论文部分内容阅读
实际静力加载测试中,测试试件的尺寸、刚度、载荷等存在较大的差异,加载油缸位移的扰动、流体的非线性、摩擦等都会对测试结果产生影响,这就要求控制系统应具有一定的自适应性。针对经典PID控制参数不能在线调整的缺陷,研究了一种基于BP神经网络的PID控制算法,利用BP神经网络具有的任意非线性表达能力,通过对系统性能的学习,实现具有最佳组合的PID控制。通过仿真,验证了所设计的BP神经网络PID控制系统比PID控制系统建模时间短,系统更稳定,超调量更小。
Actual static load test, the test specimen size, stiffness, load and so there is a big difference between the load cylinder displacement disturbance, fluid nonlinearity, friction and so will have an impact on the test results, which requires the control system should have A certain degree of adaptability. Aiming at the defect that the classical PID control parameters can not be adjusted online, a PID control algorithm based on BP neural network is studied. By using any nonlinear expression capability possessed by BP neural network, learning the system performance can achieve the optimal combination of PID control. Through simulation, it is verified that the designed BP neural network PID control system has shorter modeling time than the PID control system, the system is more stable and the overshoot is smaller.