseen loads, the Random Forest reached a Macro F1 of 95.5% and the 1D-CNN 93.2%, against 21–42% for the same models trained on synthetic data alone, and with one real fragment per class, the difference between the two training datasets was already 17–32 p.p. High accuracy on synthetic data therefore did not carry over to zero-shot diagnosis on the real bench.