基于萤火虫优化神经网络算法的气浮台自动调平衡研究

为了利用气浮台在地面精准演示分布式多星组网技术及验证多模式高分成像过程，提出一种基于改进型BP神经网络PID控制的气浮台快速自动调平衡算法.针对以往算法调平时间较长、容易得到非最优解的问题，引入仿生萤火虫算法对神经网络初始权值进行优化，提高了算法的收敛速度和稳定性.基于构建的三轴气浮台运动学和动力学模型，通过仿真实验验证了优化算法对自动调平衡具有良好的控制效果，满足多星地面仿真的调平衡要求. 

Automatic Balancing Control of Air-bearing Simulator Based on Firefly Algorithm Improved Neural Network

In order to accurately reproduce the multi-satellite networking technology and simulate the multi-mode high-resolution imaging process on the ground, the three-axis air-bearing test bed was the key device of simulation. In this paper, a fast automatic balance adjustment control algorithm was proposed based on firefly algorithm improved BP (Back Propagation) neural network PID (Proportion Integration Differentiation) control. Aiming at the problem of long adjustment time and easy to obtain a non-optimal solution, the firefly algorithm was introduced to optimize the initial weight and threshold value of the BP neural network, and improve the algorithm performance in convergence rate and stability. Based on the kinematics and dynamics model of the three-axis air-bearing simulator platform, the simulation results show that the optimized algorithm reduces the x-axis centroid offset to 2.3×10-7m in 3.1s. The algorithm has faster and higher stability on air-bearing simulator automatic balancing control, and satisfies the demand for multi-satellite imaging process simulation.