TianjinSino-German University of Applied Sciences

ABSTRACT Precisely elucidating the influence of various orientations on the fracture performance of unidirectional fiber‐reinforced composites (UD FRCs) is crucial for evaluating the structural integrity of defect UD FRCs. Existing studies lack in‐depth analysis of failure mechanisms and accurate prediction of fracture toughness for different orientations. Quasi‐static eccentrically‐loaded single edge crack tension (ESET) fracture tests were conducted in air on compression‐molded unidirectional T300 carbon fiber/7901 epoxy composites with fiber orientations spanning 0°–90° (15° increments). The digital image correlation technology was used to reveal the deformation and failure mechanisms during the fracture process. Then, the mixed‐mode I‐II trilinear cohesive zone model (CZM) was developed and used to predict fracture behavior under different orientations. It was found that increasing the orientation angle leads to a higher initiation fracture toughness, consistent with the observed trend in maximum tensile strain ε yy . The relationship between fracture toughness and ε yy has been established. Moreover, combined with microanalysis, it can be observed that as the orientation angle increases, the failure mechanism shifts from fiber–matrix interface debonding, matrix cracking, and fiber bridging to matrix shear failure, fiber breakage, and pull‐out. Finally, the CZM method combined with the established B‐K parameters relationship can quickly and accurately predict fracture toughness under any orientations, and the maximum error is less than 3%. The current research on the damage mechanism and reliability evaluation of FRCs is of great significance.

Extending the truck–drone mothership system, this paper addresses a flexible multi-visit truck and drone joint routing problem (FMTDJRP). A mixed integer linear program is proposed to minimize total travel cost. An adaptive large neighborhood search with a knowledge-based acceleration strategy yields near-optimal solutions. Experiments across varied customer distributions, drone specs, and truck-–drone cost ratios confirm flexibility, adaptability, and cost efficiency.

A radial basis function neural network-based terminal sliding mode control (RBFNN-TSMC) strategy is developed to mitigate the buffering effects in robotic arm endpoint tracking. This approach specifically targets and compensates for the system's model uncertainties and unknown external disturbances. This control scheme is composed of two parts: a novel terminal sliding mode control term and an RBFNN-based compensation term. The former ensures rapid convergence to the sliding mode surface and stable approach to equilibrium, while the latter mitigates the adverse effects of model uncertainties and external disturbances. Stability analysis, grounded in Lyapunov theory, confirms that the tracking error converges to zero. Simulation studies conducted on MATLAB/Simulink validate the algorithm's efficacy, demonstrating that it achieves high-precision trajectory tracking while maintaining robust system stability.

Industrial dispensing systems consist of many components, and the temperature characteristics of these components vary significantly. To address this, a segmented adaptive PID temperature control method is proposed in this paper. This method integrates a segmented temperature control algorithm with a variable control coefficient algorithm based on output power, which not only ensures minimal overshoot in the system but also enhances its disturbance rejection capability. Experimental results demonstrate that, under identical conditions, compared with the traditional PID method, the proposed method reduces overshoot by 2–4 °C and decreases the amplitude of temperature fluctuations after disturbance by approximately 0.2 °C.

This study examines the stochastic bifurcation phenomenon in a fractional and multistable Rayleigh–Duffing oscillator subjected to recycling noise excitation. First, using the harmonic balance method and minimizing the mean-square error, an approximate integerorder equivalent system was derived for the original fractional-order system. Subsequently, the steady-state probability density function (sPDF) of the system amplitude was obtained via stochastic averaging. The critical conditions for stochastic P-bifurcation (SPB) were then determined using the singularity theory. The stationary PDF curves of the system amplitude were qualitatively analyzed across regions delineated by transition set curves. Finally, Monte Carlo simulations confirmed the analytical findings, validating the theoretical framework. These results provide insights for improving system response control through fractional-order controller design.