粒子群算法优化BP神经网络的变载荷自平衡控制系统

发布时间：2018-05-09 18:53

本文选题：时变负载 + 自平衡　；参考：《西安科技大学学报》2017年06期

【摘要】：为解决常规PID控制难以在具有时变负载的自平衡系统中实时、精确调节负载的变化,在BP神经网络基础上,利用粒子群算法(PSO)优化BP神经网络,将神经网络的收敛速度进一步提高,并将算法应用到二轮平衡车控制系统中,对二轮平衡车进行动力学建模,介绍系统的结构、原理与实验方法,搭建二轮平衡车实验平台进行了施加突变负载情况下的试验验证。利用二轮平衡车实验平台车身上的姿态传感器得到车体倾斜输出角度,对比施加突变负载前后以及神经网络优化前后的车体倾斜输出角度变化。结果表明:粒子群算法(PSO)优化BP神经网络技术能够满足变负载二轮自平衡车控制的要求,实现了自平衡车的动态自平衡,提高了抗干扰能力,验证了优化算法在自平衡、抗外部干扰和缩短调整时间上的优势。
[Abstract]:In order to solve the problem that conventional PID control is difficult to adjust the change of load in real-time and accurately in a self-balancing system with time-varying load, the particle swarm optimization (PSO) algorithm is used to optimize the BP neural network based on BP neural network. The convergence speed of the neural network is further improved, and the algorithm is applied to the control system of the two-wheel balancing vehicle. The dynamic modeling of the two-wheel balancing vehicle is carried out, and the structure, principle and experimental method of the system are introduced. The experimental platform of two-wheel balancing vehicle was built and tested under the condition of sudden load. The tilting output angle of the vehicle is obtained by using the attitude sensor of the vehicle on the experimental platform of the two-wheel balancing vehicle. The angle of the tilting output of the car body before and after the application of the sudden load and before and after the optimization of the neural network is compared. The results show that the PSO (Particle Swarm Optimization) optimization BP neural network technology can meet the control requirements of variable load two-wheel self-balancing vehicle, realize the dynamic self-balancing of the self-balancing vehicle, improve the anti-jamming ability, and verify that the optimization algorithm is self-balancing. The advantage of resisting external interference and shortening adjustment time.
【作者单位】：上海工程技术大学机械工程学院;
【分类号】：TP13;TP18

【相似文献】