This paper introduces an innovative approach utilizing a deepneuralnetwork (DNN) to optimize the modulation scheme for time-modulated antenna array (TMAA) to verify specific side lobe and maximum harmonics levels. T...
详细信息
This paper introduces an innovative approach utilizing a deepneuralnetwork (DNN) to optimize the modulation scheme for time-modulated antenna array (TMAA) to verify specific side lobe and maximum harmonics levels. The proposed method involves training a DNN with a physics-informed loss function designed to reduce the discrepancy between the desired and actual beam patterns. This is accomplished by exclusively adjusting the periodic switching time sequence of each element within the TMAA. Specifically, the physics-informed deep neural network (PIDNN) is trained to optimize the switching-on times of for each antenna element. Simulation results demonstrate that the proposed technique achieves the desired beam patterns with significantly lower side lobe level and maximum harmonic levels compared to previously published methods. Additionally, the approach is compared to genetic algorithm (GA) which corresponds to a representative evolutionary optimization algorithm. Numerical results indicate that the PIDNN surpasses the GA in both computational efficiency and loss function evaluation.
Nonlinear mathematical models introduce the relation between various physical and biological interactions present in nature. One of the most famous models is the Lotka–Volterra model which defined the interaction bet...
详细信息
This study proposes a novel design and modeling framework for a compliant robotic gripper with high industrial relevance in precise manipulation robots. The gripper structure is optimized through stress-constrained to...
详细信息
This study proposes a novel design and modeling framework for a compliant robotic gripper with high industrial relevance in precise manipulation robots. The gripper structure is optimized through stress-constrained topology optimization, while its mechanical behaviors are predicted through a physics-informed deep neural network. Two use cases are explored, including the jaw’s stroke prediction and the working frequency modeling under physical constraints. A multi-objective genetic algorithm identifies optimal design parameters, achieving a 4.71 mm stroke, 43.62 Hz frequency, 22.25 MPa stress, and a safety factor of 3.23. The entire stroke of the gripper is 6 mm. It was revealed that the stroke is amplified twice without an external displacement amplification mechanism. A 3D-printed prototype is tested; the results indicated that there is good alignment between predictions and experimental results. The results revealed that the gripper’s large stroke enables it to adapt to objects of varying sizes. The design and modeling synthesis method for the compliant robotic gripper highlights its potential for applications in industrial manipulation tasks on robotic arms.
暂无评论