Stair-climbing posture is a common motion posture of lower extremity exoskeleton robots. In the process of climbing stairs, with the change of the gravity center of system, it has a direct impact on the dynamic stability characteristics of human-machine system. Based on this, a complete diagram of the stair-climbing posture can be set up, which can be divided into the starting stage, the intermediate stage and the ending stage. The foot trajectory, COG trajectory and ZMP trajectory of human-machine system in the intermediate stage of stair-climbing posture can be analyzed by drawing stair-climbing motion diagram of six-step. Based on the inverted pendulum theory, a virtual inverted pendulum model with variable height (VCHIPM) is proposed to analyze the influence of the change of the gravity center on the dynamic stability of the human-machine system in the upstairs posture. The test results show us that COG position can be changed by adjusting the value of a, so as to adjust the position of ZMP, reduce the error of ZMP and ensure the stability of human-machine system.

This paper aims to solve the problems related to poor motion continuity and the abrupt acceleration of a small-diameter deep well rescue robot in the process of motion characteristics analysis. According to the movement characteristics and the structural forms of the grasping mechanism and the bracket mechanism of the deep well rescue robot, the finite element analysis of the key mechanism is carried out based on the Hyperworks-Optistruct solver. According to the analysis results, the specific parameters to be optimized are obtained. Moreover, the topology optimization of the key mechanism is carried out, and the optimal design scheme of the clamping mechanism and the bracket mechanism of the deep well rescue robot are obtained. The optimization results show that, on the premise of meeting the strength requirements, the grid density distribution law is obtained. According to the variation law of the lightweight curve, the overall weight of the grasping mechanism and the bracket mechanism decreases obviously. The whole optimization process is completed, and the final optimization result is obtained.

This paper aims to unveil the symmetry–asymmetry transition mechanisms in transient fault waveforms of offshore wind power AC/DC transmission systems, addressing the critical limitation of traditional simulation methods of the fact that they cannot characterize the dynamic evolution of system symmetry, such as static impedance adjustment failing to capture transient asymmetry caused by parameter imbalance or converter control. It proposes a fault waveform simulation approach integrating mechanism analysis, scenario extraction, and model optimization. Key contributions include clarifying the quantitative links between key system parameters like submarine cable capacitance and inductance and symmetry–asymmetry characteristics, defining the transient decay rate oscillation frequency and voltage peak as core indicators to quantify symmetry breaking intensity; classifying typical fault scenarios into a symmetry-breaking type with synchronous three-phase imbalance and a persistent asymmetry type with zero-sequence and negative-sequence distortion based on symmetry evolution dynamics and revising grid-connection test indices such as lowering the low-voltage ride-through threshold and specifying the voltage type for different test objectives; and constructing a simplified embedded RLC second-order model with symmetry–asymmetry constraints to reproduce the whole process of symmetric steady state–fault symmetry breaking–recovery symmetry reconstruction. Simulation results verify the method’s effectiveness, with symmetry indicator reproduction errors ≤ 5% and asymmetric feature fitting goodness R² ≥ 0.92, which confirms that the method can effectively reveal the symmetry–asymmetry mechanisms of offshore wind power fault transients and provides reliable technical support for improving offshore wind power fault simulation accuracy and grid-connection test reliability, laying a theoretical basis for the grid-connection testing of offshore wind turbines and promoting the stable operation of offshore wind power systems.

With the growing emphasis on real-time 3D data processing in autonomous driving, robotics, and intelligent vehicles, the demand for efficient point cloud processing has expanded significantly. Early deep learning approaches to point cloud semantic segmentation relied on volumetric grids and projections, which often compromised the inherent geometric structure of point clouds. More recent methods attempt to learn directly from raw point clouds, focusing on local neighborhood information; however, optimizing computational efficiency for dynamic scenes remains a challenge. This paper presents a novel 2D-3D convolutional framework, VKPNet, for point cloud semantic segmentation that leverages Voxel Key Points (VKPs) to efficiently aggregate local features and enhance receptive fields. The proposed approach first initializes 3D point cloud features using 2D image features and applies a heuristic method to filter 3D points, extracting only those necessary for semantic segmentation, thereby reducing the input data scale. VKPs are introduced to aggregate local features at voxel cube vertices, and a 3D convolution based on VKPs is designed to expand the receptive field, facilitating effective spatiotemporal feature learning. Experimental results on the ScanNet and Semantic KITTI datasets validate the effectiveness of our VKPNet model. The framework achieves mIoU scores of 0.735 on the ScanNet dataset and 0.689 on the Semantic KITTI dataset, with a processing speed of 0.09 seconds per frame on ScanNet. These results demonstrate that VKPNet not only outperforms prior methods across various benchmarks but also achieves efficient and accurate semantic segmentation in dynamic scenes.