Yining Zhang

Signals generated during tool wear are nonlinear, non-stationary, and easily affected by machining noise, which makes reliable tool condition monitoring difficult in intelligent manufacturing. To address this issue, this study proposes a tool wear degree classification framework, FCTrans-CA, that fuses time-domain and frequency-domain information through a lightweight cross-attention (CA) bridge. Fast Fourier transform (FFT) is first used to obtain frequency-domain representations. The raw time-domain signals are processed by a multi-scale one-dimensional convolutional neural network (MS-CNN) to extract temporal wear features, while the FFT-derived representations provide complementary spectral cues. These two feature streams are fused by an asymmetric CA module in which frequency-domain features guide the selection of wear-sensitive temporal features. K-means clustering is used to divide the measured flank wear (VB) trajectory of each tool into initial-, normal-, and severe-wear stages, thereby reducing subjectivity in label generation. Experiments on the PHM2010 milling dataset show that FCTrans-CA achieves 99.43% classification accuracy on 40,648 test samples. The results indicate that cross-domain feature interaction improves the separability of wear states and provides a reproducible data-driven route for tool wear monitoring.

With the growing demand for precise and efficient control of robotic arms in the era of Industry 4.0, traditional methods for solving the inverse kinematics problem face significant limitations in terms of accuracy, computational speed, and adaptability to complex robotic configurations. This paper proposes a novel approach for solving the inverse kinematics problem by integrating the Improved Dung Beetle Optimization (IDBO) algorithm with Backpropagation Neural Networks (BPNN). This hybrid method is applied to Newton‐Raphson (NR) iterative algorithms for computing the kinematic solutions of robotic arms, effectively enhancing both optimization efficiency and solution accuracy. The IDBO algorithm, an advanced version of the traditional Dung Beetle Optimization (DBO), incorporates innovative strategies that improve convergence speed and balance local and global search capabilities, making it an effective tool for optimizing the weights and biases of the neural network. As a case study, the UR5e robotic arm is modeled using the Denavit‐Hartenberg convention. The proposed IDBO‐BPNN method is benchmarked against traditional and other optimization algorithms through simulations, demonstrating superior performance in terms of convergence speed, solution accuracy, and computational stability. Notably, the IDBO‐BPNN‐NR approach significantly reduces computation time, achieving an 80.6% reduction compared to the Random Iteration Point‐NR method and a 66.6% reduction compared to the Fixed Starting Point‐NR method. By comparing the solution parameters obtained using the IDBO‐BPNN‐NR algorithm and the Fixed Starting Point‐NR algorithm across robotic arms with varying degrees of freedom and structural configurations, the robust generalization capability of the proposed method is further validated. The results indicate that this hybrid approach is highly suitable for real‐time robotic applications, offering a scalable, efficient, and accurate solution to the inverse kinematics problem.